www.xlr8yourmac.com

Reviews and Daily News with a  Difference!
Mac Models/Mods  | CPU Upgrades  | Storage  | Video  | Audio/HT  | Apps/OS/Network/Misc  | Archives  | Search
News, Tips, Reviews or Questions to News at Xlr8YourMac.com
Apple G5 Tech Info, Updates and CompatibilityReturn to News Page

Click for Mini Pro Dual Drive!
Click for Mini Pro Dual Drive!

PowerMac G5 Tech Info/Compatibility Reports/Software Updates
Posted: June 24th, 2003
(Updated Nov. 15th for G5 1.8GHz Firmware Update
Updated Aug. 23rd, 2005 for Adaptec 29160N report
Updated Nov. 17th, 2004 for G5 (June 2004 models) firmware update 5.1.8f7
Updated: Sept. 28th, 2004 for G5 uniprocessor firmware update 5.1.5f2
Updated: Sept. 22nd, 2004 for Hammerfall rev 1.5/1.6 card damage warning
Updated: Sept. 13th, 2004 for G5 uniprocessor firmware update 5.1.5f1 for 10.3.5/sleep issue
Updated: Aug. 26th, 2004 for Liquid Cooling Pix and IBM 970FX Pwr Usage/Vore Info
Updated: July 23rd, 2004 for more Power usage Info
Updated: July 6th, 2004 for "June 2004" Model Dev Docs/CIP pages and Sonnet PCI card Compat. Info
Updated: Mar. 19th, 2004 for More G5 vs Itanium/Opteron Scientific Benchmarks
Updated: Apr. 8th, 2004 for 64-Bit Linux info
Updated: Mar. 16th, 2004 for Apple G5 Fan Control Update
Updated: Feb. 10th, 2004 for G5 disassembled photos
Updated: Dec. 2003 for G5 firmware update 5.1.4, G5 HD Cooling Tip, Softraid 3.0.2 update
Updated: Nov. 2003 for Logic Audio Platinum X and ProTools Updates, Apple Pro Apps updates & FFT benchmark results)
Updated Oct. 2003 for list of G5 compatible Acard PCI cards, Rev. 7.1 card mods note, G5 Cinebench 2003 beta, Bluetooth firmware update & ATTO UL4D/UL4S firmware/driver updates)

= Sections Link List =
G5 Tech Info/Docs/Parts install Guides | G5 OS/Software/Apps/Driver Updates & Compatibility
G5 PCI Card Compatibility Info | RAM Compatibility |  USB/Firewire
G5 Noises |  G5 Performance Tests | G5 Pix

NOTE: This page is for the single Core G5 Models (pre-fall 2005). In Fall 2005 Apple totally revised the PowerMac G5 with Dual-Core CPUs, PCI Express graphics cards and more. For details on the new Dual-Core models, see the latest Apple's current PowerMac G5 pages.
For other Powermac G5 articles, tests, reviews, etc. see the G5 section of the Systems page.

High Resolution Pictures Of The Dual 2.5GHz G5 Liquid Cooling System (from the Aug. 26th, 2004 news page)

" High Resolution Pictures Of The G5 Liquid Cooling System
I just made a new .Mac photo album of my 2.5Ghz G5 processor. I hope you like the Cube vs G5 (core) picture.
James. "

The last photo shows the G5 radiator core beside the core of a G4 Cube (they're about the same size).
G5 Hard Drive Cooling Tip: (Dec 1st, 2003) A G5 owner sent a long email with comments, temperature tests and a tip on relocating the Hard Drive Cooling Fan Temperature Sensor to better cool the G5's hard drives. See the Powermac G5 Hard Drive Heat/Cooling Tips page for details. (Article updated with notes that 10.3.3's revised fan control software helps even with sensor in stock location.)

Dual G5 1.8GHz for $2499 Announced, single G5/1.6GHz price cut to $1799 (Nov. 18th, 2003) There's been rumors of a Dual 1.8GHz G5 that proved true - Apple now lists the Dual G5 1.8GHz for $2499 (only $100 more than the price of the original single 1.8GHz which has been dropped). They also reduced the price of the single G5 1.6GHz to $1799. The Powermac G5 page has been updated for the changes and it's listed at the Apple store.


G5 Noises (Beeping and/or Other Noises during Dock or Window Movements): (from the Sept. 29th, 2003 news page.) Several readers wrote in reply to the earlier news item (below) on G5 "Beeping" noises to say that using Apple's CHUD tools to disable "Nap" mode may help. (Odd that if this is caused by processor idle mode switching that the "high performance" energy saver setting would have been more prone to it. However as I said earlier, I do not hear this on my Dual G5.) Here's a sample reader mail with the tip:

" I read on another help page that a fix for this problem is the following:

The solution requires that you download Apple's CHUD tools, which are available from
ftp://ftp.apple.com/developer/Tool_Chest/Testing_-_Debugging/Performance_tools/CHUD_3.0.0.dmg.bin
After installing CHUD, go to System Preferences, open the "Hardware" pane and turn off "Nap". (Note: NAP mode is reset to On again after a reboot.)

Apparently this option turns off the G5's ability to use processor idling, eliminating the beeping noise.
Hope this helps those with the problem. I don't have any such problem with my dual processor 2.0GB G5.
Wayne S. "

I've had several G5 owners reply to this post saying disabling Nap mode does stop the beeping and other noises they heard before. See the page of G5 noise/CHUD Tip feedback.


G5 Boot Rom (Firmware) Versions: (Update - as of April 2004 G5s are shipping with boot rom v5.15. In late 2003 Apple released G5 firmware 5.1.4 update for 2003 models and in Nov 2004 released a 5.1.8f7 update for June 2004 models. The comments below were from early owners before that release.) After the report about early G5 shipments reportedly having problems with mixing 3rd party RAM with the OEM RAM (Hynix brand reportedly) I wondered if the boot rom version had changed since the first shipments. The S/N XB337xxxxx Dual G5 here (shipped 9/15) has rom version 5.0.2f2 (as reported by Apple System Profiler). Several readers sent their G5 boot rom versions:

" Dual 2 Ghz BTO in UK
CK338xxS-NVB
(boot rom version) 5.0.5f0
Tan "

Another Dual G5 owner (very recent shipment) wrote he has an even later boot rom version:

" Mike, I got my G5 Dual 1 week ago, Just looked at the Boot rom version and it is newer than others mentiond in todays news."

    Machine model
    PowerMac7,2 (version = 2.2)
    Boot Rom: 5.0.7f0

And another report of 5.0.7f0 firmware:

" Mike, we have three CTO machines, built 9/19 (1) and 9/22 (2) boot Rom on all three - 5.0.7f0
-- Fred C.
General Manager
TheTroupe Modern Media Design & Production "

A reader with 5.0.7f0 firmware noted he still had problems using CL2 dimms (even without any CL3 OEM dimms installed based on his previous comments) - so there's still a requirement for DIMMs to have CL3 timing in the SPD. One reader later wrote he had an even earlier boot rom version than mine - 5.0.1f0.

Maybe Apple is waiting to accumulate fixes/changes before releasing a public update - but I'm hoping they will soon release a firmware update for the G5s which might solve some of the quirks that some owners have seen.


G5 System Photo Galleries:

High Resolution Pictures Of The Dual 2.5GHz G5 Liquid Cooling System (from the Aug. 26th, 2004 news page)

" High Resolution Pictures Of The G5 Liquid Cooling System
I just made a new .Mac photo album of my 2.5Ghz G5 processor. I hope you like the Cube vs G5 (core) picture.
James. "

The last photo shows the G5 radiator core beside the core of a G4 Cube (they're about the same size).

Photos of VA Tech's Xserve G5 Supercomputing Cluster (8/10/2004) Jamie sent a note that there's photos of VA Tech's new Xserve G5 based supercomputer cluster (replacing their original G5 tower models) at http://www.tcf.vt.edu/upgrade_gallery/upgrade_pics.html.

Photo Gallery of Completely Disassembled Dual G5 A reader posted a link to a .Mac page with photos of a Dual G5 completely disassembled at http://homepage.mac.com/dabaer/PhotoAlbum8.html. (Text is in German.) Note what appears to be a heatpipe cooler on the backside of the motherboard (cooler for Chipset/Memory controller I assume).

Cool looking G5 Stand: Krishna (of PC weenies cartoons) wrote with link to a japanese site with a very cool looking G5 stand. See http://www.pawasapo.co.jp/ac/rs/rs.html for photos (page is in Japanese - I've not yet tried translators on the page such as Babelfish.)

Movie and Photos of VA Tech G5 Supercomputing Cluster Walter sent a note he's posted a re-encoded (QuickTime 320x240) movie of the BBC story on the VA Tech G5 Supercomputing cluster.
Jamie Dresser of OWC sent a link to a mirror of all the original photos of the "Terascale" (1100 G5 mac cluster) VA Tech supercomputer project at http://tartarus.uwa.edu.au/~trs80/Terascale/.

First G5 Dual 2GHz owner photos/initial tests: See this page posted Sept. 4th, 2003.

G5 CPU module photo: Mr. Takashi Imai posted a photo of the Power Mac G5 processor module on his BBS at http://www.geocities.co.jp/SilkRoad/4512/970cpu1.jpg. (Thanks to M.Isobe for the link.)

High-Res G5 Photo Gallery: (from the Sept. 3rd www.xlr8yourmac.com news page)

" Hi Mike,
Thought I'd share some "artsy" high resolution G5 photos with you and your readers:
http://www.billnoll.com/g5/
Bill Noll "

Very well done... this guy should be on Apple's payroll.

Apple G5 Photos and Movie: Apple's Powermac G5 pages link to an Apple G5 movie and a few photos in the Apple photo gallery (more detailed photos are linked below).

G5 Movie on Features/Fan Speed Control: (from the Aug. 28th www.xlr8yourmac.com news page)

"Some quick shots of our 1.6Ghz G5 that came in yesterday. Pictures of the main components and all. The video shows how the fan mechanism works and you can hear the difference in the fans and all. Thought you may want to know.
http://homepage.mac.com/aaronsteele/iMovieTheater20.html
http://homepage.mac.com/aaronsteele/PhotoAlbum19.html
More pictures will come tomorrow.
Best Regards,
Aaron S."

(July 2003 links, before G5s shipped) Macitynet has some pix of the new G5 systems taken at WWDC at their G5 photo gallery and later posted page with more detailed G5 system photos. 99mac.com also has some nice photos of the new G5 systems.

G5 Superdrive Model: (from the Feb 7th, 2005 news page)

" The current superdrive that Apple put in my G5 dual 2.5 (just picked it up yesterday) was: SONY DVD RW DW-U21A
from apple system profiler. Here is full info:
Device ID: 0x0043
Revision ID: 0x0000
SONY DVD RW DW-U21A:
Manufacturer: SONY
Model: SONY DVD RW DW-U21A
Revision: AADB (firmware version)
Drive Type: CD-RW/DVD-RW
Disc Burning: Apple Supported/Shipped
Protocol: ATAPI
Unit Number: 0 (master)
It was shipped from Apple to the Delaware Apple store Jan 27, '05.
Dave C. "

Apple often uses more than one supplier for "superdrives"/DVD burners (historically Pioneer, Sony and LG) so there could be other drive brand/models used in various production runs.
Apple's June 2004 models of the G5 tower usually shipped with the 8x (DVD+R/-R) rated Pioneer DVR-107D according to reports (and later in fall 2004, they shipped a "DVR-117" model, basically a DVR-108AA with modified firmware.) The 2003 G5 models shipped with 4x rated drives - either a Sony DW-U10A or Pioneer DVR-106 model.

Note on Optical Drive Swaps/Upgrades:
Also as a FYI - according to a reader report and the notes on the Apple CIP Guide for replacing the optical drive (see this FAQ item for Apple's G5 CIP page/Guides) - there are different base screws (standoffs) for the OEM Sony DVD burner and the Pioneer DVD burners and the Combo drives. (A reader with the OEM DW-U10A sony drive replaced it with a Pioneer DVR-107 drive and said the screw/standoffs from the Sony drive resulted in the Pioneer drive's tray not causing interference with the G5's CD drop-down door/cover. (He had already removed the drive's front tray bezel, which is required regardless. For the Pioneer DVR-107D drive, you press out and up on the lower tray bezel edge to clear the retaining tabs, then you can slide it up and off the tray front edge.)
He mentioned he ordered the correct screws from Apple (normally apple does not sell parts like this to end users - so contact a local apple dealer to order them if you need to.)

(Info on the first revision 2003 G5 towers with 4x rated Superdrives follow - info from Fall 2003)
Although several G5 (single CPU) owners have reported their Superdrive was a Sony DW-U10A, I had mentioned that Apple often uses more than one source for drives. The first reader with a dual G5 sent an ASP report showing the drive was a Pioneer DVR-106 model. See the main site Systems page, G5 section for links to articles such as the first photos/info/tests of a Dual G5.
Both the Sony DW-U10A and Pioneer DVR-106/A06 support DVD+R/+RW as well as DVD-R/-RW, but may not with the OEM Apple firmware - I've asked for reports on using DVD+R/+RW media from the owners. I guess the G5's 10.2.7 added native support for the DVR-106, which was not present in 10.2.6 and earlier (requiring the modified plugin here for iTunes, ect. burn support).

(first report from the August 26th, 2003 www.xlr8yourmac.com news page)

G5 Superdrive Model: Although Apple often uses more than one OEM brand/source for their drives - this reader with a new G5 system said the superdrive was the OEM Sony DW-U10A. (I think the same drive has been used in some iMac G4s or other mac models in the past - it has native burn support in 10.2.6 and perhaps even an earlier OS X version).

"I don't know if anyone has let you know, but I just received my 1.8GHz G5. It has the Sony DVD RW DW-U10A for the SuperDrive. This was a pleasant surprise.
Greg E.
ExperCom"

From what I've heard in the past, the Apple shipped OEM DW-U10A drives did not support DVD+R/+RW burning like the standard DW-U10A does. (Panther/OS X 10.3 was the first Mac OS to have DVD+R burn support, previously I had to use Toast for burning DVD+Rs with my DVR-106 and later drives.) I asked Greg if he could test for DVD+R/+RW support. (For past reports on that drive - search the drive database here selecting Sony as the drive brand and "DVD+R/RW + DVD-R/RW" as the drive type. Omit mac model to not limit search results to a specific model.)


Apple G5 Tech Notes/Docs:

IBM article on 64-bit PowerPC (970/G5) architecture (10/24/2004) Frank at IBM sent a link to Understanding PowerPC 64-bit architecture on the PowerPC 970 family (aka the "G5") that covers "critical issues in IBM's 64-bit POWER designs, 32-bit compatibility, power management, and processor bus design".

IBM Article on 970FX G5 Power Usage/Vcore Limits (from the 8/26/2004 news page, by M.Isobe, author of several previous articles on G4 CPU module designs on the Systems page.)

" Dear Mike, ... I have found interesting document from IBM that is related to your Xserve G5 power usage reports. (this earlier page)
Please read an article titled "Power envelope and power management in Power Architecture processors" ( http://www-1.ibm.com/technology/power/newsletter/august2004/article7.html).
The article shows the power consumption of 970FX/2.5GHz in various power management mode. According to the article, 970FX with 1.3V Vdd has a reliability problem. So, the upper voltage is limited to 1.2V. I think it was the reason for the CPU Vcore was 1.2V in newer revision of Xserve G5.
The article also stated that IBM's 90nm process has a problem that prevents operation below 1.0V Vdd. So, I can speculate current situation of 970FX;

1) The reliability problem in high voltage (Vdd > 1.2V) operation inhibits faster (> 2.5GHz) chip,

2) We cannot expect ULV revision for PowerBook because 90nm 970FX does not work below 1.0V.

I am happy if this info is useful for your readers.
Best Regards,
Michiro Isobe "

I've added this to the previous page on Xserve G5 (970FX)

IBM Research Doc on PowerMac G5 Power Measurements/Usage (added July 23rd, 2004)

"I know you've listed basic info on the PowerMac G5's power saving modes (nap mode, bus slewing...) in the past, (info included on this page here. Disabling NAP mode using CHUD Tools' hardware control was a popular way to stop the 'squealing/chirping noises' from the PS.) but there's an interesting PDF file of an IBM Research Report on "Power Measurement on the Apple Power Mac G5" with info on Mac OS X and Linux using the original G5 Dual 2GHz system. (Not the 90nm 970FX.) They mentioned a 30W difference in Idle Nap Mode vs Idle Non-Nap mode for instance and measured fan rpm/outlet temperatures, CPU temperature and variations between the 2 CPUs.
Cheers, Jen."

G5 Power Management Settings Can Affect Bus/CPU Speeds: (from the "Power Management" section of the original Apple G5 Developer Docs in fall 2003)

" A common power management strategy is implemented across all Macintosh models. The basics of Apple's power management techniques are described in the following paragraphs.

Processor and Bus Slewing
To lower power consumption, heat generation, and fan noise, the Power Mac G5 computer incorporates an automatic power management technique called bus slewing. Bus slewing is designed to run at high processor and bus speeds and high voltage when the demand on the processor is high, and to run at low processor and bus speeds and low voltage when the demand on the processor is low. Switching between different processor/bus speeds and voltages is achieved by a gradual transition that does not impact system or application performance and operates seamlessly to the user. In slewing, the bus runs at half the speed of the processor.

The ranges of the slewed processor speeds are listed below:

Configuration    Processor range
1.6 GHz    1.3 GHz to 1.6 GHz
1.8 GHz    1.3 GHz to 1.8 GHz
2.0 GHz    1.3 GHz to 2.0 GHz

In addition, the Power Mac G5 computer allows the user to control bus slewing mode. The options for specifying either high, reduced, or automatic processor and bus speeds are located at System Preferences>Energy Saver>Options; then select Automatic, Highest, or Reduced.

If the Power Mac G5 computer detects a system temperature that is too high, due to high ambient temperatures or other factors, it will automatically enter bus slewing mode regardless of the selected setting."

Apple White Papers on G5 System and G5 CPU: (added Aug. 26th, 2003) Apple has Kbase docs with links to PDF files of G5 related white papers (English and Japanese PDF doc links there):


G5 Developer Doc Finialized: (Aug. 20th, 2003) The Apple G5 Developer Doc (previously preliminary) has been finalized and released this morning. The PDF revision date is August 20th.

" The final version of the PowerMac G5 Developer Note is now available http://developer.apple.com/documentation/Hardware/Hardware.html
or
http://developer.apple.com/documentation/Hardware/hardware2.html
The developer note for the newly introduce PowerMac G5 is now available in HTML and .pdf format.
(Actual direct links to HTML version is here and here is the PDF doc.)
This developer note provides technical, hardware architecture information on the PowerMac G5 announced last month.

It includes information about the major components on the logic board: the microprocessor, the other main ICs, the buses that connect them to each other and to the I/O interfaces."


G5 Developer Doc:
(Updated June 2004) Apple has updated the PowerMac G5 Developer docs for the "June 2004" G5 models.

(from the July 3rd, 2003 main site news page - see above for later release note)
A reader posted a link to the preliminary developer doc (PDF file) for the new PowerMac G5. (Includes notes on system design, Power supply (450W and 600W), CPU module, interfaces, PCI slots, chipset, ram and more.)

Tech Note: G5 Performance Primer: Apple has a Tech note/PowerPC G5 Performance Primer that outlines the design differences in the G5 vs G4 CPUs. Sections include:

  • Quick Comparison: G4 & G5
  • Summary of Differences between the G4 and G5
  • Performance Do's for the G5
  • Performance Don'ts for the G5
  • How to make your code run best on the G5

Apple G5 Customer Installable Parts Guides: (Updated July 2004)
Although many of the guides are similar, Apple has posted a separate page for G5 (June 2004) model Customer Installable Parts Instructions. Their previous page for the original G5 towers is here. Both have links to PDF guides to installing/replacing boards, memory, fans, drives, cables, etc.

Apple G5 Kbase Docs:

Apple also has a Technical Q&A Note on the G5's PCI/PCI-X slots

(G5 RAM Note: there's been problem reports from G5 1.8/2GHz owners at least from using some PC3200 memory with faster than CAS 3 ratings. I suspect Apple may require the SPD chip to have CAS 3 timings even if it's a faster rated dimm, despite the fact the Developer Docs for the G5 note that is supports "CAS latencies of 2, 2.5, 3, 4, and 5."
This happened back in 2001 after the G4 firmware updates noted in a past article here. Even CL2/CAS2 dimms had to have CL3 timing in the SPD chip or otherwise it was not recognized.)


Note on Early G5 Shipments and Problems with OEM + 3rd Party RAM Mixing In a follow-up to the previous posting on 3rd party RAM issues (i.e. if no CL3 timing in the SPD) mentioned in Monday's news and repeated below), a reader sent a note:

"Apple has sent a tech note to european resellers telling them about the RAM in the first units delivered in Europe, as described below. It seems it functions as is, but that when 3rd party RAM is added KPs (kernel panics) ensue.
I got it from a french web site (Macbidouille) who got it from a german web site (Macnews.de). Just thought you should know, since you have reported such problems (with CAS latencies) that might be related. Maybe we can have dual G5s running CAS 2 RAM !

(Update - macbidouille later reported the issue was only with OEM Hynix modules and that removing them was the recommended solution.)

    Affected units
  • 1.8GHz product serial number below CK339yyyPAZ, CK339yyyQ66, CK339yyyQ6Z, CK339yyyPB2, or CK339yyyQ68
  • 1.8GHz CTO configuration product serial number below CK339yyyNVA
  • 2.0GHz product serial number below CK339yyyPB0
  • 2.0GHz CTO configuration product serial number below CK339yyyNVB

Units with serial number greater than CK339xxxxxx are not affected.
Units with serial number starting with YMxxxxxxxx are not affected.

Apple has a kbase doc on locating the G5 S/N for those that want to check - remove the cover and it's on the lower area where the door mates to the case - Apple system profiler will also report it.
I wonder is this really an issue with the RAM shipped in them or the system/firmware of the early units? The boot rom version in this Dual G5 (s/n XB337...) per Apple System Profiler is: 5.0.2f2, which I have run with two CL3 512MB dimms + the OEM CL3 256MB dimms without any KP's (so far). If anyone has later or different G5 boot rom version let me know. (So far the latest firmware/boot rom version reported is 5.0.7f0. A Dual G5 owner w/5.0.7f0 said he still hd problems using some CL2 rated dimms - so the requirement for CL3 timing in the SPD chip is still there apparently.)
On a related note - a recent email from a G5 buyer said his Apple order listing shows "DDR440" memory instead of the usual "DDR400" listing. (Not that the memory bus is going to run at the higher speed.)

Notes on G5 Memory Modules (CAS/CL Timing Requirements):
I've used two 512MB PC3200 CL3 dimms fine in a Dual G5, but after having several G5 owners report problems with some faster than CL3/CAS3 dimms in their G5s - I've been meaning to post a note here about what I think is the problem. The first report I had on this was about a week ago from a Dual G5 owner (Martin) that said his Geil Golden Dragon and Mushkin Level II black memory (both rated faster than CL3) were not recognized in his G5. My first guess was that the PowerMac G5 required the SPD chip on the DIMM to have CL3/CAS3 timing even if the dimm was rated for lower CAS (faster) timings - despite the Apple Dev. Doc notes. (This was a problem with some CL2 dimms back in 2001 after the G4 firmware update noted in reports/tips page here back then - CL3 timings had to be in the SPD even if the dimm was rated for CL2 timing.)
Since the Apple PowerMac G4 developer doc notes that it supports other CAS latencies (under Expansion/RAM Expansion section under Figure 4-1 it says):

    "For all microprocessor speeds and for both DDR400 (PC3200) and DDR333 (PC2700) SDRAM DIMMs, the Power Mac G5 supports CAS latencies of 2, 2.5, 3, 4, and 5. "

I asked Martin to test with only a pair of the (faster rated) DIMMs installed and he said it would not boot. (Since I suspect all dimms will be run at the timing of the slowest CAS/CL timing dimm installed, I wanted him to test with the OEM CAS3 dimms removed to see if it would run with just the faster rated DIMMs installed.) Therefore I concluded that despite the note on support of lower CAS timings than 3, as with the previous firmware updated G4s, the G5 must require the DIMM to have CAS3 timings in the SPD chip, even if the DIMM is rated for lower CL/CAS timings.
I wasn't sure Martin believed my theory was correct due to the Apple Dev. Docs note (he wrote today that he does), but this weekend I found a thread in the Apple G5 forums that mentioned this was confirmed by a Mushkin tech support person and that Mushkin's product page for their memory lists PowerMac G5 specific modules with a note that says:
"Designed and Programmed for Apple Mac G5 computers".
(A reader in the Apple forums thread said he was also told by Mushkin the DIMMs had to have CL3 timing in the SPD to work in the G5.)

I'm accumulating a list of DIMMs that worked and didn't work and will post that later in a section on the PowerMac G5 Tech Docs/Compatibility/Updates page, but if you're buying memory for a G5, and it's not already listed as compatible, I'd ask the vendor if it is guaranteed to work in your PowerMac G5 model. Ironically the higher-priced (Lower CL/CAS timing) DIMMs are less likely to work than the cheaper/more common CL3/CAS3 DIMMs, but I'd still ask before buying to avoid any surprises and hassles with a return and potential restocking fees from dealers that don't guarantee/test their memory in a G5.

The question in my mind (and many others I'm sure) is can the PowerMac G5 really run at faster than CL3 timing? (assuming all install dimms have faster rated memory than CL3.) Of course the Dev Docs indicate that, so is the CL3 timing requirement in the SPD just a check at boot to make sure the system can read the DIMM's CL3 timing for cases where there's a mix of memory speeds and CL3 therefore will be the timing used. (i.e. Pairs of CL3 OEM dimms and faster pairs installed at the same time.)
If all CL2 dimms were installed for instance (and also had CL3 timing in the SPD, required to even be recognized at boot apparently) - will CL2 timing actually be used? (And is there any software that's accurate to even report this -other than trying to run a consistent memory bandwidth test perhaps. If the Benchmark/test had much run/run variation, that could make it less useful for a test like this since the difference in overall performance may be small.)

Apple G5 USB and Firewire Device Compatibility:
Since this page is already long, I've created a separate page for G5 owner reports on USB and Firewire devices (including cameras, scanners, card readers, KVM switches, etc.)

Apple G5 PCI Card Compatibility Info: (In reference to the G5 requirement for 3.3V PCI cards - but there's also possibilities of firmware/OF issues. Most recent info listed first)

Adaptec PowerDomain (Mac) 29160N Beta Driver with OS X 10.4.2: (NOTE: see older posts below for info on Open Firmware mods for booting from SCSI drives attached to Adaptec 29160/39160 cards and Adaptec's original info on G5 SCSI card compatibility.)

" I'd like to report that I have just installed an Adaptec PowerDomain 29160N SCSI card on my G5 dual 2.3GHz running OS X 10.4.2. I have an old Epson Expression 636 SCSI scanner that I'm very fond of. The computer booted up fine with the card in, but when I plugged in the scanner, turned the scanner on, then rebooted the computer, the computer would not start up.
I had to install the card driver from Adaptec found here:
apd29160x_X_v1.3beta1.hqx

After I did that I was able to boot with the SCSI scanner turned on and plugged in, and I'm actually able to use the scanner now! It did seem like it took longer for the computer to start up though.

One question, in case you might have a quick answer: I've read that some (all?) SCSI cards prevent "deep sleep." Where can I find out more about what this means, whether I now have that problem, and what can be done about it?
Thanks, Jason
(Not much can be done about it if there's a problem, but it's easy to check. Can you put the system to sleep (menu item) and wake it up OK? (and when in sleep mode does the system go into deep sleep? i.e. all fans are off, power LED pulses as normal).-Mike)

Thanks so much for the info. I just tried putting the computer to sleep and it DOES appear to go into deep sleep, with the SCSI scanner turned on or off. The tiny little light on the front of the G5 pulses, the machine is utterly silent, the firewire external hard drive spins down, and it takes a mouse click (not simply moving the mouse) to wake it up. So I guess I have lucked out! I hope my success will be heartening to others trying to use SCSI devices with a G5.

By the way, I got the 29160N off eBay (for about $40) and the one I got DID specify that it was for the Mac, if that makes any difference.

I did find the info about the openfirmware mod. (below) I assumed that its purpose is to allow you to boot from a SCSI hard drive attached via the SCSI card. Is that correct? (Yes-Mike) If so, I didn't want to bother mucking with it because I only plan to use the SCSI card for the scanner as all the SCSI hard drives I have are pretty old now and probably not much bigger than 1gb. Well, maybe I have a 20gb one somewhere. But at any rate, even if I did plug in an external SCSI hd, I wouldn't need to boot from it; I've got an external firewire hd that I can boot from if needed.
Thanks again, Jason "

One reader replied to this post saying that OS X 10.4 Server may have later Adaptec drivers than the beta ones at their website.


WARNING on Hammerfall DIGI9652 (rev 1.5/1.6) Card going up in Smoke (from the Sept. 22nd, 2004 news page)

" Hi, I have a new G5 DUAL 2GHz and want to use my DIGI9652 from my G4 where (it was) working very good.
What is happening...Push the powerbutton and "SMOKE".
The Card and the G5 Slot are burned !!.
Later i went to the RME page and read this in the not so easy finding menu.

    important information on compatibility with Apple G5 computers
    RME updates information on G5 compatibility
    All currently available RME interface cards are fully compatible to Apple's new G5 computers. They operate at the reduced voltage of 3.3V, and therefore provide the second notch at the PCI contacts to fit in any PCI-X slot. This is especially true for all PCI cards of the HDSP series, HDSP PCI, HDSP 9652 and HDSP 9632.

    Older cards are in many cases not compatible:
    Hammerfall (DIGI9636 and DIGI9652) revision 1.5/1.6: These cards have been shipped since June 2000. Until the end of 2002, cards have been shipped, which included a PCB error, making them incompatible to PCI-X slot computers. Unfortunately it is not possible to define the time frame more accurately, nor to list the affected manufacturing dates.

    DIGI96 series: The PCB error also affects all cards of the DIGI96 series until the end of 2002. Unfortunately it is again not possible to define the time frame more accurately, nor to list the affected manufacturing dates.

    Using one of the above cards in an Apple G5 computer can cause serious damage to both DIGI96 and Hammerfall series of cards, as well as damage the G5. Therefore RME strongly advises to have all cards meeting the above criteria to be checked by RME before using them in a G5. Checking and - if neccessary - modification is done for free by RME's representatives.

    Hammerfall (DIGI9636 / DIGI9652) revision 1.1: These cards, manufactured from 1999 to April 2001, are not Mac compatible at all - and thus can't be used in a G5 anyway.

    Note: the revision is printed on the PCB. The date time of manufacturing is coded as week/year on the back of the cards, within a copper area (for example 17/01 for week 17 in 2001).

Why they don't put that on the Mainpage and TOP in Extra RED color? Is it not so important? What about people that have no easy internet access...Now we will have some money for that s**t from RME, but we will see what happens... to be coninued.
Greetz "

He also sent a photo of his burned Hammerfall DIGI9652 and G5's PCI slot connector.

Sonnet PCI Card G5 Compatibility Info: See Sonnet's G5 Compat. Page (which has been revised again as of April 2005. Note a Tempo Trio owner wrote in May 2005 to say that after he tried the card in his G5, the IDE ports on the card didn't work. (Sonnet doesn't list the Trio as G5 compatible but I was told it's -electrically- compatible, so no damage should have resulted, so he may just have a bad card. I suggested he contact sonnet tech support.)
BTW - Also check their firmware updates/drivers page for any updates (in case of firmware patches/updaters, etc.).

Summary of PCI Firewire and SCSI card G5 Tests: Here's a summary of G5 owner reports (not just company claims) on using PCI SCSI, IDE or Firewire cards so far:

    PCI IDE and Serial ATA cards:
  • Acard ATA/133 RAID card (see below for full acard list)
  • Firmtek SerialATA card

    Firewire or USB cards:
  • Adaptec Fireconnect 4300 PCI card does not support deep sleep (fans run at full speed in sleep mode - see report on G5 FW/USB feedback page)
  • Belkin 3 Port PCI Firewire card F5U501 does not support deep sleep fans run at full speed in sleep mode - see report on G5 FW/USB feedback page)

    SCSI PCI Cards:
  • Adaptec 39160: Reported working Dual G5/2GHz (OF mod required for booting from devices)
  • Adaptec 29160: Reported as bootable using OF mod below in G5/1.6GHz
  • ATTO UL4S, UL4D: *See notes on Firmware/Driver update which helped one G5 owner with prev. problems.
  • ATTO UL3S: No reports yet noting success
  • ATTO UL2D: One report of working in a G5 1.6GHz (I asked if a retail or OEM card and if bootable from SCSI drives)
  • ATTO UL3D: (OEM card w/ATTO 2.0 driver update) One report of it working in Dual G5


Acard Lists G5 compatibile PCI IDE/SCSI/SATA Cards: Acard has a page with a G5 Compatibility table of their various PCI IDE, SCSI and SerialATA controllers. (Note - as of summer 2005 the past link to http://www.acard.com/eng/product/hd_list/g5.html no longer works. I couldn't find the page anymore - but did see a FAQ item there that their IDE RAID cards were not G5 compatible.)

Revolution 7.1 PCI card (mods required): (Oct. 15th, 2003) Despite a previous report/quote from M-audio tech support (and mentioned on in their Revolution 7.1 driver v1.2.7 update in September as well as their page at http://m-audio.com/news/php/g5compatibility.php which says "(The critically acclaimed Revolution 7.1 consumer surround sound card is already PCI-X compatible.)" - maybe new production models?) this owner said he was told his card required a rework for G5s for full compatibility.

"I just got off the phone with M-Audio tech support regarding some crashing I'm experiencing on my Dual 2 GHz. G5 I'm pretty sure is related to the Revolution card. The guy I spoke to told me that the Revolution card is NOT PCI-X compliant (contradicts their own posted page comments), and while it works most of the time for me just fine, I do hear occasional blips in the sound (in general), and I've been crashing about once a day.
The guy on the phone told me that for $49 plus another $5 UPS ground shipping I can send in the card and get the board modified for PCI-X compliance. I placed my order with MacMall for the card on 9/7/03... The article below I found on your website about M-Audio cards and the G5 says that the cards are PCI-X compliant according to a user's call to their tech support line. His posting is dated 9/8/03... Is it possible I just got a slightly outdated one? (I really don't know) Might there be a way to check serials or something to tell if mine is before the revision? Why would their tech support just today tell me that they need to modify the board and that they're not PCI-X compliant but they're working on that? I just don't want to be had here. Thanks guys,
Marco"

I'm guessing this may be a firmware change (just a guess) rather than a PCI voltage issue but I'm going to try and call M-Audio to see if they can say if any cards made to date don't require the mod or if all cards do. What's odd is that their Sept. 24th driver update for the Revolution 7.1 card specifically noted G5 support.


ATTO Firmware/Driver Updates for UL4D/UL4S cards: (from the Oct. 14th, 2003 news page) A reader sent a note that a new ATTO firmware/driver update for the UL4 series cards fixed a problem he had with the card in a G5. (see below for previous reports from ATTO and Adaptec SCSI card owners.)

" You might want to know that ATTO now has new firmware and drivers for the UL4 cards listed on their site. I downloaded them, updated the firmware with my G4 and then installed the card in the G5 and it now boots and so far runs great.
Doyle Y. "

I asked him which card he had - UL4D or UL4S (retail version not OEM I assume since the flasher normally won't work with OEM cards). Please include details in reports (saves us both time). I went to ATTOs site and found the links to the downloads of the firmware/driver updates. Both the UL4S drivers page and the UL4D drivers page list this update note:

" Mac OS X Version 2.10b (File Size 972k)
This is a beta driver release of the ExpressPCI UL4D/UL4S firmware and drivers. This beta release supports the G5 platform. Carefully review the Read Me for update instructions and general release notes BEFORE starting the update. G5 Users must update to this beta release for the adapters to work in this system. You will also need the latest version of the ATTO ExpressPCI Configuration Tool (v2.5) and Ultra 320 flash bundle to complete the update. We recommend that all G4 users continue to use the V2.01 driver release. "

As of almost noon Eastern, only the UL4D and UL4S cards had the update - the pages for the UL3D and UL3S card drivers did not show a new update (yet at least).

Apple Tech Note/Q&A on PCI/PCI-X slots: (Oct 6th) Apple has posted a Technical Q&A Note on the G5's PCI/PCI-X slots

ATTO OEM UL3D Card Report: (Oct. 6th, 2003) Although past reports from ATTO SCSI card/G5 owners were not good (see older reports below), here's one that noted an OEM UL3D working:

" Mike, two interesting G5 reports. I tried installing an Apple OEM UL3D SCSI card in the G5. (Dual G5 with SCSI card in middle slot) The machine started nicely and Apple Profiler recognized the card, however, Retrospect did not see my Ecrix tape drive. I then downloaed and installed the Atto version 2 drivers which are supposed to support the G5, but not the OEM card. (that driver update was noted here back in August, and other ATTO card problem reports said they had installed the latest drivers. I asked him what "revision" (firmware version) ASP reports for the card.) and viola, Retrospect sees the tape drive and works perfectly. I've tried several restores (but not a backup) and it worked flawlessly. In my experience, Retrospect is very fussy about SCSI connections so this is really a good sign. (There was an update recently to Retrospect 5.1 I think for SCSI devices. I asked him what version of Retrospect he was using)

On a sour note, I tried replacing my M-Audio Audiophile 2496 PCI card with their USB Audiophile. This worked perfectly in my G4. In the G5, however, it causes a kernel panic whenever it's turned on, even after a successful startup with it off. (Crashes on startup if the unit is on when starting). I've sent M-Audio an email, but have not heard back.
Thanks for a great site. George S.."

He later wrote with answers to questions about the Retrospect version he was using and the firmware version of the OEM ATTO card:

"Strange that you should ask about Retrospect. I was using version 5 and it worked fine. I upgraded to 5.1 (with the SCSI patch) today and it works about the same. So, bottom line is the ATTO oem card will work with either version. The firmware is 1.6.1f1. This card was purchased from OWC some months ago. OWC incorrectly states that the card will not work with the v 2.0 ATTO drivers designed for G5s. The v 2.0 drivers in fact did work (without them the card would not recognize the tape drive.

The M-Audio sound module is not a card, but connected through the USB port. I bought it because I knew the Audiophile card would not work. It has the same functionality as the Audiophile PCI card but connects through USB. In any event, any time you turn on the module you get a kernel panic within a few seconds. The module worked in my G4 flawlessly. M-Audio still hasn't responded to my e-mail.
George"

Belkin Firewire 400 PCI Card: (from a G5 owner post - see the separate page of G5 FW/USB device feedback for other reports.)

" I have a Belkin 3 Port PCI Firewire card F5U501 that worked fine in my G4. I called Belkin to ask if it will work in the G5, they said no problem. But there is a problem in sleep mode! With card installed, sometimes I come into room and computer appears to be asleep, but fans are running high speed and loud! I tap keyboard to wake from sleep, computer comes on and fans quiet down. I take out the card and I don't have the problem. Belkin says Apple has had an issue with sleep and pci cards for several years now. Apple says it is an issue for Belkin to resolve...
Does anyone have some info that may help? Is there a card out there (preferably other than Belkin since they weren't very friendly/helpful) that will work?
Peter T. "

If any G5 owners try a Firewire PCI card (FW400 or FW800) please let me know your experience (compatibility, deep sleep support, etc.)

G5 Owner Report on ATTO UL3S/UL4S and Adaptec 39160 SCSI Cards: (from 9/24/2003 email - updated with info on 39160 card use)

" I have tried both ATTO UL4S and UL3S.
With the UL4S the dual G5 will not even boot. Don't get any video. I tried two different UL4S cards with the same results in both slot 4 and 3.

With the UL3S card the computer will boot but when I try to copy any reasonable size file the computer will freeze. (problems during I/O can be termination related also, but that might not be the problem here.) Interesting ATTO's benchmarking utility shows the drives achieving 70MB per second.

Looks like none of ATTO's cards will work in a dual G5. I got two 73GB Seagate Ultra 360 I was planning to use on the G5 and as of now it looks like that will not be possible. Is anybody having any success with any SCSI cards and the dual G5 at least for high throughput drives.

So what is the solution for a high speed high capacity multi drive setup?
Doyle Y
(I asked if he had installed the latest ATTO 2.0 drivers (which noted G5 support) - he said yes. I also asked if the cards were retail or OEM models (retail he said) and if retail models, if they had the latest ATTO firmware and he said yes.-Mike)

(he later wrote)

Boy you are going to find some interesting stuff in this message.

First I talked with ATTO tech. He said the UL4S are not compatible and said they are working on a fix. As for the UL3S he could not give me an answer. But I was running the same card, cables and everything on a G4. (I had noted that the file copy problem he mentioned earlier could be termination related)

Now for the really shocker. I remembered I had a 39160 Adaptec card back on the shelf that never worked in the G4s. So just on a lark I put that card in the G5 slot 4 and guess what it seems to be working fine. ATTO benchmarks 70MB per second. Real file copy from SCSI drive 1 to 2 of a 969MB file was around 53MB sec. Opening that same file in Photoshop was the same from the SATA drive as from the SCSI drives, about 30 seconds.

I am amazed that so far I have not seen any problems from the Adaptec card.
Doyle Y. "

Previous ATTO SCSI card reports are below. Also see previous notes/tips on Adaptec 29/39160 SCSI cards below.

M-Audio Revolution 7.1 Driver Update: As of Sept. 24th, there's a 1.2.7 driver update for the Revolution 7.1 at http://www.m-audio.com/support/driversearch.php. Here's M-Audio's driver page release notes:

    Revolution Drivers 1.2.7
  • [Driver] Support for PowerMac G5 with system version 10.2.7.
  • [Driver] Support for System Version 10.3 (Panther).
  • Driver/Panel] Support for 4.0 and 5.0 audio modes.

Notes on PCI-X Slots - using a std PCI card reduces speeds of PCI-X slots: (from the 9/8/2003 news page)

(from ATTOtech website "Hot Topics" page item on PowerMac G5):

" How does using a PCI adapter in a PCI-X slot affect performance? Using a PCI adapter in a PCI-X slot will result in a performance degradation since all slots on the PCI-X bus will revert to PCI speeds, either 33 MHz or 66 MHz PCI, depending on the bus and adapters you are using."

Note: For the PCI-X G5s (1.8/2GHz) the block diagram shows two busses - one for the two 100MHz PCI slots, one for the 133MHz slot. See this page of the G5 developer docs which has a block diagram. The section on Expansion/PCI-X slots doesn't mention the effect of non-PCI-X cards on the PCI bus speed. According to ATTO's return call - the 133Mhz PCI-X slot should not be affected by a non-PCI-X card in one of the two 100MHz PCI-X slots. Of course if you owned 2 PCI-X cards, then using the 133MHz PCI-X slot for the std PCI card would preserve the speed of the two 100MHz PCI-X slots.
The ATTO Hot Topics page also includes other comments on PowerMac G5/ATTO card compatibility.

M-Audio PCI Card Compatibility: (added 9/8/2003)

"Hi, Mike. Enjoyed your site for years.
I thought I would FYI you, and your readers on M-Audio's excellent PCI cards compatibility with the G5s:
From M-Audio tech support:

Frank,
The current line of Delta Series cards are not PCI-X compliant. We are in the process of modifying all Delta PCI cards to fit in G5 PCI-X slots. Until this modified version of the Delta Series is shipped, the only M-audio card that is currently PCI-X compliant is the Revolution card. (NOTE: See the later report from a Revolution 7.1 owner that noted the card was not G5 compatible without a rework. Perhaps new mfg models are but just a FYI to existing owners-Mike)

Unfortunately, there is no estimation on the release date of the PCI-X compliant Delta Series cards. We appreciate your patience and understanding in this matter.
Please let us know if we can be of further assistance.
CK"

(See below for previous info on other brands of PCI Audio cards.)

Adaptec Page on SCSI Card Compatibility w/PowerMac G5: (added 9/8/2003) Mitchell Cohen sent a link to an Adaptec page on Powerdomain SCSI card compatibility with the PowerMac G5

" Do the PowerDomain SCSI cards support the new Apple PowerMac G5?

Does Adaptec support the new PowerMac G5 introduced by Apple? These PowerMac G5s have PCI or PCI-X compatible slots and I need to install a SCSI card, do the Adaptec SCSI cards work in these systems. This information applies to the following product(s): - AVA-2906, Power Domain 2930U, Power Domain 2940 series, Power Domain 3940 series, Power Domain 2940U2W, AHA-8945, Power Domain 29160, Power Domain 29160N, Power Domain 39160 - PowerDomain PCI SCSI Card(s)
This information applies to the following Operating System(s): - Mac OS X 10.2.x

Answer:
The Adaptec SCSI Card 2906 (ASC-2906, AVA-2906) and Adaptec PowerDomain 2930 (APD-2930, APD-2930C, APD-2930CU, AHA-2930CU) are 5 V PCI cards and will not fit into the PCI or PCI-X slots in the new Apple Power Mac G5 according to page 75 of Apple's Preliminary Power Mac G5 Developer documentation. Due to this hardware incompatibility the 2906 and 2930 will not function in these system. Adaptec is currently investigating developing a new set of SCSI cards that will be compatible and will update this article as information becomes available.

The Adaptec PowerDomain 29160N (APD-29160N), PowerDomain 29160 (APD-29160) and PowerDomain 39160 (APD-39160) are Universally keyed and will fit into either 5 V or 3.3 V PCI slots, so these cards will fit into PCI or PCI-X slot but have not been tested in the Power Mac G5. Please check back for additional information as it becomes available.

The HotConnect 8945 (AHA-8945), PowerDomain 2940 (AHA-2940), PowerDomain 2940W (AHA-2940W), PowerDomain 2940UW (AHA-2940UW), PowerDomain 3940 (AHA-3940), PowerDomain 3940UW (AHA-3940UW) are not support in any system since the the Blue and White G3."

Also remember the previous notes here on OF mods (see below) required for booting from drives connected to Adaptec 29160/29160N/39160 SCSI cards in Quicksilver, MDD and later Macs (a G5 1.6GHz owner reported that OF tweak also worked with a 29160 SCSI card).


Open Firmware Mod for Adaptec SCSI card HD boot in QS/MDD Macs (One G5 Report OK): In reply to the earlier item below on Adaptec's "Known Issues" pages on the 29160/29160N/39160 not being bootable in Quicksilver, MDD and newer Macs, a Quicksilver wrote with an Openfirmware mod workaround (reportedly from the Adaptec site - Why Adaptec doesn't note that tip in their known issues page item on the boot issue is baffling.) A MDD G4 owner also noted using an OF mod to make the 29160 bootable in that system. (Update A G5 1.6GHz owner later wrote this did work with a 29160 SCSI card in that system, which was not bootable previously.)

" This info is on the Adaptec website and is what I used to make the (29160N) card bootable (in a Quicksilver Mac). Don't know if it will apply to the new G5's though.
-Robert B.

(Info on OF mod)

    To work around this issue print out the instructions that are below and follow them exactly:

    1) Reboot, holding down command-option-O-F all at the same time at startup. This will bring up a white screen starting with "Open Firmware" and ending with a "0 >" prompt.

    2) When the Open Firmware prompt appears, type the following: (Note that the spaces between quotes are important and the instructions are case sensitive. (The β character = a space. Wherever the β character appears in the lines listed below, type a space in it's place.) After pressing return , the Mac should display an "OK" response and another "0 >" prompt.)

    setenvβfcode-debug?βtrue

    3) At the "0 >" prompt type the following and then press the return key

    printenv

    4) Verify the following line exists in the output, generally 4 to 6 lines down

    fcode-debug?      true      false 
    

    5) At the "0 >" prompt type the following and then press the return key

    mac-boot

    6) Once the system is up, select appropriate drive (using Startup Disk in the Control Panels folder under Mac OS 9 or in Startup Disk in System Preferences under Mac OS X) and reboot.

If any G5 owners try this OF tweak with a 29160/29160N/39160 card, let me know if it works to make the SCSI drive bootable.

BTW - Although the first G5 owner that reported an ATTO UL4S SCSI card would not boot in a G5 (black screen only) didn't mention any damage done - another G5 owner wrote he had his system serviced by Applecare after trying an ATTO UL4S card in the system. I'll try to get more detail from both G5 owners on this.

Adaptec 29160 and ATTO UL4S SCSI Card Tests: (Sept. 5th, 2003) Now that G5s are arriving to end users, I should be getting more reports on PCI card compatibility. Although the 3.3v PCI issue noted previously (in Apple's specs/dev docs) is a prime factor - there's also potential for other compatibility issues. (Remember even the first B&W G3 models had problems with some PCI controller cards requiring firmware updates.)

" mike, a friend loaned me his brand new G5 1.8GHz for playing around with it for a day. i tried to install both my SCSI host adapters: "Adaptec PowerDomain 29160" and "ATTO UL4S". Both with the newest available firmware flashed to them.

With the ATTO installed, the G5 refuses to boot at all. I hear only the startup-chime, and that's it. the screen stays black. (I've seen gray screens at boot in some older macs years ago after adding a PCI SCSI card; zapping the PRAM helped, but this may be a more serious problem. ATTO had told me previously that all their Ultra3, Ultra320 and all Fibre Channel adapters should be G5 compatible. There was an ATTO OS X ExpressPCI Plus driver update (2.0) last month (see August 18th news page) that noted G5 Support, but it sounds like it isn't getting far enough to even load a driver. -Mike)

With the Adaptec, the machine boots, and the SCSI-disk attached to it is mounted and fully usable, although, the mac will not boot from the SCSI-disk.
(I asked if he tried cloning the G5s boot disk to the SCSI drive before trying to boot from it. Not sure the system restore discs would install to a SCSI drive and wondering if there's some file/driver that is required for the G5 boot disk that is not present on a standard/previous OS X install. However according to this Adaptec 29160 known issues page, you can't boot from hard drives connected to a 29160 in a Quicksilver, MDD G4 or newer Mac due to an Open Firmware incompatibility. They note the same thing for the 29160N and 39160 also. But a reader sent a note there is an Openfirmware mod to enable booting with these cards in QS/MDD models. Mike later wrote to say this OF tweak also worked in the G5 1.6GHz with the 29160 card.-Mike)
best regards,
Mike B.
(switzerland)
(he later wrote)

mike, yes, i did clone (panther 7b53) from the internal s-ata disk to the cheetah. maybe a clean-install would have helped. i used CCC v 2.2.
some additional discoveries: regarding the scsi-hostadapters: i did reset the pram. this did not help.

the whole machine is built very well! behind the motherboard is an additional heatpipe, i guess it is for the chipset. the cpu-heatsink has two heatpipes in it. the machine is indeed very quiet. the internal speaker does sound very, very cheap. the cpu-doughterboard is huge. it seems to use the same connector as the G4 cpu-boards. the second DC-connector near the AGP-slot known from the G4MDD mainboard is gone.
although not supported by apple, it should be possible to route a rounded LVD-SCSI-cable internally from the PCI-compartment up to the drives. the case design does not compare to anything i've seen before. i just can't wait getting my dual G5 i've ordered... just xbenched the machine equipped with the adaptec and a seagate cheetah 15K3. it scored higher than a stock dual 2.0GHz G5. see this xbench page
best regards, mike"

Below are previous comments on PCI Adaptec and ATTO SCSI cards, as well as Audio cards - but again the 3.3v PCI requirement isn't the only potential problem. Actual reports from use are the only way to check for that.
If any other G5 owners have used PCI cards (Audio, SCSI, etc.), let me know your results. I also welcome any other hardware/software compatibilty reports from G5 owners.

Comments on G5 compatible Adaptec PCI SCSI cards: Update: As noted above, you can't boot from hard drives connected to the PowerDomain 29160, 29160N or 39160 SCSI card in a Quicksilver, MDD or later Macs and they note no fix/firmware update to address this as of Sept. 2003. This is noted in the Adaptec "Known Issues" pages for the 29160, 29160N and 39160. Since there's been no firmware update since the Quicksilver model to address this, nor any control panel for OS X, no wonder people are wondering if Mac product development at Adaptec has ceased.
(from the Aug. 25th, 2003 news page)

"Mike, A follow-up on my past question regarding compatibility between existing Adaptec SCSI cards and the new PCI-X slots in the G5...
I talked to Adaptec and confirmed that only Adaptec's high-end "universal" cards will work in the PCI-X slots on the G5. Examples of such cards are their Ultra 160 and Ultra 320 SCSI cards, high performance RAID products and fibre channel cards. Their "everyday" SCSI cards (like the OEM cards supplied by Apple in many G4 machines) will not work because they are 5 volt only designs. PCI-X slots require 3.3 volt or universal cards (universal cards work at either 5v or 3.3v). (the G5 requires 3.3v PCI cards as discussed before here - see below for previous comments on ATTO SCSI cards and Audio PCI cards.-Mike)

This probably is not a big deal to most folks, since few people still use SCSI. But G4 owners like me who still use SCSI scanners and other external SCSI devices must either invest in a new 3.3v SCSI card ($250+) or new firewire and USB peripherals. Since I'm on the verge of placing my order for a dual 2GHz G5, I must now consider that additional cost in my planning. I guess that's the price of progress. Love your site (nothing else like it on the web)
-thanks, Steve"

I had written a past contact at Adaptec a few weeks back asking for a list of G5 compatible cards to add here, but no response to date. (It's possible that like several others I knew there - he's left the company now.)

ATTO comments on PCI SCSI card Compatibility: Update - see the ATTO Hot Topics page for later posted info on PowerMac G5 ATTO card compatibility. Here's a copy of that info as of Sept. 10th, 2003:

"ExpressPCI Host Adapter and Apple G5 Compatibility
Apple's new G5 platform introduces new levels of performance for professional and individual use. Among the many enhancements in this system is the addition of PCI-X bus support in several, high-end G5 models. This provides a huge data pipe on the expansion bus for today's most-demanding applications. The latest iteration of the PCI specification, PCI-X supports up to 1GB/second shared bandwidth across the bus. This provides up to four times the bandwidth of the PCI bus in the G4!
Many of ATTO's ExpressPCI host adapters support PCI-X throughput in OS X. For SCSI connectivity, the UL4D and UL4S Ultra320 SCSI adapters both support PCI-X. For Fibre Channel connectivity, the FC 3300 and FC 3305 host adapters support it. If you have purchased a G5 system with PCI-X bus, we recommend these cards for your connectivity needs and for maximum performance. ATTO will be releasing drivers to support each of these products on the G5 very soon.
Will ATTO ExpressPCI host adapters with a standard 33 MHz or 66 MHz PCI interface (not PCI-X) work in a G5 system? Certainly. ATTO has already released updated drivers to support our Ultra3 host adapters. These drivers are available on our driver download site. (Note to date two G5 1.6GHz owner reports on using a UL4S card didn't have good luck - one noted a black screen only at boot (card had latest firmware he said, it appeared to not get far enough to even load a driver, so not sure if the 2.0 driver would have mattered), the other claimed it resulted in having the logic board replaced, but that may have been a coincident failure or other problem not due to the ATTO card.-Mike)
Very soon we expect to release drivers to support our FC 3321, FCSW and FC 2600 host adapters in the G5 systems.
How does using a PCI adapter in a PCI-X slot affect performance? Using a PCI adapter in a PCI-X slot will result in a performance degradation since all slots on the PCI-X bus will revert to PCI speeds, either 33 MHz or 66 MHz PCI, depending on the bus and adapters you are using. As such, the UL3D-66 or UL3S-66 is the more appropriate match for the high-end G5 with a PCI-X bus because these will result in the least amount of performance degradation. The standard UL3D/UL3S host adapter is appropriate for the entry-level G5 because both support 33 MHz PCI, which results in a good match. Each of our PCI Fibre Channel adapters support 66 MHz PCI operation. For best performance, use these with the high-end G5 systems. However, they will work in the entry-level G5 at 33 MHz speeds.
Unfortunately, the ExpressPCI DC and PSC SCSI host adapters are not supported in the G5 because of signal voltage limitations.
Additional questions about G5 support should be directed to our technical support staff (available between 8am and 8pm ET by telephone at 716-691-1999 or e-mail at techsupp@attotech.com). They will be able to answer additional questions about G5 support and guide you to the best host adapter solution for your system."

(earlier info from the July 2nd, 2003 www.xlr8yourmac.com news page follows)

I wrote ATTO yesterday to ask if they could clear up the confusion over compatibility with their PCI SCSI cards and the new G5 systems with PCI-X slots (the upper 2 models have PCI-X slots, the low-end version does not). Here's their reply on the 5V/3.3V confusion and notes on compatibility:

"Mike- We've also noticed that the number of questions regarding 3.3V versus 5V PCI signalling have increased dramatically over the past several months. Within the past 3 months, several Wintel machines have released that are 3.3V-only PCI-X. Now Apple has joined the 3.3V only fray with the introduction of the G5. (1.8GHz and dual 2GHz models, 1.6GHz model has only standard PCI 33MHz/64bit slots-Mike)

Two things need to be considered to determine compatibility: motherboard slot and adapter slot. Motherboard PCI slots will typically be either 5V or 3.3V and are keyed as such. Our adapters are keyed for 5V only or have Universal connectors. We do not have any cards that are keyed for 3.3V only, but in the future that may change. Currently, the ExpressPCI DC and PSC cards are keyed for 5V and will work in any 5V PCI slot. However, since the G5 is 3.3V only, these cards will not fit in that system.
Our Ultra3, Ultra320 and all Fibre Channel adapters use Universal PCI connectors. So, these will work in either 3.3V or 5V PCI slots (or PCI-X slots, too, which are typically 3.3V). All of these cards will work in the existing Mac systems or the new G5.
Peter J. Donnelly
Product Manager, Host Adapters
ATTO Technology, Inc.
(He later wrote)
I am planning on putting together a tech note and posting it to our web site to better explain what works with what, but hopefully my abridged answer below helps explain things. If not, please let me know and I'll try to clarify. Also, when we release the tech note, I'll let you know so that you can link to it, if you'd like."

As I posted in a forum thread - as with any new design system, there could be compatibility issues not related to voltage/PCI slot type also - remember the early B&W G3 days (some PCI cards needed firmware updates back then to be compatible) and the problems seen with some PCI Audio cards in the first rev. MDD G4 towers noted here last fall, although the revised motherboard was supposed to solve those issues. (The issue of the Cinewave card overheating and failing warned by Promax last fall with MDD systems was another issue, but a reader replied to this post to say that Pinnacle for several months now has been sending a cooling fan to attach to the Cinewave PCI bracket for MDD owners to prevent the overheating problem for those that request it.)

More info on PCI Card Compatibility: (from the Aug. 12th, 2003 news page here) Here's another note on PCI card compatibilty with the new G5. (If in doubt check with your card mfr before assuming compatibility.)

"The remarks about Digidesign PCI card compatibility in the new Power Mac G5s are not entirely inclusive.
http://www.digidesign.com/compato/osx/g5/ (same page linked earlier today below)
Digidesign's ProTools 24|Mix Core and Mix Farm cards (discontinued) are listed as compatible, but not the older d24 (also known as Mix I/O) or DSP Farm PCI cards. This comprises the majority of high-end Digidesign systems typically found in commercial studio environments (as opposed to personal project studios).
David Dever, Creative Director
aka-librio / J. Warner Inc."

See the linked page for complete information but here's a clip:

"....MIX Core, MIX Farm, and all Pro Tools|HD cards are 3.3 volt-powered PCI cards which comply with the G5 PCI specifications. In addition to incorrect voltage, 5.0 volt PCI cards will not fit in the PCI slots on any G5 models. The affected Digidesign PCI cards are as follows:

    Digidesign PCI Cards Incompatible with Power Mac G5
  • Digi 001
  • Pro Tools|24 system cards, in Pro Tools|24 or Pro Tools|24 MIX systems:
    d24 (core card on Pro Tools|24 systems) DSP Farm
  • Pro Tools|24 MIX system card:
    MIX I/O (Note: this is different than the MIX Core and MIX Farm cards. See MIX I/O Overview)
  • Audiomedia III (Toolbox)
  • Pro Tools Project PCI
  • Project II
  • Disk I/O (Pro Tools III Core card)
  • SampleCell II

(they later note)
Digidesign G5 Compatibility
Digidesign is continuing to work closely with Apple to add support for the new Power Mac G5's to both Pro Tools TDM and LE systems, (using compatible 3.3 volt-powered PCI cards as well as our host based systems). Complete compatibility information will become available when the new Power Mac starts shipping - currently set for later this month (August 2003)..."

See the linked page for info on some hardware exchange options. (previous reader comments follow)

" i was browsing the mac audio forums at harmonycentral.com and I found something which may be of particular concern to people considering the purchase of a G5 for use with PCI cards. Apparently, the G5s will only accept 3.3V PCI cards... anyone with a 5v or 12v PCI card will be out of luck... like users of Digidesigns Digi001 system.

There is a document on digidesign's website discussing how asides from their ProTools HD systems, NO other PCI products of theirs will work on G5!!! I would imagine this to be true of many other manufacturers as well, but not many companies cough up specs on their PCI cards on their websites. Here is the link:
http://www.digidesign.com/compato/osx/g5/
I hope this information is of some value. Please keep up the good work! Your site has been an invaluable tool for a long time and hopefully a while to come.
Take care, -henry r "


MOTU PCI card Compatibility: (from the Aug. 15th, 2003 news page)
there's a page with Motu card compatability info at this Musicmac page.



G5 Specific Software/Applications/Driver Updates & Compatibility Notes:

IBM article on installing Yellow Dog 64-bit Linux on a G5 (from Aug. 4th, 2004 www.xlr8yourmac.com news page) Frank sent a link to an IBM website article on running 64-bit YDL on a PowerMac G5, intended for developers who want to try the early betas "in anticipation of production-ready releases to come". Includes information on installation and upgrading to 64-bit YDL.

Yellow Dog 64-Bit Linux: (from Apr. 8th, 2004 www.xlr8yourmac.com news page)

" gm mike, i noticed yellow dog now has 64bit linux for the the g5 available at ydl.net.
having worked on the development of redhat's and freebsd's 64bit os for the dec alpha, i would be very interested in benchmarks for the g5.
the links:
http://www.ydl.net
http://www.terrasoftsolutions.com
Curtis "

If any G5 owners have tried this version, let me know.

G5 Fan Control Update 1.0: (March 16th, 2004) Apple notes in the 10.3.3 update details that there's improved fan control for G5s. A reader sent a note they've also posted a Fan Control Update 1.0 for G5 owners running 10.2.7/10.2.8. Running Software Update in 10.2.8 here on a G5 shows the Fan Control Update 1.0 appearing:

" The Fan Control update improves the performance and reliability of the Fan Control system of the PowerMac G5."

BTW - the reader that sent the article here last fall on relocating the G5's Hard Drive temperature sensor wrote that 10.3.3 seems to have addressed that (i.e. recurved the response for the sensor to better cool the drives.) His report is also on the Reader Reports on OS X 10.3.3 Update.

Firmware Update for G5 1.8GHz Single CPU (late 2004) (11/15/2005) Apple's support site has download pages for this update for both 10.3.x and 10.4.x users. It should appear in Software Update also or you can use the download links below.

" About this Update
This Power Mac G5 System Firmware Update is only for the 1.8 GHz, (600 MHz bus), single-processor Power Mac G5 (Late 2004) computer. This update improves the reliability of your Power Mac G5, especially after it has been inactive for two hours.

Before you install the Power Mac G5 System Firmware Update:
Use Software Update to make sure you have Mac OS X v10.4.3 or later installed on your Mac. To use Software Update, choose Apple/ Software Update. The Software Update window opens and a list of available updates appears. Select the items you want to install, and then click Install. "

Here's pages to download the update:

IIRC, earlier this year they had a Kbase doc on this issue (recommending disabling sleep, use highest performance option, etc. in Energy Saver which some readers said didn't solve the problems with this particular model.)

G5 (June 2004 models) Firmware Update 5.1.8f7 (11/17/2004) Apple posted the update for June 2004 (not original 2003) models via software update and available for download at http://www.apple.com/support/downloads/powermacg5june2004firmwareupdate.html (says its for improved system stability.)

Power Mac G5 Uniprocessor Firmware Update 5.1.5f2 (9/27/2004) Apple has posted a revised Power Mac G5 Uniprocessor Firmware Update (v5.1.5f2) that hopefully will fix the problems some single processor G5 owners (i.e. 1.6GHz model) had with the previous 5.1.5 "f1" update. (The 5.1.5f1 update was to fix the sleep problem many G5 1.8GHz single CPU owners saw after the 10.3.5 update.) See the Apple page for more info but note this comment there on removing 3rd party RAM:

" Important Note:
Remove any third part RAM before installing this Firmware Update. After the Firmware Updater has verified that it has been completed correctly, you can re-install your third party RAM."

If any G5 owners that had problems with the 5.1.5f1 update tries this let me know if it helps. Thanks.

PowerMac G5 Uniprocessor Firmware Update 5.1.5f1 (for sleep issue w/10.3.5) (released Sept. 13th, 2004)

" Apple fixes the G5 1.8 SP Sleeping issue... Finally.
After a month of powering down my machine Apple fix the issue with a firmware update.
Note to the firmware, I've seen a few people say it does not work.

When I did the update, I first made my machine sleep from the Apple menu, but when I tried to sleep from the button, nothing worked. I opened up the Energy Saver settings and under options I noticed the"Allow power button to sleep computer" was no longer selected, once that was selected again everything worked like it used to a month ago.
-Andre R.
Related links:

  • Apple PowerMac G5 Uniprocessor Firmware Update 5.1.5f1 (page removed 9/28/2004 - see 5.1.5f2 above)
  • Apple forum thread on G5 uniprocessor firmware update "
  • Apple previously posted a Kbase doc on the sleep issue that many single 1.8GHz G5 owners reported after the 10.3.5 update.

    PowerMac G5 Firmware Update 5.1.4: (released Dec. 15th, 2003. Note as of April 2004 current G5s are shipping with v5.15, rev B Dual G5s have even later versions.)

    "PowerMac G5 firmware update 5.1.4
    This firmware update provides changes for initializing and running your Power Mac G5. It provides some performance improvements for some PCI-X configurations, patches a security hole and improves fan behavior in Open Firmware. "

    After downloading the update, a window with instructions pops up. After reading the instructions, press shut down (in the window/dialog), then power the G5 back up - holding the power button in until you hear a long Tone. The G5 will then update the firmware (power LED flashes) and reboot. Checking Apple System Profiler on the Sept. 15th shipped Dual G5 here (which had a very early v5.0.2f2 boot rom as I mentioned back in Oct.) now shows boot rom v5.14f0.
    NOTE: per reader reports - this does not address the Hard Drive Temperature Sensor (low reading) issue noted previously here in the article on Powermac G5 Hard Drive Heat/Cooling Tip.

    SoftRAID 3.0.2 Update released (G5 related bug fixes) (from their email PR on Dec. 1st)

    " SoftRAID, LLC today released version 3.0.2 of their software RAID application, SoftRAID 3. Version 3.0.2 has the same feature set as version 3.0.1, but fixes bugs which were related to Mac G5's. SoftRAID 3.0.1 purchasers will be upgraded automatically by email and should contact the company if they do not receive it promptly. SoftRAID 3 sells for $129 and may be purchased from the SoftRAID website at www.softraid.com. (There's also a page there on how SoftRAID differs from OS X's included RAID.)
    Demo version download URL: http://www.softraid.com/demo.html
    User Testimonials: http://www.softraid.com/testimonials.html "

    Logic Audio Platinum X and ProTools Updates: (Nov. 20th, 2003) MU lists the Logic Audio Platinum Updater X 6.3.2 (fixes for G5 duals/10.3 compat. and more - see page for complete list) and Pro Tools Updater 6.2r2.1 (adds support for Pro Tools|HD and HD Accel on a Power Mac G5s running Mac OS X 10.2.7 or 10.2.8.)

    Apple updates Final Cut Pro, DVD Studio Pro and Shake for G5 optimizations (Nov. 18th, 2003) Apple's Pro Apps page notes:

    "New versions of Final Cut Pro 4, Shake 3 and DVD Studio Pro 2, all fine-tuned to take advantage of the performance and architecture of the Power Mac G5 and Mac OS X version 10.3 Panther, are immediately available. Version 4.1 of Final Cut Pro delivers simultaneous playback of seven fully uncompressed standard definition video streams with real-time effects on a Power Mac G5 Dual 2GHz and Xserve RAID. Final Cut Pro 4.1, Shake 3.0.1 and DVD Studio Pro 2.0.2, with the new Compressor 1.1 update, are available for current customers as free software updates.

  • Edit 7 SD Streams with Final Cut Pro
  • Encode 55% faster on DVD Studio Pro
  • Render 45% faster on Shake "
  • See the sidebar there for links to each update.

    AltiVec Fractal Carbon Demo v1.2 Update includes G5 optimization (Nov. 17th, 2003) One of the most commonly used benchmarks (for FPU/Altivec) has been updated to include G5 optimization. The AltiVec Fractal Carbon Demo page has this note on v1.2:

    "New in version 1.2:
    this app has been optimized for the PowerPC G5. This code will also compute correctly on single processor machines and non-G4/G5's. This code uses the MacMPI_X.c library for communcations and requires Mac OS 10.2 or later, or Mac OS 8.5 or later with CarbonLib 1.2 or later, and Pooch."

    Previous versions ran fine in OS X 10.1.x also.

    G5 Optimized Cinebench 2003 Beta version: (Oct. 15th, 2003) Maxon has made a G5 optimized beta version of Cinebench 2003 available now.

    "CINEBENCH 2003 G5 is a technology study. MAXON has optimized crucial parts of the render engine for the G5 with the support of Apple, which now reaches speed increases of approximately 20%. Further improvements are expected once an optimized compiler with G5 support is available, which shall produce efficient and reliable G5 code. We expect new compilers and tools to be released during the first quarter of 2004, allowing us to deliver production quality G5-optimization for CINEMA 4D and BodyPaint 3D."

    A previous report from a reader from Germany said that CT magazine tests with an early G5 beta showed about a 25% gain in the render score compared to the original 1.0 release.
    Update - I've revised my previous page of Dual G5 tests to show include the results of this beta CB2003 version (in the benchmarks section). I saw an appx 25% increase in performance of the render score compared to the previous CB2003 version, although the scene flyby FPS rates showed less gain.

    10.2.8 Update for G5: Friday Evening (Oct. 3rd, 2003) the 10.2.8 update for G5s became available (noted on the main site news page). According to the "About the Mac OS X 10.2.8 (G5) Update", here's the fixes/changes which may help with some reported USB or Firewire problems. (One reader reported it solved his Camera freezing iPhoto when using a USB connection.)

      Enhancements These are some of the enhancements that are part of the Mac OS X 10.2.8 (G5) Update:

    • The computer will now wake from sleep if both an AirPort Extreme card and some kinds of PCI cards are installed.
    • iChat works better on AirPort networks.
    • Adds support for these third-party disc burners: Logitech Corp. LCW-E48F 48x24x48x, MCE Lucid-SD DVD-R/RW Drive, Formac Devideon FWD1496-3DVD-R 4x, TDK DVD+/-R/RW Drive - DED+440, LaCie d2 CD-RW FireWire 52x32x52x
    • Bluetooth devices are now available after the computer wakes from sleep.
    • Some FireWire devices will work better if you connect them after disconnecting certain kinds of digital cameras.
    • Keyspan Twin Serial adapters work better when connected to a USB 2.0 hub.
    • FireWire devices, including iPod, are now available when the computer wakes from sleep.
    • Prevents a kernel panic from happening if you start Classic with more than 2 GB of RAM installed.
    • Improves stability when using the "Return to command state" modem string command (;) with an Apple Internal USB Modem (v.92).
    • The Apple Internal USB Modem maintains better connections to some kinds of v.34 modems.
    • Includes updated security services and the latest Security Updates."

    For G5 owners that apply this update - please send your comments if you see any fixes to previous issues you had or any other pros/cons.

    IgniterX 6.0.7 Driver Update: Aurora Video Systems has released IgniterX 6.0.7 Driver for OSX (available at http://www.auroravideosys.com/support/manualsdrivers.html) that notes:

      10/01/03 1.8MB
      IgniterX Drivers for OSX
    • Corrected RGB Range issue on non-RT cards that could cause the wrong range setting to be used.
    • Corrected Ignition issue that showed the G5 processor as not being supported.

    Adobe Photoshop CS and Illustrator CS: (Sept 29th, 2003) The Apple Adobe Creative Suite page mentions G5 optimizations for Adobe's new "CS" (Creative Suite) Photoshop and Illustrator updates. (How much so over PS 7.0x with the G5 plugin update remains to be seen, although Apple comments "up to twice the performance of previous Mac models" running the same software.) For those like me that don't need the entire Suite, there are Upgrade versions of the individual apps - the Photoshop CS upgrade for instance lists for $169.

    M-Audio Revolution 7.1 Driver Update: As of Sept. 24th, there's a 1.2.7 driver update for the Revolution 7.1 at http://www.m-audio.com/support/driversearch.php. Here's M-Audio's driver page release notes:

      Revolution Drivers 1.2.7
    • [Driver] Support for PowerMac G5 with system version 10.2.7.
    • [Driver] Support for System Version 10.3 (Panther).
    • Driver/Panel] Support for 4.0 and 5.0 audio modes.

    Notes on Wintel/Boch X86 Emulator: (just a FYI for those that saw Wintel's comments about their GUI for the Boch Open-source X86 emulator noting G5 support - that probably means the GUI was recompiled perhaps - not the open-source Boch emulator.) As noted previously (below), Microsoft confirms that Virtual PC 6 is not compatible with G5 and it won't be updated for the G5 until the next major release.

    " I have a G5 1.8 Ghz and tried to install the emulation software. Basically what I found was that after Windows 2000 Professional (what I had handy) was copied and the emulation tried to boot the system, it stalled during startup. No movement of the progress bar after at least an hour...
    Richard H., Ph.D.
    Apple Certified Technical Coordinator "

    I've added Richard's comments to the previous page of Wintel/Boch X86 emulator feedback.

    DiskWarrior 3 (orig. release) and Drive 10 1.1.2 CDs Not Bootable from G5:

    " On another note, I tried booting my G5 from my DiskWarrior 3 and Drive 10 1.1.2 disks and both failed. This happened with the "C" key down and when I selected the CDROM in Startup Disk. In all cases the machine booted to the internal hard disk drive.
    Richard H., Ph.D.
    Apple Certified Technical Coordinator"

    I'm not that surprised about the CDs - in the past I've had to update my Discwarrior CD to boot from later macs that were released after the CD was made. I'd contact Alsoft and Drive10 to see if they are working on CD updates to address this.

    Xbench 1.1.3 Update: Xbench 1.1 and later have some optimizations for the G5 (and G4s tweaks) as well as bug fixes. Here's the revision history:


    1.1.3 Changes (Oct. 7th)

  • Fixed a CPU test crash on G5s
  • Fixed CPU and cache speed reporting on iBooks/PowerBooks under 10.2.8
    (Still reports L2 at 1/2 speed and L3 at 4GHz on some CPU upgrades in my G4 DA Tower here.-Mike)

    1.1.2 Changes (Sept. 29th)
  • Fixed CPU test crash
  • Minor changes to thread computation test
  • Fixed issue limiting RAM reporting to 2 gigabytes on G5s
  • Fixed L2 cache speed and bus frequency to show in GHz if applicable
  • Fixed a crashing bug during the text test

    1.1.1 Changes (Sept. 3rd)
  • More changes for the PPC 970
  • Improved vecLib FFT test consistency
  • Fixed issue causing OpenGL test to hang until clicking in window on certain machines
  • Fixed L2 cache speed and bus frequency to show in GHz if applicable
  • Improved optimization on AltiVec basic test
  • Better optimized stream tests for G4, G5 (thanks Chris Cox)
  • Added build number to System Version

    1.1.0 Changes (Aug. 23rd)
  • Optimized for the PowerMac G5, compiled with GCC 3.3
  • Avoid use of vec_dst on PowerPC 970 (G5) for better performance
  • Changes to the AltiVec test to reduce dependence on memory bandwidth
  • Removed use of vec_dstt
  • Changes to floating point test
  • Added vecLib FFT test
  • Added automatic machine type detection
  • Fixed an issue where the results submission sheet caused the results window to become catatonic
  • Prettier submit button
  • Recalibration of CPU tests, Disk tests
  • Fixed spinning rainbow cursor during OpenGL test"

    The first G5 (1.6GHz) scores posted with 1.1.0 version so far are at this page. G4 systems scores (including duals) are at this page.

    Download page for Beta IBM XL C/C++ Compilers for OS X notes G5 (from the Aug. 28th, 2003 www.xlr8yourmac.com news page)

    "Here's some good news for the G5 - IBM is porting their XL compilers:
    http://www14.software.ibm.com/webapp/download/search.jsp?go=y&rs=vacpp3
    (download page for 55MB Beta notes "Preview Beta Compilers for Mac OS X. IBM is preparing its industry-leading C, C++ and Fortran compilers to exploit the Apple PowerMac G5")
    From what I understand, they are very well thought of, and who better to write a compiler for the 970 than IBM?
    JJ"

    If you follow the links to download the beta (55MB) it requires you to register (or already have a registered user login).


    Virtual PC not G5 compatible until next full version update (Aug. 27th, 2003) Cnet has a story today noting Microsoft has confirmed that Virtual PC is not compatible with the new G5 due to current versions relying on a feature of the PowerPC G3 and G4 processors called "pseudo little-endian mode". The article says Microsoft has no plans to release an update for G5 compatibility before the next full version release.


    Adobe releases G5 Processor Plug-in Update for Photoshop 7.0.1 - (Posted Aug. 20th, 2003)

    "Adobe Photoshop 7.0.1 G5 Processor Plug-in update for Mac OS X

    This plug-in update is designed to optimize Adobe® Photoshop® 7.0.1 for use with the Apple Power Macintosh G5 processor. It makes the following changes to an existing Photoshop 7.0.1 installation:

  • Modifies many Photoshop operations to fully utilize the G5 processor capabilities
  • Replaces the Adobe Color Engine component (ACECarbonLib) with a new version that is designed for the G5 processor"

  • Logic Audio Platinum 6.2 Update for G5s: Update - As of Sept. 2nd, there's a 6.2.1 update available at their download page, but no info on what the .1 change was (6.20 to 6.21).
    (from the Aug. 19th main site news page) Ed sent a note that Emagic's news page today has a note on a Logic Platinum 6.2 update for G5 systems. Here's a small clip from that page:

    " Dream Team: Logic Platinum 6.2 and Power Mac G5
    With a simultaneous release date, Emagic offers the first Logic version optimized for the world's first 64 Bit personal computer, the Apple Power Mac G5. As the update to Logic Platinum 6.2 is available on the same day as the G5 itself, Logic Platinum customers can profit immediately from the remarkable performance of Apple's new flagship computer.

    This new Logic Platinum version is available in the form of the Version 6.2 update, and is a requirement when running Logic on the G5. All registered owners of Logic Platinum 6.x can now find this as a free download here.

    Musicians will experience a dramatic increase in both speed and performance when using the combination of Logic Platinum 6.2 and the G5. The ability to work on noticeably larger projects than ever before, including more software instruments, audio tracks and effects plug-ins, expresses more clearly than ever both the expanding potential of native music and audio production, as well as Emagic's and Apple's ongoing commitment to the professional creative community. "

    See the linked page for more comments on G5 benefits.

    ATTO ExpressPCI Plus driver 2.0 for OS X/Jaguar notes G5 Support: There's an ATTO OS X ExpressPCI Plus driver update (2.0) that notes G5 Support - see the August 18th news page for details and links. (There were other ATTO updates noted here - config utility and firmware updates for retail cards.)



    G5 Performance Tests/Articles: NOTE: For best preformance when running tests - make sure you select the high performance option for power management located in the System Preferences>Energy Saver>Options. See this Power Management note. (To rule out the system using a lower bus/cpu speed for power savings, although if the system is running too hot those settings will be overridden.)

    (NOTE: For later G5 vs G4 system performance tests, G5 graphics card tests, etc. see the Systems page, G5 section here.)

    More G5 vs P4 Itanium and Opteron Scientific benchmarks (from the 3/19/2004 news page)

    " Hi Mike, My friend Dr. Peter Hauschildt, now at the University of Hamburg in Germany, has compiled some more performance data for his code running on a bunch of different machines.

    There is general info about what he does at this link:
    www.hs.uni-hamburg.de/EN/For/ThA/phoenix/index.html

    Some info about his code and what the various numbers in the benchmark mean, as well as his comments about running on a bunch of different compilers and architectures is here:
    www.hs.uni-hamburg.de/EN/For/ThA/phoenix/computing.html

    And a some nice charts to visualize what you've just read:
    www.hs.uni-hamburg.de/EN/For/ThA/phoenix/documents/Performance.html He's getting a cluster of 25 G5 Xserves as soon as they start shipping, so we'll know more about how they do in large parallel jobs soon. The G5 running OS X is really shaping up to be a great scientific computing platform, let's hope Apple keeps the ball rolling (and it looks to me like they will). As always, thanks for your great site, I couldn't live without it!
    Mike C.
    System Administrator
    Dept of Physics and Astronomy
    University of Georgia "

    FFT (Fast Fourier Transform) Performance Tests (G4, G5, AMD Opteron, P4) (from the Nov. 12th, 2003 news page)

    " Hi, you might be interested in the benchmarks at: http://www.fftw.org/speed
    These benchmark the performance of a large number (~50) of real-world scientific code for computing the fast Fourier transform (FFT), an important and non-trivial algorithm for scientific computation. There are numbers for a 2GHz G5, a 2GHz Opteron, and a 2.8GHz Pentium IV, including codes that exploit SSE/3dNow/Altivec.
    Steven J. "

    I'm rushed here (working on other tests) but they have detailed results pages of test results for each CPU type there. The G5 (PowerPC 970) page noted an anomoly in one of the tests (size 1024 single-precision complex transforms which sometimes returned lower than normal results) but the 2GHz G5 delivered the best performance of any CPU listed there to date.

    After Effects (Pro) Benchmark page (From Oct. 1st news page) There's a page with info/downloads of a cross-platform After Effects benchmark (for 5.5 and 6, although the Pro version is required) at http://www.media-motion.tv/aebenchmarks.html At the bottom of the page is a listing of results that show (as of yesterday when I checked) a Dual G5 2GHz as the top system.

    AMD 64-Bit PCs vs G5s: (Oct. 14th) PCworld has a test results page of PCworld's article on 64Bit personal computers ("AMD 64 vs Apple G5 Systems"). The Dual G5 wasn't the fastest system on any of their tests (Premiere 6, Word, Pshop 7.0.1 and Quake3). See their notes below the table for test system comfigurations (most of the PCs they said used RAID striped drives for instance, but not sure how much that affected test performance).
    I noticed their Dual G5 Quake3 scores were lower than my results using the 1.32 "G4" Quake3 app with dual CPU support enabled. With the standard "high quality" settings in Q3 (no changes/no config file tweaks other than enabling dual CPU support via "r_smp 1") my Dual G5/retail 9800 Pro card delivers 340 FPS at 1024x768 (vs their 294) and 223.2 FPS at 1600x1200 (vs their 207) under 10.2.8. But there's been some rumors of the OEM 9800 Pro card being slightly lower clocked than the retail model which might have affected this.

    (I've built and used PCs since the late 80's, but in the last year I've just gotten fed up with all the repeated windows patches/flaws, etc., and now I don't really care how the performance of a Windows PC compares in those tests, I'd still rather have the G5. But the low parts costs, massive selection of options to 'build your own' PC make it attractive to millions. I understand that attraction very well.)

    PC Mag compares Dual G5 vs Dual G4 vs Dual 3GHz Xeon PC PC Magazine has a review of the Dual G5 2GHz that compares performance to a Dell Pecision 650 workstation with Dual 3GHz Xeon CPUs and a Dual 1.42GHz G4. You'll want to read all the pages (page 2 has apps test notes not on the results page) but the table of test results is on page 3. (The Avid Xpress Pro tests mentioned on page 2 are not listed on the results page however, at least not as of Sept. 20th.) Although they didn't mention it, hopefully they had the G5 Energy Saver set to "Highest" performance (see the power management/bus slewing note on the G5 Tech Docs page here about how this can affect cpu/bus speeds. Although it may have been running full speed in "Auto" mode during the tests, it's still preferred for performance tests to set it to highest performance.

    Dual G5 Games/Apps Tests: I've posted a page with Dual G5 2GHz tests with an ATI 9800 Pro retail vs 9600 Pro OEM graphics cards vs. a MDD Dual G4 1.25GHz in games and apps.

    Fluid Dynamics Benchmark - G5 vs G4 vs Itanium vs Cray (from the Sept. 15th 2003 news pge) Sean Garrick, (Assistant Professor, Department of Mechanical Engineering at the University of Minnesota) sent test results of a Fluid Dynamics benchmark comparing the performance (of one CPU) between a G5 1.8GHz, G4 1.42GHz, Dell 2.4GHz Xeon P4, a 900MHz Itanium2 SGI Altix, a 1.3GHz Itanium2 SGI Altix and a Cray X1. (The purpose of the test was to see how each CPU type compared - not a test of Multiple processors.)

    G5 vs G4 vs P4 vs Itanium (Astronomy) Benchmarks (from the Sept. 15th, 2003 news page)

    " Hi Mike, Here at the Univeristy of Georgia Department of Physics and Astronomy we have the privilege of working with one of the world's top Computational Stellar Astrophysicists, Dr Peter Hauschildt. He has written code called Phoenix that simulates conditions in stars and supernovae. You can see a slightly outdated description of his code at: http://phoenix.physast.uga.edu/
    He has run extensive benchmarks on many different platforms, and has some interesting results. I've included some numbers from recent runs on a Dual Xeon 3.06 (in Hyperthreaded mode), an Itanium 2 1.3, a G5 1.6 and a few other older machines for comparison.

    • Quad 1.3 Ghz Power4 IBM p690 (IBM xlf compiler) 77.1 seconds
    • Itanium 2 1.3Ghz (Intel Fortran Compiler) 95.5 seconds
    • Dual Xeon 3.06 (Intel Fortran Compiler, Hyperthreaded) 106.2 seconds
    • Pentium 4 2.66Ghz (Intel Fortran Compiler) 119.8 seconds
    • Athlon 3000XP (Intel Fortran Compiler) 121.1 seconds
    • G5 1.6 Ghz (IBM xlf compiler) 183.2 seconds
    • Dual G4 1Ghz (IBM xlf compiler) 255.5 seconds
    • Dual G4 1Ghz (NAG Fortran Compiler) 400 seconds

    We are really curious to see how the Dual (2GHz) G5 stacks up, as it potentially has better SMP performance than the Xeons, and may do pretty well. Ours is supposed to be here in mid-october, so I'll let you know then how it turns out. Also, it's really interesting to see how much difference IBM's PowerPC tuned compiler makes, the G4 looks pretty bad without it. One more thing to note, the Itanium seems very fast, but we have not verified that it gives correct answers yet, as he just got access to that machine last week.
    If you are interested I can send more detailed results as well, showing what areas each CPU stands out in.
    Mike C.
    System Administrator
    Dept of Physics and Astronomy
    University of Georgia "

    I asked to clarify if the beta IBM Compiler for G5s was used and he said yes. (IBM Beta compiler w/G5 support noted in the Aug. 28th news page here and added to the software/updates section of this page)

    MW G5 Comparison Tests: (from the Sept. 12th, 2003 news page) Macworld as a page with some tests of each of the G5 models. Macaddict also posted their initial benchmarks of a Dual G5 2GHz later that day, limited initially to just a QT/DV to MPEG4 tests (similar to my upgrade review tests) and some Photoshop 7 filter tests. (I'm hoping they retest the other machines with 2GB of RAM in them as well as note how much memory was allocated to Photoshop. Yes, Photoshop 7 has a preference for setting the maximum amount of RAM it will use.)

    G5 vs G4 vs G3 Codewarrior/GCC Benchmarks (from the Sept. 4th, 2003 news page)

    (from a reader email)

    " Hi Mike,
    Per your request for G5 benchmarks (requested real-world tests actually-Mike) here are some comparisons between G3, G4, & G5 for one specific compute intensive task. I would not have predicted some of the results. For example, better speed in classic and differences between CodeWarrior (which I thought used gcc ??) and gcc. The lowly 1.6GHz G5 came out looking pretty good though. What I havn't included here is that even the G4 trounces a 2.4 GHz P4 Xeon and some 900 MHz (many $$) Sun Blades :)
    Best regards, Brent

    Benchmark:
    - Task is a numerically intensive (double precision floating point) field simulation code.

    - Single task running except where noted. Disk I/O is excluded from timing. Small footprint (about 2.5 Megs).

    - iBook running 10.2.3, B&W and G4 running 10.2.6, G5 running 10.2.7.

    Case 1: CodeWarrior v8 compiler, G3 target, O3 optimisation, classic application.

                            Clock      Time      Speed
    Processor     Clock     Ratio      secs      Ratio
    G3 B&W      450 MHz      0.32       164       0.34
    G3 iBook    800 MHz      0.56        92       0.56
    G4 DP      1.42 GHz      1.00        56       1.00
    G5         1.6  GHz      1.13        37       1.51
    

    Case 2: CodeWarrior v8 compiler, G4 target, O3 optimisation, MachO (terminal) application.

                            Clock      Time      Speed
    Processor     Clock     Ratio      secs      Ratio
    G3 B&W      450 MHz      0.32       170       0.35
    G3 iBook    800 MHz      0.56        98       0.60
    G4 DP      1.42 GHz      1.00        59       1.00
    G4 DP      1.42 GHz      1.00        62       0.95 (2 tasks running)
    G5         1.6  GHz      1.13        42       1.40
    

    Case 3: gcc 3.1 compiler, G4 target, O3 optimisation, MachO (terminal) application.

                            Clock      Time      Speed
    Processor     Clock     Ratio      secs      Ratio
    G3 B&W      450 MHz      0.32       198       0.39
    G3 iBook    800 MHz      0.56        94       0.82
    G4 DP      1.42 GHz      1.00        77       1.00
    

    Observations:
    - No speed hit when running in classic (actually 5% faster on CWv8_G3O3 !) as opposed to OSX native (MachO).
    - CodeWarrior produces faster code than gcc.
    - Target specific compiler optimisation (eg: running G4-targeted as opposed to G3-targeted code on a G4) did not make much difference.
    - For both CodeWarrior and gcc compilers, no performance gains (and sometimes some degradation) from higher than O3 optimisation.
    - Dual G4 takes only about a 5% speed hit per task when running 2 identical tasks simultaneously.
    - G3 faster (between 0% and 46%) than G4 when normalise by clock.
    - G5 faster (between 24% and 34%) than G4 when normalise by clock.

    CAVEAT - These results and conclusions valid for this particular benchmark only ! "

    Reader Tests on AIFF-MP3 conversion with G5 vs G4s (from the Sept. 3rd 2003 www.xlr8yourmac.com news page)

    " I wanted to let you know about a test I did at the Apple Store in Glendale, CA. I called and they said they had a G5 1.6 on display, so I went down there with an audio CD to test the speed of an iTunes AIFF->mp3 conversion against other machines at the store. I ripped an audio CD track to AIFF format on each machine's hard disk to eliminate the bottleneck caused by optical drive speed. Here are the details and results:

    Audio CD: Herbie Hancock "Headhunters"
    Track: "Chameleon" - funk/jazz tune, 15 minutes and 44 seconds long, ripped to hard disk as an uncompressed 16 bit, 44.1 kHz stereo AIFF file using iTunes.

    Task: Using iTunes, convert the AIFF file to mp3 ("Higher Quality" setting, 192 kbps). I didn't have a stopwatch on me, so what I'm reporting here is the highest speed multiple iTunes reported during conversion. Machines tested (all had 256 MB RAM and iTunes 4.0.1 (117), and on all I quit all other applications, turned off every sound effect in iTunes, and unchecked "play songs while importing" and "create file names with track number"):

    • G5 1.6: - 23.2x
    • G4 Dual 1 GHz (MDD): - 20.6x
    • G4 1.25 (single processor): - 18.4x
    • iMac 1 GHz: - 13.1x
    • iMac 800 MHz:- 10.4x

    I'm surprised at how well a single 1.25 G4 holds up compared to a single 1.6 G5. If one speculates mathematically what a 1.6 G4 would do, it's 23.5x, which is the same as the 1.6 G5. I wonder how much of a speed increase we'll see when iTunes is optimized for the G5. Hope the above is useful to other readers of your site.
    Yarone
    AppleTLC"

    If any G5 owners have other real-world test results from production systems (not pre-release Dual G5, etc. - systems they actually own. i.e. iMovie Tutorial export, QT DV->MPEG conversion, FCP or FCExpress tests, etc.) let me know. (Include system details in reports - thanks.)

    Cinebench 2003 Comparison of G5 1.6GHz/G4 Dual 1.25GHz/Dual G4 1GHz/AMD 2000+ PC (August 22nd, 2003) I realize this is just a benchmark (not real-world tests) and its not optimized for the G5, so I'm just posting as a FYI. (See also the comments from Maxon below)

    "Last night I ran Cinebench 2003 on my 2001 Quicksilver dual 1 Ghz Powermac, my 2003 1.25 Ghz MDD PowerMac, and my homebuilt AMD 2000+ PC running Windows XP. I compiled the results with the 1.6 GHz G5 Cinebench benchmarks from Chaosmint.com and then posted them on my website at http://www.theandyzone.com/Computer/shootout.html. Take a gander. Feel free to link to it and/or send any comments.
    Andy"

    I'm not surprised by those results, as I've said before PCs have always scored higher in CB tests. Here's a related comment from a reader: :

    " Hi Mike, Just regarding the Cinebench scores - if you delve into the forum that Brucie posted earlier on today, you'll come across this maxon Cinema4D forum thread
    In that post, a guy called Richard from the Maxon development labs says this:

      "OK, some news directly from the MAXON development lab:
      Of course all the following numbers are not final, no promise at all !!!!!!!
      This is based on the information we have right, now, there is still a of of work to do and we still have to wait for a new compiler...
      With the current CineBench a single G5 1.8GHz scores at about 188, the optimized version will maybe score at about 238...

      A hypothetical single G5 2.0GHz could score at about 210 on the old CB, optimized could be 265...

      A dual G5 2.0 could maybe score at about 480 with the optimized version of CB...
      Depending on the new compilers and our findings (thanks a lot to Apple for being extremely helpful and cooperative) we might even crack the 500 score for the dual G5 2GHz...
      Again, no promise and of course no release date ;)
      Cheers, Richard"

    This compares to the 1.8ghz G5 score that Brucie posted of 188. Which means they're aiming to get a fair bit out of optimisation for the chip! Just as a means of further comparison...
    http://www.imashination.com/bench.html
    You'll see the top score these people have recorded is for a dual Xeon 2.4ghz - with a score of 502. If the G5s make it up to over 500, that'd be great.
    - james "


    Dual G5 2GHz Tests with PShop7, FCP4 and Soundtrack: (from the Aug. 15th, 2003 news page - note this was before Adobe released the G5 PS7 Plugin update) There's a page at soundtracklounge.com with Photoshop7/Final Cut Pro 4/Soundtrack tests using a "pre-production" Dual 2GHz G5. The comparison G4 machine was a single 933MHz CPU. (It would have been nice to see how it compared to a dual G4 1.25GHz or dual G4 1.42GHz model, but the results are still interesting.)

    After Effects (short) Comparison Test with G5 Dual 2GHz (from the June 26th main news page) I'm getting this 2nd/3rd hand, so I can't verify this, but just a FYI since in the past Mac vs PC After Effects tests (G4 duals vs PCs) showed the PCs much faster as I remember (reportedly Adobe's software didn't seem to take advantage of dual CPUs nearly as well as Apple's Final Cut Pro - but that app is not cross-platform). I personally despise the platform war debates (it brings out the worst in everyone it seems) but here's an email I received today. (Note also it's not a full Night Flight test - only the first 10 frames.)

    " Hi Mike, I belong to the After Effect-List, which exchanges information, tips, and techniques for both the PC and Mac versions of Adobe After Effects. One of the listees was at WWDC, and ran a specific AE project that we all use to evaluate processor speed and performance.
    Here is a copy and paste from his post this morning (Thursday):

    ==============================================
    "Hey everyone, I've been at Apple's developer conference and had a chance to install and try out After Effects on a new G5.
    I ran the Night Flight file that has come to be the standard for AE benchmarking. Since I didn't want to sit there and watch it render for hours, I ran just the first 10 interlaced frames from the project's pre-set render queue...
    http://www.aefreemart.com/tutorials/3DinAE/nightflight/nightflight.html
    Here are my results for this test on the three computers I have available to me:

    1 x 1.0 GHz G4 PowerBook 17" - ~30 minutes (3 min/frame)
    2 x 2.66 GHz Pentium Xeon from Boxx - 11 min, 39 sec (1.2 min/frame) (actually 1.165 min/frame)
    2 x 2.0 GHz PowerMac G5 - 6 min, 1 sec (0.6 min/frame)

    I ran the Xeon test on a couple different identical machines to make sure mine wasn't just running slowly, but got identical results.
    Of course my Mac bias is well-documented, but I'm sure many people here can vouch for me as an honest person. If the results had gone the other way, I'd just keep my mouth shut and let someone else break the bad news.
    Other observations about this test that may ultimately work in the Mac's favor:

    1) The machine was not running 64-bit Panther, but only a tweaked version of 32-bit Jaguar. Likewise, AE is obviously not yet compiled to take advantage of the G5 chip in any way. Both or these situations will automatically be rectified in the future.

    2) Night Flight is very CPU-intensive, but not very disk I/O intensive. I think the 1 GHz system bus and other details on the G5 will provide greater gains for typical projects that rely more heavily on I/O."
    ==============================================

    All pretty interesting!!!!
    DAVID S.
    pixelcraft studios "

    Another mail from an AE List subscriber said the performance test above was from Brendan Bolles, a respected AE plug-in developer and included a direct link to the post at http://www.fido.se:8100/Lists/AE-List/Message/36461.html.
    It will be interesting to see the results of more complete tests when the G5 systems ship.

    G5 vs G4 Java Benchmark Performance: (from the June 30th, 2003 main site news page)

    " Hello, I write you to try to clear some things up a bit. In many forums there are arguments about how the G5 compares to the G4. The Xbench results may have caused the G5 look a bit poor against the G4. (this may be due to compiler or issues with some instructions on the G5 mentioned in the Apple Tech note/G5 Performance Primer-Mike)
    Since I was at the live streaming event from Berlin to SF, I was able to test the new G5 Dual 2Ghz a little bit.
    http://homepage.mac.com/jdonath/Apple_Event_Berlin/PhotoAlbum41.html
    The benchmark I was able to run is a web based Java Linpack benchmark.
    http://www.netlib.org/benchmark/linpackjava/. While it is actually a benchmark of how good the implementation of Java is on the specific OS it is a good way to compare the old G4 to the G5.
    http://www.netlib.org/benchmark/linpackjava/timings_list.html.
    In this list are the results of different CPU's performing this bench. If you scroll down at around 90Mflop/s you will find the top of the line G4 Dual 1.42Ghz.
    My machine a G4 933Mhz scores 66Mflop/s under OS X 10.2.6 or under 9.2.2 77.4Mflop/s. If you now scroll to the very top of the chart you will find the Dual 2Ghz G5 scoring an amazing 236Mflop/s.
    This benchmark is very dependant on the memory bandwidth, this is why nearly all P4's score very high.
    Despite the fact that the Java implementation lacks a bit under OS X it clearly shows the power of the G5. This benchmark is in no way optimized for the G5 and also lacks SMP features. But the real strength of the G5 would be demonstrated in a Linpack double precision 1K x 1K because of its 64Bit capabilities.
    Regards, Christian D. "

    One reason for some lower-than-expected G5 results in Xbench (and some other apps/benchmarks perhaps) may be due to the instruction issues noted in Apple's Tech note: G5 Performance Primer noted here in last week's news page. (compares G5 to G4 architecture, notes on deeper pipelines in the G5, problems with some instructions, pipeline 'bubbles' and dos/don'ts for G5 programmers.)


    The Fast and Furious G5 Case Mod:

    " New G5 case mods already in the works:
    http://homepage.mac.com/kristofor/images/g5mod.jpg
    Thought you might appreciate the lighter side. Made it myself. ;)
    Kristofor C. "

    G5 intercooler and Nitrous optional... Just needs a few more decals and lowering the case until it drags the table when moved then it'll be ready for "Super2NR TV" or "High Rev Tuners"...

  • Click for 6G SSDs!


    OWC XYM SPECIALS!
    (Ad/Sale items)

    = UPGRADES by Mac =
    Upgrades just for
    YOUR Mac!

    = Used Macs =
    (Click for List)

    SSDs from under $50!
    Fast SSDs for Most Macs/PCs

    = ThunderBolt =
    Drives & More

    = HARD DRIVES =
    Up to 6 TB HDs!
    Hitachi, WD, Seagate, Samsung HDs

    = 2.5in HDs & SSDs =
    Notebook Hard Drives and DIY drive/case kit bundles

    = MEMORY =
    Lifetime warranty RAM Upgrades!

    = OPTICAL DRIVES =
    Internal and External Superdrives/Blu-Ray drives

    = VIDEO/LCDs/TV =
    Graphics cards, TV tuners, DVR, adapters and more.

    = SOFTWARE =
    Apps, Utilities, OS, VM, Games and more

    = WIRELESS =
    WiFi and Bluetooth Devices/Adapters/More

    = Repair Service =
    for iPhone, iPad, Macs

    = iPad/iPhone/iPod =
    Accessories, Cases, Repairs & More

    NuGuard iPhone Case *Extreme* Drop Tested!


    XLR8YourMac T-Shirts


    FasterMac.net
    ISP/Hosting

     
    = back to www.XLR8YOURMAC.com =


    = Other Site Topic Areas =
    Mac Mods/Upgrades | CPU Upgrades | Storage | Video | Audio/HT | Misc/Software | Search | Recent


    Copyright © , 1997-2014. All Rights Reserved
    All brand or product names mentioned are properties of their respective companies.

    Legal: Site Privacy and terms/conditions of use.