"MCE Technologies, LLC, an Apple Developer, today announced an internal CD-RW/DVD-ROM drive upgrade for the PowerBook G4 (Titanium) 400MHz and 500MHz models. The MCE PowerBook G4 Combo Drive upgrade is designed to replace the original CD-RW-only or DVD-only drive that shipped inside those PowerBooks. Customers may ship their PowerBooks directly to MCE or they may have the upgrade done at an Authorized Upgrade Center. MCE will begin accepting PowerBooks for upgrade on June 3, 2002.

"The MCE PowerBook G4 Combo Drive upgrade enables customers to burn CD's AND play DVDís on their PowerBook G4 without the added weight and hassle of extra hardware and cables to drag around." said Arnie Ramirez, President, MCE Technologies. "It also brings them up to par with the currently shipping PowerBook G4 models which ship standard with a combo drive."

The MCE PowerBook G4 Combo Drive upgrade is a slot-loading CD-RW/DVD-ROM mechanism that replaces the original CD-RW-only or DVD-only drive that shipped with the 400 and 500MHz PowerBook G4 Titanium. The drive writes to CD-R media at 8x speed, to CD-RW media at 8x speed, reads CD-ROM discs at 24X speed and reads DVD-ROM discs at 8x speed. The drive is bootable, and is compatible with both Mac OS 9 and Mac OS X as well as Appleís iTunes, Disc Burner, and DVD Player software. The MCE PowerBook G4 Combo Drive is also certified by Dantz Development as compatible with Retrospect.

The MCE PowerBook G4 Combo Drive has a retail price of $299. Customers may arrange for the upgrade via world wide web or by calling MCE Technologies directly. Turnaround time is expected to be one to two business days. Repair and upgrade shops interested in becoming an Authorized Upgrade Center for this product should contact us either by phone or send an e-mail to dealersales@mcetech.com "

The slotted drive I suspect should also fit the G4 Cube (which uses a notebook drive with a standard ide cable/power adapter).

Please use caution when installing the service pack 1 if you are using the Microsoft custom enterprise agreement (CEA) version of Office X.
The CEA versions do not have individual CD Key numbers - in fact, here at University of Michigan, none of the Office install disks either list the CD key nor does the installer ask for it at any time.
After installing the SP1, however, Office asks you for the CD Key at first launch. Doh! It took a 15 minute call to Microsoft Corporate Sales to explain the situation before I was able to get an appropriate key from them.
But - since they supplied us this morning with one key, I do not yet know what will happen when multiple machines (potentially hundreds) on the same network all try to use Office at the same time. The CD key-less CEA version originally did not give warnings when multiple copies were in use on the same network, unlike the retail version.
I can let you know later today or tomorrow about what will happen once we update more machines.
Thanks!
Mike S., MD"

Here's two more emails noting the problem (again about a *dozen* readers have noted they did not have any problem with the update or having to re-enter a S/N).

"Same problem here, Mike -- CD key worked for Office v.X 10.0 install. After I installed SR1, upon launching any Office app I was prompted to enter my name and CD key. Entering the original CD key resulted in an "invalid key" error.

In order to re-install 10.0 and have it actually work, I had to delete both the Microsoft Office v.X folder in Applications, and the Microsoft folder in [username]/Library/Preferences. Deleting only the folder in Applications allowed for a "successful" re-install of 10.0, but then none of the apps would launch. To get a working installation I had to delete that prefs folder in my library folder.

One more thing: when the original Office v.X came out, there were hundreds of reports on MacFixit and across the Mac Web about this same issue from folks who'd ordered Office v.X as an upgrade to 2001. In that case, MS had apparently shipped out invalid CD keys.
Perhaps in this case MS is trying to weed out pirated versions by making everyone call them for a new CD key?
Matt"

And the only other report of a S/N problem:

" I had the exact same thing happen this morning on my Office X install.
It wouldn't accept my SN number. My SN number hasn't been outside of my CD storage ... crate, so I know MS didn't block it because it was a known number.
Mark B. "

(Original reader email first noting the problem earlier today)

"The sr1 update resets your office so you have to type your serial # again. Here's the problem, after digging through all my boxes of computer stuff to find my key # on the back of my cd case, it said mine was not a valid cd key! So I backed my Microsoft user data folder in my documents folder, reinstalled office x and entered my key and it worked but then when I tried the sr1 update again it said my key was still invalid.
I have not given my cd key to anyone but the cd case did sit on my desk for several months, so I suppose someone else could have used my key from it. My office does not have a network so I would not be able to tell by the multiple office apps with the same key # stopper.
I have not yet called Microsoft to see what they will say, because I do not have time to wait an hour on hold today. So I will just reinstall office x and not update it with the sr1 update for the moment.
Christian G."

"... i was reading the read me from the insall update of Office X. And you know what, in the chapter 3.1, 3.2 - they speak of 10.1.5, saying that with it Excel and word can use antialising. So we may expect this update for the days to come.
Greg"

There's been rumors that 10.1.5 may also solve the Radeon 7000 PCI card issue with OS X in Beige G3 and older Macs (noted here in mid-Jan. at the Radeon 7000 card feedback page) and might have other graphics chip related surprises...

"IBM has released some technical documents about 750FX.
750FX technical summary
PPC750FX Differences Document

According to the documents, 750FX has the same pipeline as previous G3 processors. The performance advantages of 750FX are,


• 512KB on-die L2
• Dynamic frequency control with dual PLL
• Wider L1 data cache reload bus (256-bit); same as 750CXe.
• Wider L2 reload path (256-bit)
• Additional FPU reservation station; same as 750CXe.
• Higher precision results from the reciprocal estimate instructions; same as 750CXe.
• Enhanced 60X Bus: pipelines back-to-back reads to a depth of 2.
• 150-200MHz FSB with 1.2-1.5V low-voltage signal
• TAU [thermal assist unit] support
• 27 Bus ratios from 2 to 20
The documents also note some features have not been implemented in initial revision.
• Miss-under-miss L1 data cache
• L1 cache and tag parity, L2 tag parity
• L2 support for 2 outstanding misses (1 data and 1 instruction; or 2 data)

And, 750FX still lack MESI coherency protocol.
M. Isobe "

I tested Logic Audio Platinum 4.7.0 and 4.8.1 on my new PowerBook G4 800, 512 megs ram (standard middle model on Apple Store). I could not test Logic 4.7.3 (since there were no updater to that vers. on emagic.de), but I got the same results from 4.7.0 and 4.8.1 so this should be the same numbers I would have gotten from 4.7.3.

I used Mac OS 9.2.2 with VM turned off of course. Stereo Platinum verbs were used with standard settings. I used an audio file about 3 secs long which was looped. [He later wrote that using non-looped, separate files still produced the same results.-Mike]

I managed to get 18 Stereo Platinum verbs. When turning on 19th the processor choked, but 18 played fine when I looped it for a while. These numbers translate to about 11 (11.25 to be exact) platinum verbs on this processor running at 500 mhz, so the first PBs processor is still a bit faster mhz for mhz (or perhaps rather memory throughput is better). Not that it matters though, since there are no such processors in PBs running at 800 mhz.

(For those that missed the past reports on PB G4s - here's John's results using Logic 4.7.3-Mike)

• PB G4/667 - 9 platinumverbs
• PB G4/550 - 7 platinumverbs
• PB G4/500 - 13 platinumverbs

(He later wrote a G4/450 (@470+MHz) upgraded S900 ran 6 platinumverbs and had a GaugePro memory bandwidth score of just under 70MB/sec.-Mike)

So there you have it, the best Apple portable for music production!
/John
(I asked about the bitrate, etc. of the sample file)

I used a relatively clean version of OS 9.2.2 with lots of unnecessary extensions off. And no Palm Desktop software. It could possibly have been tweaked a bit more, but in my opinion I wouldn't have gotten another platverb if I had spent time doing that.

The audio file was indeed 16bit/44khz. The Soundmanager version is the one that is installed with OS 9.2.2, version 3.6.7, the driver is Mac AV and as far as I know, it is not possible to work at a higher rate than 44khz or perhaps 48khz with that driver. Or higer bitrate than 16bit. The other tests on PowerBooks should also not have been at any higher rates than 16bit/44khz so my test should be comparable to those.

I've not done extensive research on the bitrate/khz issue, since I never work with higher rates than those. I guess it depends on what the final sound is used for.
John
(I asked for ideas why the test seems so bandwidth hungry)

About the test being bandwidth hungry. I'll start out by saying I'm not really an expert on these things, but:
As far as I know, doing reverb is a very complex task. If you would want to mimic a rooms sound characteristics exactly it would take trillions of operations which would of course not be practical for real-time reverb. This is due to the sound bouncing off of everything in the room changing it's characteristics every time, so basically the sound that reaches your ear is made up of an extreme amount of different variations of the original sound. I guess it could be compared to ray-tracing when doing 3D.

Instead the audio stream is going through the cpu several times, and each time reverb is applied. All streams that go through the cpu are saved to produce a final reverb sound. So I guess this could be the cause of it being bandwidth hungry, because essentially one sound becomes many different ones and every one is copied and sent through the cpu again to produce a new sound. In the end all sounds are combined into an output audio stream. This ought to require lots of memory AND bandwidth. Also, when considering this is done with the new combined sound you start seeing why it is such a complex task.

If someone knows better than me, please enlighten us all. I'm very interested in this subject. Also, another thing I'm considering is contacting emagic to see what they have to say about it. "

I asked the reader that submitted results of the Platinumverb tests with the PB G4/500, PB G4/550 and PB G4/667 last December for more info on this test settings, etc.

(do you remember the settings/bit rate, etc. for your tests?)
Not really, but I suspect it was interleaved stereo files, 16-bit (?), 44.1kHz.


(The PB G4/800 owner said he used 'default settings' and a 3 second file looped for the test - I'm no audio guy so don't know how that compares to the tests you ran/sent last December. )

I was using continuous files, not short looped regions, but I doubt that would affect much aside from disk throughput.
(Note the PB G4/800 owner later wrote using non-looped/separate files didn't change the results-Mike)


Ask him which audio driver he was using - Mac AV? [Yes, he did use Mac AV perf comments above-Mike] And what settings were checked in the driver's settings panel in Logic. It's possible to tweak a little more performance by using larger process buffers, etc, but in my experience the difference is minimal - less than 1 extra Platinumverb's worth of grunt.


(if you remember any details/settings, bitrates, etc. from those tests, let me know.)

The test is to create a stereo track, add a stereo audio file to that track, and start stacking up stereo Platinumverbs in the track's insert slots. When you fill that track (8 inserts), make another stereo track, use a _different_ stereo audio file on it, and add reverbs to that. The files should probably be interleaved stereo, 24-bit, 44.1khz, I'll run my test again using those.
John Pitcairn"

I asked another reader about why with 16bit/44KHz files there seems to be such a demand on bandwidth (as another reader pointed out, 16bit/44KHz files are only about 172KB/sec). I've been meaning to ask the developers of Logic Audio about this also but haven't yet - too many other irons in the fire every day and night...

" I believe one of the reasons for the very high demand on memory bandwidth and CPU is that most professional audio applications (i.e. Logic and Cubase) processes audio internally as 32 bit floatpoint files, regardless of the bitrate of the original file (see this link:
http://www.emagic.de/english/products/logic/lap.html )
This means that the Audio app besides processing the original audio and routing it to the reverb plugin also have to do at least two times bit conversion (back and forth) now what if the reverb plugin processes audio in 24 bit? more bit conversion back and forth...

So I believe signal path is more like this
pr. audio track (stereo is two tracks):
reading from HD into RAM buffer->out of buffer-> into CPU (bit conversion)->out of CPU-into Audio App RAM-> (send to Reverb plugin) ->into CPU->out of CPU->into reverb plugin times 2 (unmodified signal + modified audio "signal" for mixing unmodified with modified) times number of reverbs ->into Audio app->into CPU (bit conversion again) ->out of CPU->Audio card DA converter-> out (Amplifier/headphones)...
pfew that was a long explanation, and maybe the signal path back and forth is even longer..... I guess reading/processing/reading is true for simply playing audio, like in iTunes..

I must note that I haven't been able to confirm these theories of mine about the signal path with any developers, so it is just my "enlightened" guess based upon my experiences, my knowledge of how audio apps work, and my knowledge of real world hardware setups and signal routing (between tape recorders, mixing consoles and effect racks)

I believe processing (rendering) reverb is in some ways comparable to 3D rendering with Radiosity on (reverb is a kind of 3D sound) meaning that every piece of audio has to be processed more than once: audio signal leaves the "loudspeaker", spreads and hits the walls of the "room" and gets reflected and hits the "microphone" this repeating for as long the user have configured the length of the reverb, and the size of the "room".
(once again an "enlightened" guess")...
Chakaranda
Copenhagen..etc "

(For my first impressions on the new 2002 PowerBook G4/800, including some performance test comparisons in iMovie2, Photoshop 7 and 5.5, 2D scrolling, Quake3 performance, etc., see this article.)

A reader asked if anyone had tried the Samsung 240T 24" LCD display (1920 x 1200) with a Mac using the DVI port. (Most Mac DVI port cards show a max DVI res. of 1600x1200 - although Apple lists some of these cards as 23" Cinema (1920x1200 display) compatible with their DVI->ADC adapter.)
If you have tested the 240T using the DVI interface with a Mac, please let me know (include info your mac model and graphics card used). Thanks!

USB PenDrive Reports:
• Kanguru 256MB USB Flash Drive used w/Powerfbook G4 (OS 9.2)
  (listed under 'other' brands)
  
  DVD-R DVD Recorder Drive Reports:
• IDE Pioneer DVR-104 in B&W G3 (OS X)
• IDE Pioneer DVR-104 in G4/AGP (OS X)
  
  DVD-R + DVD-RAM Drive Reports:
• Firewire Que (DVD-R/DVD-RAM) used w/G4 AGP (OS X)
  (problems noted)
  
  DVD-ROM Drive Reports:
• IDE Artec 16x in GB4/AGP (OS X)
  (listed under 'other' brands)
• IDE Lite-On 16x in B&W G3 (OS X)
• Firewire Sony 16x/40x [IDE drive in FW case] used w/Dual G4 (OS X)
  [noted setting drive to Cable Select solved issues w/ADS Pyro case]
• SCSI Pioneer U03S [didn't note speed] in 6400 (OS 9.1)
  (asked about drivers, patches, etc.)
• IDE Sony DDU1621 (16X/52X) in B&W G3 (OS 9.2)
  (FYI CD install guide here covers 8600/9600/Beige G3/B&W G3/G4s thru Digital audio)
  
  Combo DVD+CDR Drive Reports:
• IDE LG 4120B (12x8x32x + 8x DVD ROM) in B&W G3 (OS X)
• Expansion bay Matshita UJDA710 (8x4x24+DVD ROM) in PB G3 2000 (OS X) [used guides here posted last year]
  
  IDE CDRW Drive Reports:
• Philips 241240 (24x12x40) in B&W G3 (OS 9.2)
  (not bootable w/OS X CD, no itunes 9 support)
• Philips 32x10x40x in Beige G3 (OS 9.2)
  (noted 'interface unstable' errors)
• Sony CRX120E 4x4x2 in Beige G3 (OS X)
  
  Firewire CDRW Drive Reports:
• Acer 24x10x40 (in FW case) used w/G4 AGP (OS 9)
  (discs won't mount - asked if he installed Toast)
• EZquest Boa 40x12x48x used w/Dual G4 (OS X server)
  (drive mech. is Lite-On, supported by iTunes/DB)
• Iomega Zip 650 ATAPI 12x4x32 in FW Case used w/iMac slot loading (OS X)
  (no itunes support)
• Lacie U&I 32x10x40x used w/B&W G3 (OS X)
  
  SCSI CDR/W Drive Reports:
• Lacie 12x10x32 (sanyo mechanism) in 6500 (OS 8.6)
• Teac R-55S (4x12) in Performa series (OS 9)
• Yamaha 8424S (8x4x24x) in 95/9600 (OS 8.1)
  
  IDE Hard Drive Reports:
• Maxtor 740DX-6L 60GB in B&W G3 [rev 1] (OS X)
• Seagate ST380020A 80GB in G4 no AGP [yikes] (OS X)
  (Reader FYI Illustrated guide to adding a 2nd HD here in B&W G3 rev 2 to G4/AGP systems)
• Western Digital 1200JB 120GB in B&W G3 using IDE PCI Card (OS X)
  
  Firewire Hard Drive Reports:
• EZQuest Cobra+ 60GB used w/Umax S900 (OS 9.1)
  
  Removable Media Drive Reports:
• SCSI Iomega ZIP 100 (external) used w/Dual G4 (OS X)

You can find full owner reports (latest shown first) by searching the database by drive/brand/interface/mac models (the latest reports are shown first in searches).
For guides to installing CD/CDRW/DVD drives or Hard drives in many mac models, see the IDE Articles page. The Firewire articles page also has guides on case kits, installing drives, etc.
If you've added a IDE, SCSI, Firewire or USB hard drive, CDRW, tape drive, etc. make sure you add a report to the database. [If you post an updated entry - make sure you use the same name, etc. as you did before so I can find your past entry. Thanks.]

(Entries today will be added the following newsday. Incomplete entries are deleted. Do not post questions in the database, it's for drive reports not questions on what drive to buy - for that try searching the database for reports from owners of your mac model on the drive type/brand/interface, etc. you're interested in.)

• Other World Computing G4/500 in Beige G3 (rated 8)
• Other World Computing 500Mhz in Beige G3 (rated 8)
  (didn't note if G3 or G4 CPU)
• PowerLogix G4/533 [550] in B&W G3 (rated 8)
• G4/400 AGP OC report

(Warning - Overclocking may not be reliable and could lead to hardware failure or corrupted data.) You can find the full reports by searching the database selecting the indicated Mac model and upgrade card brand/type. If you've upgraded the CPU on your Mac, please post an entry in the database. Search the database for entries from most every upgradable Mac model *before* you buy. (Searchable by mac model/upgrade brand). For detailed reviews with performance tests and install tips, see the CPU Upgrades page.)

• 1 year ago today
• 2 years ago today
• 3 years ago today