Last update for the night - more news tomorrow at Noon.

Mac Treasure Tracing Club sent the URLs to what I assume is the company reported to be making the new iMac notebook/consumer portable for Apple. In a quick scan I did not see any Apple specific info there but here is their message:

" It is addition information.
Alpha Top shrine that AsiaBizTech reported is a subsidiary mobile product of GVC company. http://www.gvc.com.tw/eng/gvc/index.htm

GVC Group map
http://www.gvc.com.tw/eng/gvc/home.htm
Mac Treasure Tracing Club
http://www.tcp-net.ad.jp/danbo/ "

CarrierZIF Tech Note: I had gotten two comments today that the XLR8 CarrierZIF was not being shipped due to an issue with some Motorola ZIFs and Macs with interleaved memory. My tests so far were in the PowerCenter Pro - which does not interleave memory and I don't own a Motorola made ZIF of any kind at the moment. When Bottom Line sent word they had stock I wrote them, despite the fact their tech told me they were having great results so far (even in Macs with interleaved RAM). At the time I had no real details on why a Motorola ZIF would matter. Apparently only 'X' series Motorola ZIFs (not standard production versions) have the interleaved RAM issue with the CarrierZIF. I had mailed Gary Dailey of XLR8 on the issue this morning and the following is his reply:

Mike,

Here is the information you've asked for.
The CarrierZIF is an extremely solid product as verified by our testing, and yours. [My tests so far were using a non-interleaved Mac and not with these 'x' series Motorola ZIFs-Mike] 100% of the production qualified chips tested to date have worked without fail. The CarrierZIF allows us to test a wide array of CPU daughtercards.

As stated in all of our documentation we recommend Apple branded or XLR8 branded PowerMac CPU G3 ZIF upgrades only. Now, onto some recent conversations...

You have referred to a problem with some Motorola ZIF cards. We do not currently recommend nor support these cards. Our understand is that these chips are from a release of a multitude of prototype (X) chips that Motorola has released through some of the smaller mail-order companies. I have included the Motorola documentation on these chips.

These (X) chips have problems that manifest in certain edge situations in some machines. This problem is not evident on CPUs shipped to regular 750 specification, or any IBM CPUs.

The problem that you are referring to on some Moto ZIFs were duplicated in our labs. We can duplicate this on about one in 50 card/Motorola ZIF combinations. At the request of some dealers "burdened" with these prototype Motorola ZIF cards we are testing and working on a 100% perfect fix.

At this moment, the work-around is De-interleaving. [Not having like Dimms in Ax/Bx slots prevents interleaving-Mike]
Although we can individually tune a board to mask the edge case and perform without fail, we are working on a global fix that in addition to support for "Current Apple G3 upgrade cards, and XLR8 MACh Speed G3Z", will also allow the prototype Moto ZIFs to work in all cases and with interleaved memory.

Lastly, I would like to add that we do promote that for maximum performance in all cases, a person invoking MVP can get highest performance (as verified by Macbench) by de-interleaving memory, then increasing the bus speed (the motherboard bus speed can be pushed higher with de-interleaved memory than interleaved memory).

We will provide more information as information is gained.
Regards,
Gary Dailey

The gist of all this is that the current CarrierZIF design will require deinterleaving RAM to work with OEM ZIFs that are Motorola 'X' part number versions (XLR8, Apple ZIFs are not affected and neither are Powerlogix ZIFs I suspect -as most of them have IBM CPUs as shown in my reviews). Gary also sent a Motorola PDF file that contains a list of Motorola 'X' Part number info (called Pilot Production). The table lists these as having "Modified Voltage and Temperature Specification". I have also been told by readers that Bottom Line has halted shipping of the card.

On the subject of 'X' Series Motorola parts - remember the Motorola XPC106 PCI/Memory controller chip that Apple used in the Beige G3? It too was not their standard MPCxxxx part. Well I just verified that my Blue and White G3 also has a XPC106 chip on it.

Bad News on B&W G3/WD Expert: I've just finished several tests on the data corruption issue with the Western Digital Expert 18GB Drive in the Blue and White G3 - I'm sad to say even disabling ATA66 Mode (via a PC) did not fix the problem. Here is what I did:

1. I performed the SMI file test as noted in yesterday's news (FWB comment) and the checksum failed when run from the Expert drive. (I verified that the SMI file was good by running it from the stock 6GB drive). Copies ran from the WD Expert drive failed a checksum test every time.
2. I then installed the TurboMax PCI IDE card in the B&W G3. I jumpered the WD Expert as Master (since it was the only drive on the cable) and had to reinitialize it with drive setup (this happens each time you switch a drive from the TurboMax to an onboard IDE controller in my experience).
3. I then copied the same SMI file to the Expert now running off the TurboMax card - it passed.
4. While I had the drive running on the TurboMax, I ran the ATTO Tools Benchmark to see how performance compared to the onboard IDE of the B&W G3. Read rates were about identical - slightly over 30MB/sec peaks, but write rates were a bit lower (27MB/sec vs just over 30MB/sec with the B&W G3 IDE). Not bad really, better than I expected.
5. Since the WD Expert's ATA66 mode is known to cause data corruption on many PC IDE controllers, I then moved the drive to a PC and ran WD's ATA66 utility - it showed ATA66 mode was enabled, so I disabled it.
6. I then moved the Expert back to the onboard IDE of the B&W G3. SMI files would still report an error (it seemed worse in fact - it reported a bad file even before the checksum was completed).

I was really hoping that disabling ATA66 would fix the issue as it does on PCs, but apparently there is something different about the B&W G3's IDE. This is very disappointing. B&W G3 owners can't take advantage of one of the fastest IDE drives on the market due to this problem. Using FWB's driver and modified DMA mode changes fixes the problem, but a reader said it severely impacted performance - 27MB/sec rates fell to 12MB/sec he said. I hope in the future Apple can address this problem with either another firmware update or revision to the disk driver (without the severe negative impact on performance seen with the FWB fix). I'm still wondering about the reader that reported the FWB comment who said his WD Expert did not have the problem. My B&W G3 has the 1.02 Firmware update and is running OS 8.6.

I added a warning on this issue earlier today to the reader review of the WD Expert, my B&W Info, B&W G3 Compatibility page and my B&W G3 Slave drive article. I'll be updating them with my findings as soon as I can.

In case any readers were wondering about the Maxtor DiamondMax Plus drive I used in my original B&W G3 Slave drive article, - it works perfectly. There were no errors on the SMI file checksum test on both partitions (verified again tonight).

Symantec has released v4.04 of Norton Utilties:

" What's new in 4.0.4?
Resolves an issue with some 333Mhz iMac's where FileSaver crashes when attempting to create its FileSaver data files, or on restart after initial installation of Norton Utilities.

Resolves an incompatibility between Norton FileSaver 4.0.3 or earlier and Anubis v3.x or APS Power Tools v4.x drive formatting software. "

PCWorld has a review of removable media drives including an IDE version of the Orb 2.2GB drive, which was the fastest drive. The bottom of the pack was the Superdisk. I've had one report of an IDE Orb being used in a B&W G3, but the owner reported disk changes were not recognized until a reboot.

Although it's a moot point now after proving the drive suffers from data corruption errors when running from the onbaord IDE of the B&W G3, I tested the Western Digital Expert 18GB drive in my 400MHz B&W G3 as a slave drive to the stock 6GB Quantum. At first boot the drive did not mount (it was formatted prev. under OS 8.1), but Apple drive setup recognized the drive and I simply reinitialized it. The performance in the B&W G3 was the best I've seen for any IDE drive, with peak rates exceeding 30MB/Sec in the Atto Tools tests. Sustained rates were similar to the 9600/ProMax tests in the review. I assume the faster memory bus (and possibly slower Mach 5 Kansas motherboard memory speed issue) are reasons why. Of course a 400MHz G3 with 1MB of backside cache helps too. As noted in the update above, the TurboMax PCI IDE card did not have the SMI checksum error seen with the B&W G3 onboard IDE and is about as fast. It's just a shame you have to shell out another $129 to run the drive reliably.

Although a reader yesterday said his WD Expert did not have the problem, today another B&W G3 owner wrote that all 3 of his WD Expert 18GB Drives installed exhibit the .smi image data error noted in yesterday's news link to FWB's B&W G3 Compatibility page. Even the drive connected to the CDROM channel has the same issue he said. The FWB 'fix' (changing DMA mode) dropped his peak rates from 28MB/sec to 12MB/sec he said.

I've seen two cases of HTML file corruption under OS 8.6 when writing to a PC format ZIP disk - the header (first part) of the file is garbage. I also noted *extremely* slow Finder performance when opening a PC formatted ZIP disk that contained about 50 or so files and folders in the root level. In Photoshop 5 this almost gave the appearance of a locked system. The data corruption happened on both the PB G3 and B&W G3 (the only systems I have running OS 8.6). I never saw this problem with OS 8.5 or earlier.

Larry of Other World Computing sent a note he's offering $10 off on the $299 366/183/1MB ZIF Upgrade. (To see owner comments on this upgrade select Other World as the brand name and search the Rate Your G3 Upgrade database):

"For your readers only!
Receive $10 off the Z366APL of Z366APLF when ordered online. MUST PUT XLR8YOURMAC in the comments field to receive the discount. This discount will appear on final invoice copy delivered with the upgrade.
Offer valid until 5/31/99.
Best Regards,
Larry O'Connor
Other World Computing, Inc."

I know many of you have already bought the upgrade, most report running 400/200 in the beige G3. Best ZIF bargain on the market right now in my opinion. It should also work fine in the XLR8 CarrierZIF card as well. Also check with OWC on trade-in credit for your original ZIF ($15 for 233, $25 for 266, $50 for 300MHz listed this morning).

A reader comments on reviews of G3 cards in the latest MacWorld:

"Interesting note--have you received your copy of MacWorld July issue? Have a good laugh when you get it or read it, if you get the print version.

They rate XLR8's 400MAChSpeed G3 at less than 1280CPU...it's the lowest rated card, and they claim it would not boot a 9500/132 at 50 Bus setting. As incredible as their "review" was I referred them to your pages(with URL's) with a sharply worded email about how MacBench 5.0 tests are traditionally performed. Their writer didn't know **** about testing G3 cards.

I have never read anything so poorly written or documented in my life, and I've seen a lot of "reviews" from Toasters to Mercedes Benz.

Even in my UMAX S900 without turning off the MB cache I got 1404CPU/1517FPU @428/214/1MB(2:1) last night. Newer RAM also replaced my 128 modules so all is well. ...
latergater,
David"

I'm often puzzled by MW's CPU reviews, as are many readers. Perhaps they had a 'bad' card or there was some other issue with their test system. As noted in my review the XLR8 400/200/1MB CPU card was the most stable CPU card I've tested in my 9500 based Genesis, 9600/350, etc. No other CPU card I've tested would boot at 50Mhz bus in the Genesis or 9600/350 - but the XLR8 400 did from day one. And as regular readers know, I've had more reader reports than I can count that also had great results with it. Many of these are noted in the Rate Your G3 Upgrade Database.

Robert Levy wrote with a note that reinstalling Speed Doubler under OS 8.6 solved the commonly reported problem with shutdowns:

"i had the 'application not quit' message with 8.6 and speed doubler. i tried all the solutions that i read on the net. increasing memory, changing faster file copying, i even disabled it. nothing worked. then i read that speed doubler needs to be removed or uninstalled by the installer to really get rid of it cause it does something to the system resources. after uninstalling it and reinstalling it - the restarts now work and the application quit message is gone.
levytv"

CPU Rumors: Here's the latest from the rumor mill:

• One comment that Motorola may not make G3 CPUs over 400MHz, focusing on the G4 I assume.
• The 10X Bus/CPU ratio G3 CPU may appear this summer. This would really help older macs with lower bus speeds achieve higher CPU speeds (8X is the current max Bus/CPU ratio).
• The G4 is undergoing a die/mask change which could delay G4 based systems until the first of next year.

These were all sent from my Anonymous Mail Form - which leaves no trace as to origin, but based on comments made, this comes from knowledgeable sources. I'd don't post completely off the wall comments. The 10X G3 rumor has been noted here in the past, let's hope it is true.

Ian Skopan sent a note that Quicktime Beta24 is now available. He sent a long list of updated components (too long to list here.

Jeff Carvalho sent a note that PowerLogix's new G3 Cache Profiler has caching support for BootX settings and Linux PPC. He sent a copy of his post to the Linuxppc user list:

" I don't know if this has been posted yet, but for those using a PowerLogix G3 card in their machines, the new PowerLogix G3 Cache Control Panel now has caching support for BootX settings and LinuxPPC. Very cool!

Check it out at: www.powerlogix.com/support/g3cpman/advanced.html

This control panel/extension set should also work with non PowerLogix cards and a host of native machines from the iMAC,beige G3, and B&W G3 for better cache and bus speed control! Check it out...completely worth it for stability purposes of your machine and linuxppc
jeff carvalho "

In the continuing thread on the Yamaha 4416S CDR - this reader reports that his drive has problems waking from sleep (disappears from the SCSI bus):

"Hi Mike,
I just want to ask if anybody has come across a problem with the 4416S not waking up from sleep mode. I am not sure that removable drives can go to sleep mode but mine seems to have trouble waking up. This problem is easily reproducible in my machine. I just let it go to sleep mode and when I wake it up the 4416S is no longer showing in my SCSI chain. Toast 5.5.6, HDT 3.0.2, and SCSI probe 5.1.1 does not see the drive. Restarting the computer does not solve the problem either. Only thru shutdown and power up does the drive become available. I have already updated the firmware on the 4416S to 1.0g. Help.
Erwin Sengco"

I've asked him for specifics on his system/scsi controller/OS version. I'm not running a 4416S on a Mac, but if you are and can test this please contact me.

Powerlogix has posted a warning on using generic PC100 DIMMs not made to Apple Specs in B&W G3s which they claim can cause problems after upgrading the CPU. I've never heard of this before and since the memory bus is fixed at 100Mhz (unless you change the Jumper Settings), I'm not sure why ZIF upgrade would suddenly cause problems. Perhaps some PC100 DIMMs are marginal quality. I've had no problems personally, nor has anyone reported it (even those using PC100 generic DIMMs). As noted on my original B&W G3 info page, the B&W G3's memory controller (according to apple's docs) reads the DIMM's SPD (timing parameters) chip and sets the memory bus timing to that of the slowest installed DIMM in cases where they are mixed. I wish Powerlogix had been more specific on what the actual problem was, or identified specific brands/chip types that were tested to be problematic. I welcome reports from 3rd party ZIF upgraded B&W G3 owners that have seen this problem - but as noted in my Rate Your G3 Database - most ZIF upgraded B&W G3 owners don't seem to have the problem.

A reader sent a copy of his driver update inquiry from VillageTronic which notes there may be updates soon:

">Hi Ingo, We will have a new glide release maybe today. And the first RAVE release is just being worked on. The first RAVE release should be ready within the next 2 weeks I think.
With kind regards
Tobias Seiler
Customer Support Service
Village Tronic "

If you don't check every day's daily news (updated several times a day normally) you might miss something valuable. Check the summary list below for highlights of the last two days news and the archives page for older new page links. For a blast from the past, read what was on the news page here exactly a year ago.

Other Net News:

• The Dirty Little Secret About Macs (theiMac.com)
• New Intel Price Cuts Threaten AMD (ZD/Net)

Thursday's News Summary

• Rage128's arrive at company (reader report)
• FWB B&W G3 Comment on data corruption w/some IDE drives
• Yamaha 4416S and Blue CDRs (Verbatum work/Memorex don't?) reader report
• Microconversions closes/iWonder opens
• Reader tip on using MotoDV card with Final Cut Pro
• HP announces first DVD+RW Drive
• Explanation of CDR disk colors
• Reader notes Toast update required for 4416S
• 3D Game News updated
• Still waiting on Powerlogix Z-Force
• Better Hardware & New Software... (Audio Column update)
• News.com article on iMac notebook (consumer 'book)
• TurboMax review updated for U2 Cheetah comparisons
• Starmax/Newer Tech speed bump reported
• Reader report on Phase5 G3/Umax C500 data corruption fix
• Beige G3/Newer Tech ZIF switch settings
• Readers raves about Falcon 4, bug report/patch
• Reader comment on PL G3 Cache Profiler 1.0
• Request for B&W G3/firewire compat. Videomail app
• Bottom Line weekly specials
• Reader comments on Omni Voodoo2 drivers
• Promax IDE/Softraid reader report
• Macsonly review of PF 466/233 ZIF
• Frequently Asked Questions Updated
• Rate Your G3 Upgrade database updated
• Other Net News

Wednesday's News Summary

• Microconversions closing? (reader report/order line message)
• Readers report Turbomax PCI card boots with OS 8.6 (contrary to earlier report)
• Apple releases iMac firmware update 1.2
• TurboMax IDE Controller review updated for U2/Cheetah comparisons
• Falcon 4 Mac in stores at CompUSA
• Newer Tech ZIF CPU Switch Settings (reader report)
• NetCD v1.6.6 released (works with Kenwood TrueX cdrom)
• Hardware accelerated Quake 2 for Mac OS X server
• Pentium III/Windows 98 clarification on PS5 Performance
• Tribes II first look (dual platform game)
• 3D Game News updated
• G3 CPU Upgrade Ratings database updated
• Other Net News

For other recent site news/features links, see the Sidebar to the right and the Recent Site Features page.

If you own a G3/300 or faster Mac, you might want to get our PPC Checker utility that will report if your CPU is copper or aluminum based.

Considering a G3 Upgrade? Check the CPU Upgrade Owner's Survey results and search for owner reports by Mac model and/or card brand. The database is being updated daily for new reports. Also make sure you ask your dealer about any issues with current builds of some models (ie PF G3) as I have gotten several reports of issues with certain mac models in the last few months (Mach 5, PowerBase, 7300). I urge any owners that have had problems to report this via the Survey Form. Always give the most weight to recent owner reports (dated in the listing).