Dumb logic board question

[yntzl]

Hello everyone,

I have a Power Mac G3 B&W with a dead logic board (as described here http://68kmla.org/forums/viewtopic.php?f=13&t=21030) and I'll buy a new mobo, the question is, my G3 is Rev 1 and the motherboard that I wanna buy is Rev 2, everything will work fine when I swap the boards or It will be a compatibility issue?

Thanks.

 

[mcdermd]

Everything should be just fine. The rest of the box is simply peripherals. I think the only difference on the logic board is the one ATA chip.

 

[Schmoburger]

Correct... revised ATA controller is the only real difference. The 1st revision had issues with corrupting data when using HDD's larger than 10Gb on the primary IDE bus, and/or running a slave drive. Whether you will experience this issue with a rev.1, and how severe it will be varies from computer to computer and HDD to HDD. My 450 was purchased as a 300mhz Rev.1 initially and I ran a 40Gb Seagate drive and a 10Gb Seagate on the internal primary bus, and ran the corruption tester and it came up with zero errors... conversely the other one I have was originally also a 300 rev.1 and we put a 120Gb WD drive in it and it was writing errors here there and everywhere. I elected to buy a Rev.2 board for mine anyway however, just to be safe and make sure I'd never have problems, and put in a 450mhz CPU and a rev.2 graphics card, which is slightly faster than the rev.1 card and is identified by the presence of a small black finnned heatsink on the GPU which is absent on the slower card.

Also as an aside, the way to tell a Rev.2 logic board visually is by looking at the 'CMD' branded chip in the corner of the board near the bottom PCI slot. If it has the suffix '402' then it is a Rev.2 board, if not it is a Rev.1.

 

[Renegade]

The 1st revision had issues with corrupting data when using HDD's larger than 10Gb on the primary IDE bus, and/or running a slave drive.

IRC, corruption may occur even with a <10 GB single HDD, if that drive is "fast" enough.

I bet a 9GB Seagate Medalist (ST39140A) would get data corruption.

 

[Schmoburger]

Yep, this is correct... the generally accepted convention among G3 users/hackers/modders for years has been to the effect thast basically any drive that isnt an identical to what is instsalled out of the factory is likely to give issues when used on the factory internal bus.If it doesnt it's a bonus, but even then, just because one drive works fine, doesnt mean the next will.

There are a number of ways that people have used to get round it... the simplest is simply to run a factory drive as a standalone boot volume on the onboard bus, and install a Mac compatible PCI ATA card to run the larger drive. A variation on this is installing a PCI firewire card with an internal port and running the second hard drive through a Firewire/ATA bridge off the internal firewire port as a firewire device. The other option is to run a bootable SCSI card, however SCSI has severe limitations by todays standards, the biggest being now that SCSI doesnt support drives of capacity greater than 36Gb. Also, SCSI can be fickle.

Me however, I simply opted to replace the logic board (when they were still readily available on the secondhand market in the mid-2000's mind you), as it was a far simpler, cleaner,efficient and also cheaper way of going about it, and ensured continued piece of mind as I was able to run two brand new HDD's instead of relying upon an aging 6Gb Quantum drive to remain in a functional state.

 

[Renegade]

I've always wondered why there was no class action lauched against Apple for such a design flaw.

 

[uniserver]

probably because it was only something they noticed when someone upgraded the hd. after years.

 

[uniserver]

The Apple Power Macintosh G3/300 (Blue & White), based on the Yosemite architecture, features a 300 MHz PowerPC 750 (G3) processor with 512k of backside cache, 64 MB of RAM, a 6 GB Ultra ATA/33 hard drive, a 32X CD-ROM drive, and an ATI Rage 128 GL graphics card with 16 MB of SDRAM.

 

[Renegade]

I remember reading reports about data corruptions on XLR8YourMac by mid' 1999.

 

[uniserver]

its like anything

how many ford pintos blew up killing people before they decided,,, Um we gotta acutely do something about this.

maybe not enough people got rowdy enough, so they were like eh,,,

Or maybe if you bitched enough they would replace the LB for free.

 

[Schmoburger]

I can only imagine it was an issue Apple knew about before the Rev.1 was shipped, as the servers all shipped with SCSI drives. Like, it didnt take the end user long to work out there was a serious issue... as Renegade said, XLR8 documented it within a few months of it being released to the market, so surely the operational testing guys at Cupertino knew about it. Same deal with the yoyos... They were downright dangerous and it was worked out very early in the piece. They were introduced in 2001? and well... I was still shipped one as later as 2005, and my brother purchased one new in around 2007 from Apple?... They were rubbish, they were dangerous, the whole Apple community knew about it, yet as far as I am aware, a recall was never initiated. And it wasnt just a possibility of them shorting out, they were notorious for shorting out and bursting into flames... I have two friends from high school who had it happen to them and Ive seen the evidence to prove it.

But yeh, I guess 13 years after the fact it's an irrelevant point, but it still does bring one to ask some questions about Apple's ethics... the same way the Pinto, Explorer, BA Falcon/Territory. and Crown Victoria make one question the ethics of Ford Motor Company.