Another AMD Cover Up?

[Paleo Pete]

I just stumbled on this article on Tom's Hardware Guide while trying to find some info. Never did find out the info I was looking for...

It's basically old news that the K6-2 at 350MHz or higher had a compatibility issue with Windows 95, due to a timing loop as I understand it, and AMD wouldn't admit it for a long time, then finally had to post a patch when it became pretty much common knowledge. Didn't encourage me to trust AMD too much...

Now AMD seems to be at it again. I'll have to read the article again to halfway understand it, but it seems the Irongate boards using GeForce cards that are supposed to support AGP 2x actually don't, they only supprot AGP 1x, due to some sort of bug, and AMD has been selling a fix, therefore taking people's money twice. And they are up to their old tricks of not admitting it, instead blaming it on nVidia's hardware...

This makes me wonder about some of the problems I've seen in different posts where I couldn't figure out what was happening, except that I noticed they were always AMD processors...

Maybe I'm being a bit harsh, but the article posted above makes me glad I've always been an Intel guy...

Let's not turn this into a flame fest, that's not the idea here, just letting folks know, and see what opinions we come up with...but feel free to correct any mistakes I made, I've just started to look into this, and don't have it all straight in the old bald noggin yet...

------------------
If you had everything...Where would you put it?

[url="http://www.geocities.com/paleopete/"]Computer Information Links (http://www5.tomshardware.com/blurb/00q1/000210/index.html)

[Samantha]

Hi Pete,

It's a little hard to follow on Tom's site how current this issue is. It seems to be a problem more with poor motherboard quality than with irongate. As the article notes, the two big OEM vendors are using irongate on custom motherboards. I don't know what AMD would have said about the issue. Something along the lines of, "some motherboard makers have screwed up"? However, if the problem is in AMD-certified motherboards, then I'd agree that they are being less than open. Nonetheless, I would sympathize, as they've had a devil of a time getting platform support.

------------------
S~~

[ixl]

Don't know about the current issue, but Pete, the 350 MHz thing is actually a Microsoft problem, not an AMD one. They wrote a timing loop into the OS, which is bad programming. It fails on AMDs above 350, and Intels above 700. Why the difference? Because there's an instruction used that takes half as many cycles on AMD chips as on Intel ones. I think MS has a KB article on it...
Interesting stuff!

------------------
Charles M. Kozierok
Webslave, The PC Guide (http://www.PCGuide.com)
Comprehensive PC Reference, Troubleshooting, Optimization and Buyer's Guides...
Note: Please reply to my forum postings here on the forums. Thanks.