Did Cyrix use PR ratings or is it AST who was playing tricks?

[Breadfan]

Ok, so a few computers from our pentagon field office came home last week, and I just got around to checking them out.

One is an Athlon classic 500mhz on a Soyo board which runs great (it better...heheh, I built it.) The other is some funky AST "P166". Figured it was a Pentium 166mhz, but realized it was a Cyrix cpu.

The thing came back to the office DOA. No boot, just fans. Started the normal troubleshooting routine, minimum ram, only the power switch adn speaker, no peripherals, etc etc.

Still nothing. So I figured I'd swap out the CPU, and put in a lowly 100mhz Pentium from the shelf. No go. But this is where I noticed some intersting things.

First off, the bottom of the computer had a sticker claiming 166mhz.

I take the CPU out, and it has a silver heatspreader on top. I figured this is jsut a normal heat spreader like a K62 or p4 even, but that was about when it just fell off. Literally...sorta blew off in the breeze, lol.

Anyway, underneath was not a bare core like I expected, but the typical Cyrix 6x86 graphic on another heatspreader just like the little tiny aluminum one.

Thats what sorta confused me. There was no exposed core and this heatspreader was the same size as the little cap. So I really didn't see a point of that little silver one...all it would do would limit thermal efficiency...it was adding no cooling power or core protection whatsoever.

Then I noticed on the Cyrix CPU it said "133mhz".

I was never a Cyrix fan, but if I remember I think they did use PR ratings...well sorta, they'd say it was a P166 b/c it was as fast as a pentium 166 (which if I aslo remember correctly, wasn't usually the case :) ).

Still, AST claimed 166mhz, which obviously isn't the case.

Made me wonder a few things...maybe AST said 166 and put the silver cover on just in case anyone had the cpu out. It was supposed to be glued on, so I doubt it would've come off as easy.

Or is this just AST misinterpreting Cyrix's pr ratings? I'd really like to get it running and see what the default clock on the CPU is. If it's running at 166mhz, then that would mean AST was overclocking from the factory.

Well that'd be neat to check, but I'm not sure if I'll get that board up. It seems pretty dead, but I've gotten deader things to work so I'll give it a shot later in the day :)

Would be an interesting and unethical practice if AST had oc'd the 133 cyrix to get a 166 rating, and possibly just as unethical to put a 166mhz sticker on the computer when its only running at 133...pr or not.

So, just to check, was Cyrix using the PR stuff all the way back on their 133 chips?

Mike

[moonunit]

Looks like they did use PR ratings.

http://www.sandpile.org/impl/pics/cyrix/m1/17_top_c.jpg (See this piccy:)
(Naughty linking I know, but it saves you sifting through)

http://www.sandpile.org/impl/m1.htm (and its parent page:)
(to keep webmasters happy)

That should clear things up a little.

All hail Google!

[RED Hot Machine]

AST are not the only ones, I know Packard Bell used Cyrix chips. We had one that was sold as a 300mhz machine when the chip was really only running at 233. (Its at 266 now, but thats a different thread :) )

Originally posted by moonunit
Looks like they did use PR ratings.

http://www.sandpile.org/impl/pics/cyrix/m1/17_top_c.jpg (See this piccy:)
(Naughty linking I know, but it saves you sifting through)

http://www.sandpile.org/impl/m1.htm (and its parent page:)
(to keep webmasters happy)

That should clear things up a little.

All hail Google!

You've got your links the wrong way round :)

They should be like this

See picky (http://www.sandpile.org/impl/pics/cyrix/m1/17_top_c.jpg)

and its parent page: (http://www.sandpile.org/impl/m1.htm)

[Breadfan]

yeah, that picture is the cpu...I checked it, scraped off some more of that stranger thermal goo/glue and found it does have the "PR166+" rating on it.

I'm gonna put up pictures of it when I get home, I snapped a few with my digital camera but don't have cables or a CF reader here to d/l 'em, so won't be able to until tonite.

But the little cap that was on it is such a joke, I know its there just to cover up the logo on the cpu. As you can see in the linked pic, the gold peice on the cpu with the logo on it is the orginial heat spreader...well there was another one on the cpu from this AST machine.

I think the main point is that AST didn't claim it to be a PR166 at 133mhz, but actually claim it to be 166mhz.

If I can get it to post I can see if the post screen lets you know its only a 133mhz cpu...

I know..."who cares if its 133 or 166 its SLOOOW and OLD" but I for one find it interesting to see what AST was trying to pull off, and was seeminly successful at.

Anyway, its fun to dissect older equipment, especially premade stuff like this. You always get to see soem interesting if not effective cooling designs...this AST has a bracket with an 80mm fan blowing right across the hsf unit (which has a dying 50mm fan).

Mike

[repo man11]

Not quite a year ago, I was using a computer that had a PR166 Cyrix. Yes, it ran at 133. I was disappointed that it wasn't much faster than the Evergreen upgrade CPU that I had in my 486. Of course,I had overclocked that CPU to 150! This is how it goes when you use other peoples cast off equipment. I fooled with overclocking the Cyrix, the Biostar board was capable of 200 MHz, but they had to be about the worlds worst overclocker.

[Breadfan]

Here's the pic of that fake heatspreader. Kinda funny...its VERY thin, so could offer no structural support....its barely thicker than aluminum foil...

Sure back at those speeds excellent thermal properties weren't as necessary as with today's cpu's, but comeon...

I can't see another reason for this other than to cover up the CPU.

The AST is dead. We had a busy day today, our asbestos lab is going through an accredidation check so we can stay accredited. I had to spend a while today explaining to the guy doing the check just why their database is secure...ugh...he was a know-it all even though he didn't know what he was talking about. Was going on something about our Win2000 network and I had to inform him that we are using Novell, lol...

Anyway, I didn't have the time to ressurrect the AST, and I probably won't bother. There's a 2gig hdd, a cdrom drive, a floppy, and an old AT power supply plus the CPU and EDO ram that could still find a home in some of our older systems around the office.

Mike

[Caffinehog]

Seems to me that it most probably was a re-marked chip fraudulently sold as a better one. Changing multipliers could be all that was needed. Putting that cap over the writing would keep people from seeing what speed the processor was legitimately meant to do.

[eobard]

6x86's are multiplyer locked (at 2.0) My 6x86 cpu is a "p150+" which means it runs at 120mhz but according to cyrix it's just as efficient as an intel Pentium at 150mhz. The 133mhz chip is rated the "p166+" and the 150mhz chip is rated the "p200+".

[Cowboy X]

Cyrix played the Pr game first and failed. They should rather have bragged about price and beating higher mhz rated cpus.

AMD is doing that too but they can more afford to do so as their chips offer the pr rating and more . Cyrix always had a poor fpu and that was their downfall ........ I had a cyrix pr233+ chip with mmx . It ran at 188 Mhz and performed 2d and buisness apps almost as well as a pentium 2 233mhz chip but it's fpu was only on par with a 166 mhz mmx intel pentium .............. the good thing was it cost less than most penitum 150 mmxs at the time. I was at school then and that is why I bought it. I never had any trouble from it and later sold the cpu , it was my first successfull overclock 'til I fried the cheap QDI board :)