elmatero wrote on 2026-04-25, 05:29:

There was also ultra rare 83x2 PR-233 produced by IBM .

https://www.cpushack.com/CyrixID.html

If we want to include early/very late/rare it’s important to note that many Cyrix chips with a specific physical speed started out one speed rating then moved up (or down) to another later. (See incomplete chart)
There is Some evidence that Cyrix overseas did some tall tales selling things with different markings compared to domestic
[as 95x2.5 (black top) was noted some places as existing as pr333] which as far as I know never existed in NA.

Good examples of the same chip being marketed as different speeds

PR266/PR300

PR333/PR350 ~366(at least it had a different FSB)

The historical PR400 had a different clock speed but I don’t think it was actually released when it was mentioned historically (it showed up early in the computer shopper but nobody had inventory )

Pr350’s of any variety are rather rare and worth documenting

Back when I was on an old bravenet forum during the last 3 months National was in the Cyrix business insides (folks that bought a lot of Cyrix chips for pc business) were claiming they were selling all sorts of strange things, different colors, different labeling /typesets, wierd voltages and wierd speeds.

Now 26 years later I haven’t seen photo evidence of most of these, just verbal mentions here and there.

Makes me disappointed nobody from vcfed interviewed the Cyrix folk before they passed

Yeah I have an IBM 6x86MX PR233 3x66MHz FSB that matches the MHz of the Cyrix MII PR266 - 200MHz

Hello. This is also the CPU used in my first personal computer(Cyrix 6x86 PR120).
I've always wanted to collect its best CPU.

weilun777 wrote on 2026-04-27, 07:01:

Hello. This is also the CPU used in my first personal computer(Cyrix 6x86 PR120).
I've always wanted to collect its best CPU.

If the motherboard was fixed voltage, single plane, that might have been the PR 166+, if the PR120 was an L version and there was split plane, possibly PR-200+
but usually split plane would allow MMX/MII support, then you'd be hunting unicorns with us in the "do the 400 or 433 functionally exist?" realms. But the motherboard could have been limited by bus speed to maximum 233Mhz actual clock speed too, or by the strength of the voltage regulator to under 15W or similar.

I bought a Cyrix 333GP 2.9v and did some quick testing to find settings that work:

2.0v 225MHz, 2.2v 237MHz, 2.4v 250MHz, 2.9v 285MHz. It would post at 3.2v 300MHz but it's definitely unstable. It works fine at lower voltage/speed, so I guess it's one of the 250nm chips despite the 2.9v label. As others have posted earlier, it probably takes one of the "2.2v" chips to get 300MHz and up.

Also, it was difficult to get the web site to work and post this. Something about dating robots? This may be one of my last posts if it's going to be this much of a hassle.

I'd like to know which Cyrix qualifying tests passed with the PR433, but not with the PR400. From my past testing, both appear to run fine at 3x100 with 2.2 V.

How far has one of these been overclocked, yet was still able to run a full suite of Windows games and software? In my tests, the PR400 ran fine up to 333 MHz. Almost everything ran OK at 350 MHz, except for the Final Reality benchmark. I didn't try overclocking my PR433 chip.

nickles rust wrote on 2026-04-27, 21:07:

I bought a Cyrix 333GP 2.9v and did some quick testing to find settings that work:

2.0v 225MHz, 2.2v 237MHz, 2.4v 250MHz, 2.9v 285MHz. It would post at 3.2v 300MHz but it's definitely unstable. It works fine at lower voltage/speed, so I guess it's one of the 250nm chips despite the 2.9v label. As others have posted earlier, it probably takes one of the "2.2v" chips to get 300MHz and up.

Also, it was difficult to get the web site to work and post this. Something about dating robots? This may be one of my last posts if it's going to be this much of a hassle.

Website works 100% fine for me.

feipoa wrote on 2026-04-28, 03:15:

I'd like to know which Cyrix qualifying tests passed with the PR433, but not with the PR400. From my past testing, both appear to run fine at 3x100 with 2.2 V.

How far has one of these been overclocked, yet was still able to run a full suite of Windows games and software? In my tests, the PR400 ran fine up to 333 MHz. Almost everything ran OK at 350 MHz, except for the Final Reality benchmark. I didn't try overclocking my PR433 chip.

Unofficially my VIA pr433 2.2 volt easily went to 400mhz, my board didn’t go any further.

It appears I sold off the chip when I downsized circa 2010 (ish)

I was referring to the Cyrix MII-PR433, not the VIA C3 series.

feipoa wrote on 2026-04-28, 13:45:

I was referring to the Cyrix MII-PR433, not the VIA C3 series.

Mine was a “Cyrix” MIIv-pr433, date code was well after national was dead so made by via .

I don’t remember even having to fudge voltage , used it with a SIS 540 motherboard

nickles rust wrote on 2026-04-27, 21:07:

I bought a Cyrix 333GP 2.9v and did some quick testing to find settings that work:

2.0v 225MHz, 2.2v 237MHz, 2.4v 250MHz, 2.9v 285MHz. It would post at 3.2v 300MHz but it's definitely unstable. It works fine at lower voltage/speed, so I guess it's one of the 250nm chips despite the 2.9v label. As others have posted earlier, it probably takes one of the "2.2v" chips to get 300MHz and up.

Also, it was difficult to get the web site to work and post this. Something about dating robots? This may be one of my last posts if it's going to be this much of a hassle.

I would imagine that someone on your ISP has been a naughty boy. If you don't have a static IP, then turning your cable/fibre/DSL modem off overnight one night might get you a "cleaner" IP in the morning.

rmay635703 wrote on 2026-04-28, 13:53:

feipoa wrote on 2026-04-28, 13:45:

I was referring to the Cyrix MII-PR433, not the VIA C3 series.

Mine was a “Cyrix” MIIv-pr433, date code was well after national was dead so made by via .

I don’t remember even having to fudge voltage , used it with a SIS 540 motherboard

Some sources have Via cranking those out well into the noughts, so I don't know where they all went. Developing markets most likely.

rmay635703 wrote on 2026-04-28, 13:33:

my... pr433 2.2 volt easily went to 400mhz, my board didn’t go any further.

Which Windows 9x benchmarks, games, or other software did you run at 400 MHz? From my experience, running the chip with difficult software at 350+ MHz caused crashes. I think if these actually ran well at 400 MHz, they would have marketed them as such.

If someone has a Cyrix machine running, I've love if you could try my benchmark tool

https://starrts.vogonswiki.com/downloads/stwin072.exe Windows 9x
https://starrts.vogonswiki.com/downloads/stdos072.exe DOS

Designed to target 2D RTS performance. I really want to see how Cyrix compares to Intel for this benchmark.

mwdmeyer wrote on 2026-04-28, 23:45:

If someone has a Cyrix machine running, I've love if you could try my benchmark tool […]
Show full quote

If someone has a Cyrix machine running, I've love if you could try my benchmark tool

https://starrts.vogonswiki.com/downloads/stwin072.exe Windows 9x
https://starrts.vogonswiki.com/downloads/stdos072.exe DOS

Designed to target 2D RTS performance. I really want to see how Cyrix compares to Intel for this benchmark.

I have a few Cyrix based systems setup, but put away. Do you have a table of results we can see for StarRTS? This will let me know what's lacking. I haven't been keeping track of all the new benchmarks being pushed on VOGONS.

Yep have online database https://starrts.vogonswiki.com/