I briefly used the Winchip 200 back in 1997/98. It was the cheapest upgrade for my Socket 5 motherboard.

I can't say I loved it. In fact in some areas (FPU) it was a non-upgrade. Particularly weak was the FPU. I remember Winchip 200 having similar framerate in Quake as the Pentium 90, so it was one of upgrades where it occasionally paid off to keep the original CPU 😀

I eventually upgraded to Celeron 300 and never looked back. Never tempted to experiment with Winchip 2/3. Who would be when Celeron was selling for a similar price? (yes, you had to upgrade the motherboard, but it was vastly superior). There is no surprise Winchip was a commercial failure. I think chips from Cyrix were a slightly better deal at the absolute low end if you had a motherboard that could run them. That was actually the only benefit of IDT C6. Unlike K6/6x86 it ran on any motherboard where it physically fit, incl. oldest Socket 5 boards.

Sorry for my non appreciation. I only appreciate it now as an obscurity 😀

Thanks for the reply! It is always interesting to read peoples experiences.

I have read that the WinChip1 FPU was particularly bad and I wouldn't be surprised if a Pentium 90 was similar or even better than a 200!

The WinChip2 at least in my testing seems significantly faster than the WinChip 1 (although I haven't tested it yet). My W2A chip results are here: https://youtu.be/hM3eo1d9-38 but I will also post raw data in the next few days.

I think the Cyrix 6x86 had a better chance as it was around earlier while really the WinChip was already "old" when out.

It is interesting though that you did see it as an upgrade from your Pentium 90, which gives me some good ideas for future benchmarking!

Only ever came across 2 WinChips a 486 and Pentium class cpus.

I had Winchip 200 in 2003 https://www.wimsbios.com/forum/topic3540.html...
It was slower in benchmarks than Pentium 133 it replaced 😀 BTW. I still have this patched BIOS from J. Steunebrink.

Wow very nice. Did you notice it faster in anything you did or was the P133 just all around faster? Did you go back to the P133?

Please add DOS BENCHMARKS!

Kotu Koylu wrote on 2021-05-10, 09:27:

Please add DOS BENCHMARKS!

It is interesting, I see so many videos with DOS benchmarks I wanted to try something a bit different. Maybe the WinChip is more a DosChip? 😉

I'll try and do some in future!

mwdmeyer wrote on 2021-05-10, 06:26:

I've just started benchmarking some WinChips and are interested to know if anyone had one back in the day and what they thought about them?

I have a range of WinChip 2 (pretty fast!) results that I will post once I have together. Some very interesting things really, it's not a bad chip at all.

I think a lot depends on whether you compare it to clock-for-clock equivalents, or to its period-correct competitors, as it was very late to market for an So7 CPU.

Certainly for ALU-intensive stuff it performed decently, clock-for-clock, and around 1998 FPUs still weren't being utilized a lot for the desktop applications this CPU was targeted at. But if you look at performance vs other things being sold at the same time (K6-2, Celeron), particularly in anything touching FPU, it was very low.

I'm actually looking for a cheap Winchip 1 or 2 to see if I can get one working on a very eccentric So5 board that refuses to boot with any AMD or Cyrix CPUs I've thrown at it so far. The biggest redeeming feature of the Winchip was that it ran at 3.3-3.5v, so could be used as an upgrade for boards without voltage regulators. In that sense the real comparison should be with other CPUs that run on 3.3V. That also explains why the 2A was a complete commercial disaster: it needed VRM anyway, which meant it had to compete with K6-2, M2 and P55C, which it clearly couldn't.

Yes I have benchmarks from Pentium 133 to K6-3, but yes when discussing it I am looking at similar MHz.

I do compeletely agree that there were a lot faster options when it was out.

The chip I have is a single voltage but a 2A so seems pretty good: https://www.cpu-world.com/CPUs/Winchip2/IDT-W … 3DFF200GTA.html

Raw data (I'm still adding to it): https://www.vogonswiki.com/index.php/Super_So … et_7_Benchmarks

mwdmeyer wrote on 2021-05-10, 09:04:

Wow very nice. Did you notice it faster in anything you did or was the P133 just all around faster? Did you go back to the P133?

Although the raw numbers spoke for the Pentium, the overall smoothness in Windows 98 was better on Winchip.
So Winchip stayed on this motherboard forever 😀

mwdmeyer wrote on 2021-05-10, 09:28:

Kotu Koylu wrote on 2021-05-10, 09:27:

Please add DOS BENCHMARKS!

It is interesting, I see so many videos with DOS benchmarks I wanted to try something a bit different. Maybe the WinChip is more a DosChip? 😉

I'll try and do some in future!

I watched ur latest videos. I felt something is missing 🙂 please add dos benchmarks to ur chart

I didn't had them back in time, but i'm collecting them now.
Winchip 1 (c6) was roughly a 486 on steroids, one of the slowest cpu for socket 7
Winchip 2 had performance similar to the k6-2 and it is the only non amd cpu with 3dnow support. It used the voltage of 3.52 to be compatible with older socket 7 motherboards
Winchip 2a, still looking for it
Winchip 2b I will test it as soon I will be able to repair my aopen motherboard.

I suppose the performance between 2, 2a, 2b will be the same but we will see I suppose.

The most interesting aspect to me is the 3DNow support of WinChip 2. I've enjoyed trying various 3DNow-supporting games with the WinChip2 240 that I have. That AMD-written Quake2 3DNow patch works well enough.

The only WinChip in my collection is a 200MHz C6. I bought it back in 1999 for the sum of £5 "new in box". A very good price for an older CPU at the time.

Originally it was an "upgrade" for a 100MHz Pentium, but it was really more of a downgrade and it got replaced after some months with a Cyrix MII. Ultimately the C6 ended up replacing a Pentium 75MHz in a Socket 5 PC, where it still resides to this day and works very well.

The C6 isn't a great CPU, but I do love it's oddness, at least on the 1 I have - glued heatsink/fan powered straight from the socket.

I've done some testing on the C6 recently and actually not as bad as I thought. Slower than the Pentium 133 in most areas but I would think similar to a Pentium 90-100.

I've added my C6 results here (sorry getting a bit hard to read!)
https://www.vogonswiki.com/index.php/Super_So … et_7_Benchmarks

The WinChip C6 looks faster than the Cyrix 6x86L PR200 (2x75MHz) in both int and fpu.

mwdmeyer wrote on 2021-05-12, 09:52:

The WinChip C6 looks faster than the Cyrix 6x86L PR200 (2x75MHz) in both int and fpu.

It's interesting you say that, I remember around 2001 I got hold of a Cyrix 6x86L PR200+GP CPU as part of some scrap components from somebody who upgraded their PC. I tried it for a while as a replacement for the C6, but ultimately ended up putting the C6 back in as I wasn't happy with it's overall performance, plus some software compatibility issues that the C6 didn't have.

Yeah the 6x86 (non-MX) CPU is pretty awful IMHO. Actually I feel one of the worst for socket 7. The MX version is a lot better.

Lots of compatabilty problems, no MMX and smaller issues like needing 75MHz/2.9v volts etc.

I was given a 6x86L PR200+ to upgrade my Pentium 133 back in the day and swapped back pretty quickly.

mwdmeyer wrote on 2021-05-12, 09:52:

The WinChip C6 looks faster than the Cyrix 6x86L PR200 (2x75MHz) in both int and fpu.

Interesting. That's not how I remember it.

In fact there are still contemporary reviews which show C6 to be lagging behind PR200 or even PR166 cpus:

https://www.tomshardware.com/uk/reviews/idt-w … -c6-cpu,37.html

That's likely with the original 6x86 though (running at 166 MHz with 66MHz bus). 6x86L (150/75) might work differently. Maybe that's why?

Wow. I still remember reading those reviews when they were fresh 😀