I've heard of it. That's about it. I am pretty sure the Winchip 3 was only released on paper, because none of the CPU collectors seem to have one. Samples probably exist somewhere though.

Is that not the one that was released as the Cyrix III samuel core?

Nope, the Winchip 3 appears to have been a revision of the 2 with bigger L1 caches... the Winchip 4 was the basis of the C3, with longer pipelines than earlier models: https://www.ele.uva.es/~jesman/BigSeti/ftp/Ca … /MPR/121605.pdf

someone on cpuworld has some, if you google image search.

BloodyCactus wrote:

someone on cpuworld has some, if you google image search.

Would you like to supply the link? I can only find C6 and winchip2 on google image search and cpu-world.

http://www.cpu-world.com/forum/viewtopic.php?t=1180

Kamerat wrote:

http://www.cpu-world.com/forum/viewtopic.php?t=1180

Holly guacamole!

Now the only thing we need is for the owner to bench these along with his MII, Joshua and the VIA Cyrix III

Good luck with that. CPU-World members don't seem to care much about having working hardware.
But now that some Winchip3 samples existed, I guess I need to update my "want list".

The same individual has a Cyrix Joshua?

feipoa wrote:

Good luck with that. CPU-World members don't seem to care much about having working hardware.

I do. I only buy working chips (or at least ones for which I have a reasonable expectation to work). For me it's not about keeping them on a shelf in a display case, but using them. At least for x86 hardware that is.

feipoa wrote:

Good luck with that. CPU-World members don't seem to care much about having working hardware.
But now that some Winchip3 samples existed, I guess I need to update my "want list".

The same individual has a Cyrix Joshua?

I have no idea, but who knows?

And while I can see your point, I can see theirs as well. And without those collectors, there might not even have been a CPU-World for us to get very nice chips from 😀
From the view point of some of the "display collectors", perhaps they see people who actually use this hardware as people who end up killing all available hardware in a slow process because all of us make mistakes sooner or later. And there's many friendly folks amongst them.

To be a store-it kind of CPU collector, I would at least want to verify that the CPU is working in a motherboard before sticking it in a box or in a frame. I have one rare chip in my collection which does not work and just knowing that bothers me. I have this itchy feeling to get rid of it - like a rotten apple on my fruit tray.

The vast majority of my chips, including rare ones, came from ebay scrappers who didn't know what they have, although I did fill some gaps in my collection from collectors. I'm not saying there is anything wrote with these baseball card style collectors, it is just not my preference. I'm going to PM the guy with the Winchip3 to see if he is willing to let me benchmark it.

feipoa wrote:

To be a store-it kind of CPU collector, I would at least want to verify that the CPU is working in a motherboard before sticking it in a box or in a frame. I have one rare chip in my collection which does not work and just knowing that bothers me. I have this itchy feeling to get rid of it - like a rotten apple on my fruit tray.

The vast majority of my chips, including rare ones, came from ebay scrappers who didn't know what they have, although I did fill some gaps in my collection from collectors. I'm not saying there is anything wrote with these baseball card style collectors, it is just not my preference. I'm going to PM the guy with the Winchip3 to see if he is willing to let me benchmark it.

That is exactly how I feel. My dead Rise MP6 bothers me inordinately, at least half a dozen times I've put it in the recycle pile only to snatch it back at the last minute. It's now at the bottom of a box hidden from my sight.

I hear quite a bit about dead RiSE mP6 CPUs, a lot more than other CPUs. I wonder if they were particularly fragile electronically? Did you get yours from LZF? He had a tonne of dead mP6 chips. I had to pay double to buy a tested one. I now have a rare RiSE mP6 333 chip that I haven't gotten around to testing. I figure there is a 50% chance it doesn't work, but at least it will make for a good trade chip on CPU-World.

In another 10 years, perhaps your dead RiSE mP6 will be worth something as a trade chip as well.

Did anyone try heating dead rise mp6s? I believe there could've been some kind of solder ball detachment problems, like in some video cards. Heating doesn't fix the problem for good though, but would at least let the CPU run for some time.

feipoa wrote:

I hear quite a bit about dead RiSE mP6 CPUs, a lot more than other CPUs. I wonder if they were particularly fragile electronically? Did you get yours from LZF? He had a tonne of dead mP6 chips. I had to pay double to buy a tested one. I now have a rare RiSE mP6 333 chip that I haven't gotten around to testing. I figure there is a 50% chance it doesn't work, but at least it will make for a good trade chip on CPU-World.

In another 10 years, perhaps your dead RiSE mP6 will be worth something as a trade chip as well.

Could you perhaps tell me more about these dying mP6 chips? I got only 2 of the more common 266's but wouldn't want to kill the few I have due to something that I could've easily prevented 😊
Do these actually have selectable CPU multipliers or are they locked?

Winchip 4 was predecessor IDT's Samuel Core if i'm right ?

The Winchip 3 is only avaible as Engineering Sample and did not made it into market.
Also Winchip 4.

Tetrium, they are not multiplier locked. I ran mine at 2.0, 2.5, and 3.0x. I don't know what is killing them as they were DOA. What GPA says might be worth looking into. These chips are BGA on a PGA adapter PCB and this aspect distinguishes them from most other socket 7 CPUs.

The question I have is, what temperature to bake the chips at and for how long? Oven temp, which usually maxes out of 300 C, or wood stove temp, which can easily go over 400 C?

matze79, yes, that information seems to be consistent with what I read elsewhere. People still get the engineering samples though. Data for a Winchip3 belongs on a comprehensive benchmark database. Speaking of which, I still need to add the mP6 333 and winchip 2B data. I'm curious if the Winchip 2B has any advantage of the 2A per-clock.