Reply 41 of 58, by feipoa
Reply 42 of 58, by Unknown_K
Anybody here snag this?
Collector of old computers, hardware, and software
Reply 43 of 58, by darry
Unknown_K wrote on 2020-05-21, 19:23:
Anybody here snag this?
My whole vintage setup did not cost that much . My pockets aren't deep enough
Reply 44 of 58, by feipoa
Reply 45 of 58, by vmunix
I wonder if that part has the bug in the branch prediction unit.
Trailing edge computing.
Reply 46 of 58, by feipoa
All 133 MHz parts don't do branch prediction in Windows reliably; DOS is OK though. To get a 133 MHz part with branch prediction work in Windows, you will need to find an S1R3 100 MHz or 120 MHz part which can be overclocked to 133 MHz on a motherboard that supports the 66 MHz FSB.
Reply 47 of 58, by Socket3
Interesting thread. I have a Lucky Star 5VP3 socket 7 board that will run at 8, 20 and 40 MHz unofficially. I discovered 8MHz accidentally when trying to get a K6-2 to run at 6x66MHz. It's stable at all these speeds, but at 8MHz multiplier selection is disabled, the board will always force 1x regardless of multiplier jumper settings. I believe the 8MHz setting is a debug feature (please correct me if I'm wrong). Don't know about 20 and 40MHz.
Reply 48 of 58, by rmay635703
feipoa wrote on 2011-12-14, 16:57:
I really would be curious to know just how the MII-433 @ 300 MHz compares up to the Pentiums. One day I'll get around to the ultimate socket 7 comparison. My gut feeling is that it is somewhere between a P233MMX and a P266MMX, perhaps even a slow slot1 celeron.
My 2.2v Mii-433 overclocks to a real 400mhz at that speed it does move fairly well but never benched anything 3D to see.
I believe someone did compare an overclocked PR433 vrs a celery 300a and came up with a few interesting results .
If Cyrix had sold real 400mhz parts in 1998ish they might have survived
Reply 49 of 58, by feipoa
Not sure if you are aware, but that question is almost a decade old. This question has been answered in the Ultimate 686 Benchmark Comparison. While I have an MII-433GP, I don't think I'd want to run extensive benchmarks at 400 MHz on it. I ran the MII-400 up to 350 MHz and extracted results for 400 MHz. The response vs. freq. was linear.
Reply 50 of 58, by gdjacobs
feipoa wrote on 2020-05-21, 20:46:
Yes, someone here did. Don't ask me who. It is up to him/her to reveal him/herself. What matters to me is that a [properly] scanned manual is coming my way.
More power to 'em. At least users on this forum are unlikely to try packing them full of play-dough before learning the limits of LN2 cooling.
All hail the Great Capacitor Brand Finder
Reply 51 of 58, by feipoa
Reply 52 of 58, by mpe
Reply 53 of 58, by feipoa
There are some QFP IBM 5x86c chips that will suffice then. I had one doing pretty well in DOS at 150 MHz with just a heatsink/fan... until I overdid it and it stopped working. I bet some liquid nitrogen cooling, or even a reliable peltier, could have yielded Windows operation.
Reply 54 of 58, by lukas12p
So I have bought AMD X5 160 and this happened..
Reply 55 of 58, by rmay635703
Reply 56 of 58, by The Serpent Rider
Yes, obvious fake. Marking won't go so easy from a genuine CPU and there's slight difference in ceramic color.
people will go pretty far to fake
If it's relatively popular and sought after and can be easily faked - it would be faked without a doubt.
Get up, come on get down with the sickness
Open up your hate, and let it flow into me
Reply 57 of 58, by amadeus777999
Reply 58 of 58, by lukas12p