If so would you be willing to run a few benchmarks and games like Quake and make a video to post to Youtube? I really want to see a 486 pushing 200mhz with my own two eyes. I've had a few of these chips over the years and never got lucky. It would be awesome if there was a build of CPU-Z that worked with win95/3.1 where you could post the verified numbers online.
wrote:CPU-Z that worked with win95/3.1
There is one for Windows 95, but I suspect Windows 3.1 support is unlikely, you could install WIN32s and try it I guess. If you do I would be interested to know if my SIV utility runs.
wrote:If so would you be willing to run a few benchmarks and games like Quake and make a video to post to Youtube? I really want to see a 486 pushing 200mhz with my own two eyes. I've had a few of these chips over the years and never got lucky. It would be awesome if there was a build of CPU-Z that worked with win95/3.1 where you could post the verified numbers online.
I have mine at 160mhz. ( Very Stable ).
That’s as high as I could get it to run.
But it probably has to do with having the right motherboard and bios.
I think your best bet will be with the “AMD 5x86-133-P75 ADW” CPU.
Maybe with a Gigabyte motherboard.
There are a few people claiming to have their X5-133 running at 200 MHz, but to be honest, I doubt it will be stable long enough to even fully boot into Windows 95. You can run a few benchmarks and brag about it, but seriously, overclocking a x86 CPU by 50% just isn't going to work.
I have a pile of 5x86s and none of them are stable 180MHz. 180 with working cache is rare enough. 200MHz with the cache on is a pipe dream. Too bad nobody can reverse engineer a 486 on a smaller process and stick it on a PGA adapter to make a super upgrade CPU.
"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium
wrote:I have a pile of 5x86s and none of them are stable 180MHz. 180 with working cache is rare enough. 200MHz with the cache on is a pipe dream. Too bad nobody can reverse engineer a 486 on a smaller process and stick it on a PGA adapter to make a super upgrade CPU.
How about tracking down someone from Cyrix to explain what it took to do the reverse engineering in the early 90s? I think that part can be done. From there it's making the higher frequency version manufacturable, which main barrier is going to be cost.
I always had an amusing thought about Powerleap's PL54C-MMX Socket 5 to Socket 4 convertor. If an Intel was able to take a Socket 4 Pentium and convert it to a Socket 3 Pentium Overdrive, and Powerleaps adapter takes a Socket 5/7 Pentium and converts it to a Socket 4 - isnt there just a little bit more work required to adapt Powerleap's PL54C-MMX' to Socket 3?
I figured it was never attempted because most Socket 3 systems did not have the needed multiplier/fsb speeds and the PL54C-MMX's Socket5->Socket4 compatibility was probably a 50% chance at best... but it sure would be interesting to see.
Check out DOSBox Distro:
https://sites.google.com/site/dosboxdistro/ [*]
a lightweight Linux distro (tinycore) which boots off a usb flash drive and goes straight to DOSBox.
Make your dos retrogaming experience portable!
overclocking a x86 CPU by 67%
133 -> 200 = 50%.
I really want to see a 486 pushing 200mhz
Easy, use liquid nitrogen or wait till winter.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
The Intel Pentium overdrive for socket3 was actually using a modified P54C with extra L1 cache...so it's not exactly a socket4 chip. However, early prototype versions of the POD did supposedly use a 5V core from the socket4 chips. In addition to the extra cache, I'm sure POD must also have some onboard circuitry to halve the databus. Probably possible with an adapter, but the PL 54C/MMX is never going to go in a socket3.
"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium
wrote:I have a pile of 5x86s and none of them are stable 180MHz. 180 with working cache is rare enough. 200MHz with the cache on is a pipe dream. Too bad nobody can reverse engineer a 486 on a smaller process and stick it on a PGA adapter to make a super upgrade CPU.
wrote:How about tracking down someone from Cyrix to explain what it took to do the reverse engineering in the early 90s? I think that part can be done. From there it's making the higher frequency version manufacturable, which main barrier is going to be cost.
What would those chips be good for? I mean you would be able to benchmark a very fast 486 but then what? Games that need faster CPUs can be run on a Pentium or better. Even later games will require additional instructions and at some point the CPU is so restricted by bus bandwidths that increasing clock speeds won't do much for speed.
If you need more power than a real Am5x86 can deliver it's probably best to just change platforms.
I had a 5x86-150mhz marked chip in my laptop for a time as an upgrade
Would the 150mhz marked 5x86 chips clock higher than the standard 133mhz chips?
wrote:What would those chips be good for? I mean you would be able to benchmark a very fast 486 but then what? Games that need faster […]
wrote:I have a pile of 5x86s and none of them are stable 180MHz. 180 with working cache is rare enough. 200MHz with the cache on is a pipe dream. Too bad nobody can reverse engineer a 486 on a smaller process and stick it on a PGA adapter to make a super upgrade CPU.
wrote:How about tracking down someone from Cyrix to explain what it took to do the reverse engineering in the early 90s? I think that part can be done. From there it's making the higher frequency version manufacturable, which main barrier is going to be cost.
What would those chips be good for? I mean you would be able to benchmark a very fast 486 but then what? Games that need faster CPUs can be run on a Pentium or better. Even later games will require additional instructions and at some point the CPU is so restricted by bus bandwidths that increasing clock speeds won't do much for speed.
If you need more power than a real Am5x86 can deliver it's probably best to just change platforms.
A 1ghz 486 chip using 20 year old fabs would be possible and cost a couple bucks a chip in volume
It would be a small surface mount and need voltage regulator
If some other market had use for a small x86 it could be possible.
Would likely need to integrate on chip dram and more advanced multiplier control to be of much use
Would be neat to see if modern 14nm process would be capable of making super high-speed low-voltage 486 CPUs.
Anyone here have some contacts at TSMC? 🤣
wrote:Would be neat to see if modern 14nm process would be capable of making super high-speed low-voltage 486 CPUs.
Anyone here have some contacts at TSMC? 🤣
Check out the Vortex86DX:
It is more of a i586 though.
Up to 1Ghz and uses a whopping 2.02w at 800Mhz:
http://www.dmp.com.tw/tech/vortex86dx/
https://en.wikipedia.org/wiki/Vortex86
wrote:There are a few people claiming to have their X5-133 running at 200 MHz, but to be honest, I doubt it will be stable long enough to even fully boot into Windows 95. You can run a few benchmarks and brag about it, but seriously, overclocking a x86 CPU by 67% just isn't going to work.
Well, the Celeron 300A often ran at 504MHz on air stable, thats an 68% overclock. I had the Celeron M 350 1,3GHz running at 2,6GHz on air cooling back in the days which is an 100% overclock. Overclocking a top of the line CPU by 67% is a much harder/impossible task.
A 1ghz 486 chip using 20 year old fabs would be possible and cost a couple bucks a chip in volume
That would still require major redesign. You can't just die shrink the original design indefinitely.
I must be some kind of standard: the anonymous gangbanger of the 21st century.
wrote:What would those chips be good for? I mean you would be able to benchmark a very fast 486 but then what? Games that need faster CPUs can be run on a Pentium or better. Even later games will require additional instructions and at some point the CPU is so restricted by bus bandwidths that increasing clock speeds won't do much for speed.
If you need more power than a real Am5x86 can deliver it's probably best to just change platforms.
Your question is like a mainframe designer asking what good is an 8080 microprocessor in 1975.
wrote:wrote:What would those chips be good for? I mean you would be able to benchmark a very fast 486 but then what? Games that need faster CPUs can be run on a Pentium or better. Even later games will require additional instructions and at some point the CPU is so restricted by bus bandwidths that increasing clock speeds won't do much for speed.
If you need more power than a real Am5x86 can deliver it's probably best to just change platforms.Your question is like a mainframe designer asking what good is an 8080 microprocessor in 1975.
That analogy makes no sense.
He is right. A super duper charged 486 would be fun, but it isn't in any way practical - even for the purposes of retro computing.
wrote:Check out the Vortex86DX: It is more of a i586 though. Up to 1Ghz and uses a whopping 2.02w at 800Mhz: […]
wrote:Would be neat to see if modern 14nm process would be capable of making super high-speed low-voltage 486 CPUs.
Anyone here have some contacts at TSMC? 🤣
Check out the Vortex86DX:
It is more of a i586 though.
Up to 1Ghz and uses a whopping 2.02w at 800Mhz:http://www.dmp.com.tw/tech/vortex86dx/
https://en.wikipedia.org/wiki/Vortex86
+1, I have one of these in an industrial SBC (half-ISA card w/ VGA, USB, serial, etc).
