ychh0 wrote on 2024-10-09, 12:27:

According to below site, there are many variation of information on its surface.
https://www.cpushack.com/cyrix-486-cpus/

I worked on this list. I tested all the chips with S0R5 or S1R3 markings in that table.

ychh0 wrote on 2024-10-09, 23:11:

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3 […]
Show full quote

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3.4F) I guess it may be one of them (R50, R51, R52) is related to 4V setting.
And 5kohm trim pot is OK for this mod? For example below one.
https://www.amazon.com/Projects-Variable-Resi … r/dp/B08MDJ4JD7

It seems all of Cyrix 5x86 133MHz have voltage information on them and it is 3.6V or 3.7V. I guess this information is for other producers to set voltage. I suspect Cyrix 5x86 133MHz inherently has very narrow operation window of voltage therefore Cyrix might sold them to mainly CPU module producers who can set voltage. Cyrix might not be able to sell 133MHz version to commercial normal users if its voltage operation windows is too narrow. If this assumption is right, 3.6V-3.7V could be sweet spot for 133MHz operation according to experiments results of Cyrix.

feipoa wrote on 2024-10-09, 08:38:

I recall 3dmark99 and 99max having issues with Cyrix 5x86 and 6x86 CPUs. The results were abnormally low. I think mkarcher made a fix for this, somewhere in here: Re: 3dmark99 MegaThread
I haven't tested it yet. Nonetheless, 3dmark99 shouldn't hang on Cyrix 5x86 CPUs. The first thing you should do is increase your voltage by adding a trim pot in place of the resistor that corresponds to the 4 V setting on the M919.

I doubt that the trimmer you linked has more than one turn revolution. You are better off with a trimmer that has 10 or 20 turns so you can dial in the voltage a bit better. Something like a Bourns PV36 series, https://www.digikey.ca/en/products/detail/bou … 02C01B00/666515 I'm using a 5K trimmer, as shown:

You will remove the resistor at location R51 and solder on the trimmer. Use the same orientation and leads that I am using. For the unused trimmer lead, solder it to the central pin of the trimmer. Note that temporarily removing the JP4 header will make soldering a lot easier.

Thanks!

Let me buy similar one.
In summary, remove R51 and connect pin #1 (CCW) to input side of smd pad of R51 and pin #2 (WIPER) to output side of smd pad of R51. And pin #3 (CW) is connected to Pin#2 (output side of SMD pad of R51). Is this right?

feipoa wrote on 2024-10-10, 04:58:

I worked on this list. I tested all the chips with S0R5 or S1R3 markings in that table. […]
Show full quote

ychh0 wrote on 2024-10-09, 12:27:

According to below site, there are many variation of information on its surface.
https://www.cpushack.com/cyrix-486-cpus/

I worked on this list. I tested all the chips with S0R5 or S1R3 markings in that table.

ychh0 wrote on 2024-10-09, 23:11:

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3 […]
Show full quote

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3.4F) I guess it may be one of them (R50, R51, R52) is related to 4V setting.
And 5kohm trim pot is OK for this mod? For example below one.
https://www.amazon.com/Projects-Variable-Resi … r/dp/B08MDJ4JD7

It seems all of Cyrix 5x86 133MHz have voltage information on them and it is 3.6V or 3.7V. I guess this information is for other producers to set voltage. I suspect Cyrix 5x86 133MHz inherently has very narrow operation window of voltage therefore Cyrix might sold them to mainly CPU module producers who can set voltage. Cyrix might not be able to sell 133MHz version to commercial normal users if its voltage operation windows is too narrow. If this assumption is right, 3.6V-3.7V could be sweet spot for 133MHz operation according to experiments results of Cyrix.

feipoa wrote on 2024-10-09, 08:38:

I recall 3dmark99 and 99max having issues with Cyrix 5x86 and 6x86 CPUs. The results were abnormally low. I think mkarcher made a fix for this, somewhere in here: Re: 3dmark99 MegaThread
I haven't tested it yet. Nonetheless, 3dmark99 shouldn't hang on Cyrix 5x86 CPUs. The first thing you should do is increase your voltage by adding a trim pot in place of the resistor that corresponds to the 4 V setting on the M919.

I doubt that the trimmer you linked has more than one turn revolution. You are better off with a trimmer that has 10 or 20 turns so you can dial in the voltage a bit better. Something like a Bourns PV36 series, https://www.digikey.ca/en/products/detail/bou … 02C01B00/666515 I'm using a 5K trimmer, as shown:

The attachment M919_variable_VRM_1.JPG is no longer available

The attachment M919_variable_VRM_2.JPG is no longer available

You will remove the resistor at location R51 and solder on the trimmer. Use the same orientation and leads that I am using. For the unused trimmer lead, solder it to the central pin of the trimmer. Note that temporarily removing the JP4 header will make soldering a lot easier.

Yes, it's pretty simple. Best to use solder tips that are bent. Does your board have one or two BJTs?

I don’t understand what is BJT
I’m planning to install trimmer on other small board and using wire connect to smd pad. Then space problem will be solved. Let me try.

Thanks!

feipoa wrote on 2024-10-10, 05:37:

Yes, it's pretty simple. Best to use solder tips that are bent. Does your board have one or two BJTs?

The BJTs are the two large transistors in my photo with heatsinks on them.

Then mine has two, the same as yours.

feipoa wrote on 2024-10-10, 07:11:

The BJTs are the two large transistors in my photo with heatsinks on them.

feipoa wrote on 2024-10-10, 04:58:

I worked on this list. I tested all the chips with S0R5 or S1R3 markings in that table. […]
Show full quote

ychh0 wrote on 2024-10-09, 12:27:

According to below site, there are many variation of information on its surface.
https://www.cpushack.com/cyrix-486-cpus/

I worked on this list. I tested all the chips with S0R5 or S1R3 markings in that table.

ychh0 wrote on 2024-10-09, 23:11:

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3 […]
Show full quote

Could you let me know which resistor should be changed? There are several resistors near voltage setting jumpers.(Mine is M919 3.4F) I guess it may be one of them (R50, R51, R52) is related to 4V setting.
And 5kohm trim pot is OK for this mod? For example below one.
https://www.amazon.com/Projects-Variable-Resi … r/dp/B08MDJ4JD7

It seems all of Cyrix 5x86 133MHz have voltage information on them and it is 3.6V or 3.7V. I guess this information is for other producers to set voltage. I suspect Cyrix 5x86 133MHz inherently has very narrow operation window of voltage therefore Cyrix might sold them to mainly CPU module producers who can set voltage. Cyrix might not be able to sell 133MHz version to commercial normal users if its voltage operation windows is too narrow. If this assumption is right, 3.6V-3.7V could be sweet spot for 133MHz operation according to experiments results of Cyrix.

feipoa wrote on 2024-10-09, 08:38:

I recall 3dmark99 and 99max having issues with Cyrix 5x86 and 6x86 CPUs. The results were abnormally low. I think mkarcher made a fix for this, somewhere in here: Re: 3dmark99 MegaThread
I haven't tested it yet. Nonetheless, 3dmark99 shouldn't hang on Cyrix 5x86 CPUs. The first thing you should do is increase your voltage by adding a trim pot in place of the resistor that corresponds to the 4 V setting on the M919.

I doubt that the trimmer you linked has more than one turn revolution. You are better off with a trimmer that has 10 or 20 turns so you can dial in the voltage a bit better. Something like a Bourns PV36 series, https://www.digikey.ca/en/products/detail/bou … 02C01B00/666515 I'm using a 5K trimmer, as shown:

The attachment M919_variable_VRM_1.JPG is no longer available

The attachment M919_variable_VRM_2.JPG is no longer available

You will remove the resistor at location R51 and solder on the trimmer. Use the same orientation and leads that I am using. For the unused trimmer lead, solder it to the central pin of the trimmer. Note that temporarily removing the JP4 header will make soldering a lot easier.

Trimmer mod clearly improved 133MHz operation. Thanks again for your advices..

After setting voltage as 3.7V, GL-Quake only run test passed 24+ hours running. Now, I'm thinking GL-Quake only run test is not adequate to evaluate stability.
(Finally at around 27hours it froze. This is different from before. Before it quit to Windows with error after 1x hours running. It seems those errors were due to low voltage)
And also, I did quick 6hours test. At first ran 256kbps MP3 (using Winamp 2.05) for an hour and then for around 30 minutes ran several apps simultaneously with MP3 playing (such as Microsoft Front Page, Excel, Powerpoint, Word, Acrobat 5.0, another domestic word processor, CPU-Z Vantage with benchmark and so on). And then 3DMark99 Max test (Voodoo2), and GL-Quake around 4 hours. All test were done without problems.
I'm happy with the results, though it stopped 27hours of GL-Quake running, it is enough for my purpose.
( Cyrix Enhancements set up as /BTB_EN=off /LOOP_EN=on /LSSER=off /RTSK_EN=on /FP_FAST=on /MEM_BYP=on /DTE_EN=on /BWRT=on /LINBRST=on)

BTW, 3DMark99 Max is still unstable with Matrox G200. I believe it is from Matrox G200 drivers. With/Without Cyrix enhancements random freezing occur. Even operation at 100MHz (25MHz*4) still random freeizing occur. I couldn't have finished a single 3DMark99 Max test ever with G200. I guess it may be from Matrox driver compatibility with specific domestic (language) Windows. In fact, I couldn't install G200 Windows NT 4.0 driver in this system because it does not work with this language versoin Windows NT. (I'm using universal Windows NT video driver instead)

Glad to hear things improved with increased voltage.

What Matrox driver version are you using? I think I tested with version 4.33c in Windows 95. Using later versions caused some problems, but I forget which now. For NT4, I think I used 5.06 or 5.07.

Driver version is 4.33c with Windows9x and ME. For Windows NT I don’t remember exact version but tested with several versions including even lower than 5.0x. I tested with Matrox Millenium II also with several driver versions and results were the same and couldn’t use with Windows NT.

feipoa wrote on 2024-10-13, 09:54:

Glad to hear things improved with increased voltage.

What Matrox driver version are you using? I think I tested with version 4.33c in Windows 95. Using later versions caused some problems, but I forget which now. For NT4, I think I used 5.06 or 5.07.