Bruno128 wrote on 2024-08-01, 21:41:

It's nice to see you back
Testing version 0.3.0 with W83194BR-39B / SL-65FVB

Also tried other values nothing happens.

Thanks for testing!

That's very weird that it is reading the FSB all wrong. Are you sure you have the right PLL?

enaiel wrote on 2024-08-01, 22:56:

Thanks for testing!

That's very weird that it is reading the FSB all wrong. Are you sure you have the right PLL?

This one https://theretroweb.com/expansion-chips/4407

Bruno128 wrote on 2024-08-01, 23:03:

This one https://theretroweb.com/expansion-chips/4407

Ah, unfortunately you can't go by specifications. Often different revisions of the same motherboard use different PLL chips. You need to visually identify the PLL chip on your motherboard to be absolutely sure.

enaiel wrote on 2024-08-01, 23:40:

Bruno128 wrote on 2024-08-01, 23:03:

This one https://theretroweb.com/expansion-chips/4407

Ah, unfortunately you can't go by specifications. Often different revisions of the same motherboard use different PLL chips. You need to visually identify the PLL chip on your motherboard to be absolutely sure.

I’m saying it’s literally this one on the photo.

Bruno128 wrote on 2024-08-02, 07:04:

I’m saying it’s literally this one on the photo.

Oh, okay. In that case, could you run the dump command and send me the output? It will not actually change the frequency but just do a dry run.

VIAFSB W83194BR-39B 66.82/33.41 -d > dump.txt

enaiel wrote on 2024-08-02, 11:42:

VIAFSB W83194BR-39B 66.82/33.41 -d > dump.txt

1VIAFSB v0.3.0 - DOS FSB utility for VIA chipsets.
2VIAFSB: Trying to set FSB to 66.82/33.41 using PLL W83194BR-39B...
3VIAFSB: Found supported VIA Southbridge: VT82C596/A
4VIAFSB: Using SMBus Config Address: 0x90
5VIAFSB: Found SMBus Address: 0x5001
6VIAFSB: Using instead SMBus Address: 0x5000
7VIAFSB: SMBus is enabled
8VIAFSB: VIA Southbridge Revision ID: 0x00
9VIAFSB: Using PLL W83194BR-39B...
10VIAFSB: PLL W83194BR-39B is supported
11VIAFSB: Found PLL at SMBus Slave Address: 0x69
12W83194BR-39B: Read 13 bytes (hex): 00 CF FF FF 87 F3 00 B0 27 00 97 62 51 
13W83194BR-39B: FSB_BYTE(0) read (hex bin): 00 00000000
14W83194BR-39B: FS_SEL_BIT(1) is NOT set in FSB_BYTE(0). Getting FSB from Latches...
15W83194BR-39B: FS5 FS4 FS3 FS2 FS1 FS0 (bits): 0 0 0 0 0 0
16W83194BR-39B: Got key for (0 0 0 0 0 0) (hex bin): 00 00000
17VIAFSB: Got FSB from PLL W83194BR-39B: 80.00/40.00
18VIAFSB: Requested FSB 66.82/33.41 is supported by PLL W83194BR-39B (PCI divider 2)
19W83194BR-39B: Found key for FSB(66.82/33.41) (hex bin): 03 00011
20W83194BR-39B: FS5 FS4 FS3 FS2 FS1 FS0 (bits): 0 0 0 0 1 1
21W83194BR-39B: Writing FSB_BYTE(0) (hex bin): 0E 00001110
22W83194BR-39B: Writing 13 bytes (hex): 0E CF FF FF 87 F3 00 B0 27 00 97 62 51 
23VIAFSB: Unable to set FSB 66.82/33.41 using PLL W83194BR-39B

And I also would like to make a request for another PLL ready for testing, ICS 94228 datasheet

Testing version 0.3.0 with ICS950405AF / Asus K8V SE Deluxe
Dump looks like this, in reality it halts/freezes the system.

1VIAFSB v0.3.0 - DOS FSB utility for VIA chipsets.
2VIAFSB: Trying to set FSB to 180.00/0.00 using PLL ICS950405...
3VIAFSB: Found supported VIA Southbridge: VT8237/R
4VIAFSB: Using SMBus Config Address: 0xD0
5VIAFSB: Found SMBus Address: 0x0401
6VIAFSB: Using instead SMBus Address: 0x0400
7VIAFSB: SMBus is enabled
8VIAFSB: VIA Southbridge Revision ID: 0x00
9VIAFSB: Using PLL ICS950405...
10VIAFSB: PLL ICS950405 is supported
11VIAFSB: Found PLL at SMBus Slave Address: 0x69
12ICS950405: Read 24 bytes (hex): F7 FF FF F5 7F 7F 18 01 CC 77 80 E2 E7 06 18 00 F7 FF FF F5 7F 7F 18 01 
13ICS950405: FSB_BYTE(0) read (hex bin): F7 11110111
14ICS950405: FS_SEL_BIT(-1) is NOT set in FSB_BYTE(0). Getting FSB from Latches...
15ICS950405: FS5 FS4 FS3 FS2 FS1 FS0 (bits): 0 0 0 1 1 1
16ICS950405: Got key for (0 0 0 1 1 1) (hex bin): 07 00111
17VIAFSB: Got FSB from PLL ICS950405: 200.40/33.40
18VIAFSB: Requested FSB 180.00/30.00 is supported by PLL ICS950405 (PCI divider 6)
19ICS950405: Writing BYTE_COUNT_BYTE(6) (hex bin): 0F 00001111
20ICS950405: Writing 15 bytes (hex): F7 FF FF F5 7F 7F 0F 01 CC 77 80 E2 E7 06 18 
21ICS950405: Read 24 bytes (hex): F7 FF FF F5 7F 7F 18 01 CC 77 80 E2 E7 06 18 00 F7 FF FF F5 7F 7F 18 01 
22ICS950405: Found key for FSB(180.00/30.00) (hex bin): 09 01001
23ICS950405: FS5 FS4 FS3 FS2 FS1 FS0 (bits): 0 0 1 0 0 1
24ICS950405: Writing FSB_BYTE(0) (hex bin): F9 11111001
25ICS950405: Writing 15 bytes (hex): F9 FF FF F5 7F 7F 18 01 CC 77 80 E2 E7 06 18 
26VIAFSB: Unable to set FSB 180.00/30.00 using PLL ICS950405

This is the detection although the frequency options are a bit weird and none of them work

Works fluently with Asus A7V133-VM! ICS94215.

dj_pirtu wrote on 2024-10-24, 05:59:

Works fluently with Asus A7V133-VM! ICS94215.

I like "fluently" 😀 Great word for this fix!

Please add support for VIA VP3 chipset based motherboard SOYO-5ED5.
PLL is ICS9148BF-04
https://www.alldatasheet.com/datasheet-pdf/pd … S9148BF-04.html

Regarding above VIA VP3 motherboard, it's VT82C586B based.
I've implemented some code on my own by using PLL ICS9148BF-04 datasheet and app reported that SMBus is just disabled and it failed to enable it.
I confirm that doesn't work in such early VIA motherboard.

Amazing tool, could you try to add support for the Epox/ProNix EP-8KTA3+?

The southbridge is a VIA VT82C686B, and the PLL is this one: https://www.alldatasheet.com/html-pdf/179408/ … /HIP6301CB.html .

Mobile CPUs up to Athlon XP Barton are unofficially supported (only problem on my mobile Barton 2500+ is that the max FSB of the board is 133 and the max multiplier of the board is 12.5, so it won't go above 1.6 GHz, but at least it has twice the cache of previos generations. It also works with AMD Geode NX CPUs (highly binned Athlon XP Thoroughbred-B with unlocked multipliers, again up to 12.5), so in theory a wide range of speeds should be accessible.

Chaintech 7AJA2 has: Cypress W230H clock generator (PLL)
https://www.digikey.at/en/products/detail/roc … /W230H/12128234
https://rocelec.widen.net/view/pdf/rhrzt4ub8v … d=true&u=5oefqw

Would you please add support for Cypress W196?

Hopefully you are still taking requests. I've a Tyan S1854 (Apollo Pro 133A) 694X-596B and a Cypress W196 PLL. Here is a link with the data sheet: https://theretroweb.com/chips/5784

CPUFSB works, but the nag screen is annoying and there's no way I know of to make it go away. TIA

I'm back with a release after another 2 years...

Made some more long overdue updates to VIAFSB and published the code as before to Github: https://github.com/enaiel/viafsb.

v0.3.1
* Code refactoring and cleanup.
* Added print config feature to help verify against PLL datasheet.
* Fixes for PLLs ICS9248-127, W83194BR-39B, W83195R-08.

I tried everything I could to try and change FSB frequencies across PCI dividers, but failed. Some PLLs allow you to program them with fine grained control to set any frequency in the given range by setting a VCO divider (n) and REF divider (m) using a formula, instead of setting the frequency using values from a ROM table. Some PLLs allow you to program safe frequencies for a Watchdog timer that will send a reset pulse and switch the frequency if it detects instability after changing the FSB. However, no matter what I tried, all I got was hard freezes.

I added a print config feature (-c|--config) to print out the config tables of a PLL similar to a datasheet to help diagnose config issues. Using this feature I found an error in my ICS9248-127 config, which I fixed. I also found errors in the datasheets for W83194BR-39B and W83195R-08 which I also fixed. They neglected to mention that the latched frequencies were inverted! These 3 PLLs should hopefully work now.

I know it's been a long time coming, but I appreciate any testing you are able to do with this release.

Grem Five wrote on 2022-04-06, 02:08:

I have not had any luck with the winbond machines […]
Show full quote

I have not had any luck with the winbond machines

[*] DFI CA63-SN W83194BR-39B

<snip>

As you can see from pics viafsb isnt showing right FSB just loading the pll

[*] Asus P3V 4X W83195R-08

<snip>

I noticed on this one it says Detected VT82C596A southbridge when its actually a VT82C596B I dont know if that has anything to do with the error or not.

Hi @Grem Five, it took me 4 years, but I think I finally figured out why it wasn't showing the right FSB - the datasheets for the two PLLs neglected to mention that the latched frequencies were being returned inverted!

Bruno128 wrote on 2024-08-01, 21:41:

It's nice to see you back Testing version 0.3.0 with W83194BR-39B / SL-65FVB […]
Show full quote

It's nice to see you back
Testing version 0.3.0 with W83194BR-39B / SL-65FVB

<snip>

Also tried other values nothing happens.

This applies to you as well, @Bruno128.

mockingbird wrote on 2022-03-23, 18:02:

Awesome! […]
Show full quote

Awesome!

Can you please add support for the following:

PLL205-03XC (Abit KT7A)
ICS94215AF (Azza KT3-AV)

Thanks

mockingbird wrote on 2022-05-09, 03:12:

Can you please add support for ICS9248AF-127?

mockingbird wrote on 2022-05-18, 04:14:

Can you please add support for W230-03H (Biostar M7VKD)?

Hi @mockingbird, did it work for the PLLs you requested: PLL205-03, ICS94215, ICS9248-127, and W230-03H? I just fixed a misconfiguration on my side for PLL ICS9248-127, so it's worth a re-test.