VOGONS


First post, by GabrielKnight123

User metadata
Rank Oldbie
Rank
Oldbie

I've got a CPU speed problem the CPU is an Intel 200MHz MMX BP80503200 SL23S/2.8v and the motherboard is a DFI G586IPV Rev.C

I've set the jumpers on the motherboard from the manual and its getting a post CPU speed of 166MHz, the motherboard has the two power regulators next to the CPU with the extra jumpers for setting 2.8v (MMX) JP11, JP20 and JP21 compared to some of the versions of this motherboard without these extra jumpers and regulators

I've set JP6 for 66MHz CPU bus clock 1-2 off, 3-4 on.
I've set the multiplier to 3x for JP8 to 1-2 off, 3-4 on.
I've set the voltage to 2.8v (MMX) JP9 1-2 on, JP11 1-3, 2-4 on, JP20 3-4 on, JP21 1-2, 3-4 off.

I know the bus clock works on JP6 because if I set it to 50MHz clock it posts as 100MHz for 2x but 125MHz for 3x and 100MHz for 3.5x, if I set the clock to 66MHz and 2x it runs at 133MHz and if I set it to 66MHz and 3.5x its still at 133MHz so the 3x and 3.5x multiplier is off. I reseated the CPU and now thats all I can think of to try

It has an Award Modular Bios v4.51PG at post and in the bios is Rom PCI/ISA Bios (2A59GD4F) I think this is the latest bios all though I'll try a newer one if there is one.

The motherboard manual I used is here:
https://www.elhvb.com/mobokive/archive/Dfi/ma … ipv/5ipv423.pdf

Attachments

  • Motherboard.jpg
    Filename
    Motherboard.jpg
    File size
    1.64 MiB
    Views
    694 views
    File license
    Public domain
  • Jumper 2.jpg
    Filename
    Jumper 2.jpg
    File size
    1.84 MiB
    Views
    694 views
    File license
    Public domain
  • Jumper 1.jpg
    Filename
    Jumper 1.jpg
    File size
    1.72 MiB
    Views
    694 views
    File license
    Public domain

Reply 1 of 9, by Gmlb256

User metadata
Rank l33t
Rank
l33t

The CPU is actually running at the specified CPU speed (except perhaps when selecting the 3.5x multiplier), it is just that during POST it doesn't display the correct frequency which means that the motherboard may don't actually have the latest BIOS. This can be somewhat checked by looking the date at the bottom of the POST screen.

VIA C3 Nehemiah 1.2A @ 1.46 GHz | ASUS P2-99 | 256 MB PC133 SDRAM | GeForce3 Ti 200 64 MB | Voodoo2 12 MB | SBLive! | AWE64 | SBPro2 | GUS

Reply 2 of 9, by GabrielKnight123

User metadata
Rank Oldbie
Rank
Oldbie

During post at the top for the year under the bios version v4.51PG is 1984-96 and at the bottom of the screen is:

09/23/96-i430VX-W877-2A59GD4FC-00

I have tried a known good Pentium 200mhz mmx from another board and its the same, speedsys and other programs say its only running at 166mhz but at least speedsys reports it as a P554c

Could there be a different jumper setting than the manual for the multiplier? I'm still holding back on that though ide like to know about the bios first

Reply 3 of 9, by zwrr

User metadata
Rank Newbie
Rank
Newbie

You can try 66MHz*1.5 and see what happens。

Cyrix 486DLC-40, 386-VC-H, 16MB, GD5422, ES1868F


Intel 486DX4-100EW, VI15G, 16MB, WD90C33, ES1868F


AMD5x86-133, HIPPO-15, 32MB, S3 Vison 964, ES1868F


K6-3+ 500, T2P4, 128MB, Millennium II, Voodoo 2 12MB, SoundBlaster AWE32


Reply 4 of 9, by zyga64

User metadata
Rank Oldbie
Rank
Oldbie

There is a difference between P54C (non MMX) and P55C (MMX) in interpreting multiplier.
Please watch this great explanation from Scorp: https://youtu.be/J0NLGfocviU?t=282
But this can be remedied by soldering a resistor - details also in this video 😀

Last edited by zyga64 on 2023-11-21, 08:54. Edited 2 times in total.

1) VLSI SCAMP /286@20 /4MB /CL-GD5422 /CMI8330
2) i420EX /486DX33 /16MB /TGUI9440 /YMF718+GUS
3) i430FX /K6-2@400 /64MB /Rage Pro PCI /ES1370+ALS100
4) i440BX /P!!!750 /256MB /MX440 /SBLive!+Vibra16s
5) iB75 /3470s /4GB /HD7750 /HDA

Reply 5 of 9, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

Jumpers are directly connected to the CPU's BFx pins - it shouldn't be BIOS dependant, only if the correct speed is shown at POST or not. But if softwares show that it runs at 166, that's probably the case. I would probably try some other multiplier settings to see if they'd work. Until you wrote that you tried a known good 200 with same result, I was considering, that you might have a counterfeit 166 with locked multiplier, but that doesn't seem to be the case ...

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 6 of 9, by GabrielKnight123

User metadata
Rank Oldbie
Rank
Oldbie

Thanks for the video zyga64 I'll look into adding a resistor on the BF0 to see if it helps but I might need some help with which of the jumper legs to use on this motherboard I can find the BF0 pin on the underside of the CPU socket though, zwrr the setting for 1.5x uses the same jumper setting for 3.5x JP8 1-2 off, 3-4 off, it runs at 133MHz for both settings its a bit strange that there is one settings for two multipliers

Reply 7 of 9, by zyga64

User metadata
Rank Oldbie
Rank
Oldbie
GabrielKnight123 wrote on 2023-11-21, 11:45:

Thanks for the video zyga64 I'll look into adding a resistor on the BF0 to see if it helps but I might need some help with which of the jumper legs to use on this motherboard I can find the BF0 pin on the underside of the CPU socket though, zwrr the setting for 1.5x uses the same jumper setting for 3.5x JP8 1-2 off, 3-4 off, it runs at 133MHz for both settings its a bit strange that there is one settings for two multipliers

It's by design. Because processors which supports x1.5 doesn't support x3.5, and vice-versa.
Take a look here: https://www.pchardwarelinks.com/cpuspeed.htm

1) VLSI SCAMP /286@20 /4MB /CL-GD5422 /CMI8330
2) i420EX /486DX33 /16MB /TGUI9440 /YMF718+GUS
3) i430FX /K6-2@400 /64MB /Rage Pro PCI /ES1370+ALS100
4) i440BX /P!!!750 /256MB /MX440 /SBLive!+Vibra16s
5) iB75 /3470s /4GB /HD7750 /HDA

Reply 8 of 9, by Gmlb256

User metadata
Rank l33t
Rank
l33t
GabrielKnight123 wrote on 2023-11-21, 08:00:

09/23/96-i430VX-W877-2A59GD4FC-00

Definitely not the latest BIOS for that motherboard.

There are later ones which can be found here, yours has the COAST module so pick the one mentioning 07/31/98-i430VX-W877-2A59GD4FC-00.

VIA C3 Nehemiah 1.2A @ 1.46 GHz | ASUS P2-99 | 256 MB PC133 SDRAM | GeForce3 Ti 200 64 MB | Voodoo2 12 MB | SBLive! | AWE64 | SBPro2 | GUS

Reply 9 of 9, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

I have noticed that sometimes the multiplier settings are sometimes scrambled on locked p1 mmx cpus

There are two registers you must fiddle with

Bf0, and bf1

You should also make sure these pins are pulled high when pull low (jumper) is removed

You can remove cpu and measure these pins for 3.3v with mobo turned on and bf0,1 jumpers removed.

You can check them again with mobo off and jumpers installed for continuity to ground (black, negative)

Past that, if that checks good just fiddle with those settings until you get 200mhz as I said some of rhe locked cpus had their multis scrambled.

I have some same S-spec cpus that between three different specimens will give three different multipliers without changing the mobo 🤣

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)