Hi ChrisK,

today I tried to run my system at 160 Mhz. I did your modifications and the system boots, but it is very unstable. At the beginning it did not even boot in to Win95. I found out, if I disable the L2 Cache, it will boot in to Windows. Have no idea how this is connected. But even once it booted in to Win95, it crashes way to often to call it a stable state. Booted in to DOS, it will not even finish the Quake time demo without crashing.
I did not mess with the ISA clock.
I added a jumper for the resistor as well, so I can switch easily back to 133 Mhz. If I want to enjoy a stable system 😉

Regarding HDD size, i use a Dynamic Drive Overlay. With my Bios version from 10/10/95 it has the 500 (something) MB limit. I am using a IDE-SD adapter with a 16GB SD card.

Regards,
Peter

Hi peda126,

at what voltage does your CPU run?
I have added a 3.3V converter since my system did not have one when I got it. It was 5V-CPUs only. I additionally added some jumpers to set the CPU voltage between 3.42 and 3.62V in some steps.
At the moment it is set to 3.56V and the system is running obviously fine in DOS. I have to admit that I do not have a Windows installation running since I am still looking for the newer BIOS from 21-OCT-96 (the old tulip website in the internet archive lists it but it wasn't archived so it can not be downloaded ;( .
I was hoping Paadam would have this installed and could make a backup of this because he said he had a larger harddisk running...

But for the overclocking I think the CPU voltage could be the problem.

You should also need no jumper over the removed resistor for the 40MHz FSB thing. The resistor essentially sets S1 of the clock gen to GND and by removing it and adding the jumper to 5V you can select between GND & 5V by just shorting or opening the jumper (clock gen has internal pull-downs on the S2/1/0 pins). S2 is fixed to 5V by a PCB trace and S0 is set LOW/HIGH via J23.

Regards,
Chris

Mine is from October 1996, I attached a screen.

But for some reason my board refuses to work at 33MHz bus all of the sudden. Post nicely at 25 Mhz (and is stuck at starting MS-DOS screen, also does not boot from CD even though it displays choice whether I want to boot from CD or Windows (it is on the CD, so it finds boot sector). Any ideas? I have not yet tried other CPU, will do that soon.

That's exactly the BIOS version I am looking for!
It would be great if you could make a backup of it if your system is running again.
My older BIOS shows a flashing option on the first boot screen, via F3 key I think, and maybe there's a backup option in it, too (never entered it yet). Don't know if that's still there in your newer version.

Did you ever mod anything in your system regarding overclocking, voltages, ...?
I remember someone wrote he'd damaged the onboard CMD640 IDE controller by some fault but I could not find it here in this thread.
I would guess it is related to the IDE interface. I don't really think it is CPU related. But if you can check with another CPU this possibility could be eliminated.
The CMD640 is clocked by the chipset with the PCI clock, so changing the FSB also changes the clock for the IDE controller.
Maybe you can also check with another harddrive, disconnect the CDROM, ...?
If that also fails an ISA IO-card with IDE could also be worth a try.

I hate it when one error follows the next one. Every time when you think that's it now I can have fun another part stops working or does strange things.
My original harddrive has apparently also gone RIP. It suddenly showed write errors when starting Windows after I tried to install/update the graphics drivers.
But in my main machine it seems to work ok. On the other hand a compact flash card works like a charm in the Tulip, so where's the problem? Weird...
Think I will do a low level scan of the drive when I have some time... may be it's really the drive. Or just the Windows installation gone bad. Who knows.

Hope you can find the fault!

CMD640 controller and a hard drive showing write errors you say?

https://www.mindprod.com/jgloss/eideflaw.html

https://en.wikipedia.org/wiki/CMD640

Yes. I know about the flaws but I thought they were mostly (only?) relevant when doing data transfers between two IDE drives connected to the same chip.
I always had only one drive connected. So I thought this could not be the issue. But like it seems a floppy does also count...

But what is this "fast mode" (from your first link):

Definitely Safe:
If you have a non-Intel BIOS and run only DOS, DESQview, Windows 3.1, Windows-95 and never use the "fast mode" simultaneous disk I/O feature on floppy or tape backup/restore, you are safe.

I am running DOS from CF-card and DOS & Win95 from HDD. Surely there was data copied to/from floppy sometime but whether it was done with "fast mode" I cannot say.

If I read further, under "speculation", it comes to my mind that I have played around with the 32-bit option for HDD access (don't know the exact notation) in the BIOS when looking for the maximum supported drive size. It is said that Win 95 is generally safe as is DOS in 16-bit mode. Maybe this could have been the initial fault which broke some files on the drive and eventually caused the write errors (as long as it is really no hardware fault and the 32-bit BIOS option relates to this 16-bit DOS mode).

Anyway, thanks for the hint.

@ChrisK, Would you be willing to share the gerber files for the L2 module to "the community"? I'd love to build one myself! 😁

Hi InjecTioN,

since there is a small (although easily patchable) bug in the current design files I would prefer to not make these files public. I could correct this but obviously there is very little demand so I'd like to not spend any time on it at the moment. I hope you can understand.

May I ask for what machine you would like to build one? Do you have a Tulip TC40 too? If not I would be very curious where else these modules fit.

If this is an option for you I can offer you an unpopulated PCB from my prototype batch & all necessary information to finalize it with your own parts.
Just let me know.

Hi Chris! Did you get my PM?

Hi Paadam,
I've seen your PM but unfortunately didn't have time to fully complete the module yet. Seems you can only have family OR hobby time...
I'm working on it!

Btw could you solve the problem with your machine not booting at 33 MHz?

I haven't had any luck finding the newer BIOS for the TC40. Only found an entry in the internet archive:
https://web.archive.org/web/2001040708 ... xpand=3#3
but unfortunately no luck in getting the appropriate file (40961021.EXE BIOS Upgrade TC40 21-OCT-96).
I also tried some Phoenix flashing tools for DOS but they require a platform.bin. Haven't tried the Windows versions yet.
Uniflash doesn't work. The aforementioned F3 option in my boot screen doesn't work. Grrr....
Any hints appreciated!

I found this same-looking memory/cache module in my big bag of unsorted SIMMs. Could it be something compatible with the TC40 board?

The connector is pretty standard and used for very different cache module designs.
On the first look I would guess that it will NOT work in the TC40 because of the missing logic required for decoding of the control signals.
There's only a 373 latch for address line buffering. The TC40 / Intel Aries chipset require some more decoding logic, hence the PAL on the original cache module.

I have one or two very similar looking modules which definitely also don't work. Better don't try it because you also don't know if that module creates some short circuits in the power supply lines.
I'll see if I can dig out some more info about your module tomorrow.

ChrisK wrote on 2021-10-21, 13:34:

Hi Paadam, I've seen your PM but unfortunately didn't have time to fully complete the module yet. Seems you can only have family […]
Show full quote

Hi Paadam,
I've seen your PM but unfortunately didn't have time to fully complete the module yet. Seems you can only have family OR hobby time...
I'm working on it!

Btw could you solve the problem with your machine not booting at 33 MHz?

I haven't had any luck finding the newer BIOS for the TC40. Only found an entry in the internet archive:
https://web.archive.org/web/2001040708 ... xpand=3#3
but unfortunately no luck in getting the appropriate file (40961021.EXE BIOS Upgrade TC40 21-OCT-96).
I also tried some Phoenix flashing tools for DOS but they require a platform.bin. Haven't tried the Windows versions yet.
Uniflash doesn't work. The aforementioned F3 option in my boot screen doesn't work. Grrr....
Any hints appreciated!

I managed to get a file here: https://web.archive.org/web/20030509042028/ht … ip/40961021.exe

Via here: https://web.archive.org/web/20070122114635/ht … rmimm/bios.html

I have not dealt with it yet, was busy whole summer building a garage for another hobby (old Porsches 🤣). But the machine is nearby and I will focus on it soon.
I could try to get a BIOS dumpsome time if needed.

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just there. Thank you so much, Cyrix200+!!!

Now reading the info file within it, the ominous F3 key comes into play again 😁
Seems it has to be pressed at POST while a diskette with the update on it is in the drive, then updating the BIOS.
Question is will it make a backup of the old one? Wouldn't be very funny if it breaks the machine after all.
I think I'll need a Schnaps or two before pressing F3 😳

Cheers and good night!

ChrisK wrote on 2021-10-21, 18:57:

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just […]
Show full quote

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just there. Thank you so much, Cyrix200+!!!

Now reading the info file within it, the ominous F3 key comes into play again 😁
Seems it has to be pressed at POST while a diskette with the update on it is in the drive, then updating the BIOS.
Question is will it make a backup of the old one? Wouldn't be very funny if it breaks the machine after all.
I think I'll need a Schnaps or two before pressing F3 😳

Cheers and good night!

Ha, you're welcome.

Sometimes you just need someone's differently trained Google algorithm to find something... 😀

ChrisK wrote on 2021-10-21, 18:57:

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just […]
Show full quote

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just there. Thank you so much, Cyrix200+!!!

Now reading the info file within it, the ominous F3 key comes into play again 😁
Seems it has to be pressed at POST while a diskette with the update on it is in the drive, then updating the BIOS.
Question is will it make a backup of the old one? Wouldn't be very funny if it breaks the machine after all.
I think I'll need a Schnaps or two before pressing F3 😳

Cheers and good night!

Hi ChrisK,

I have a spare TC40 mainboard with a fried IDE controller, but floppy works. If you dont want to put your machine at risk, i am willing to try to update this board first. Because its only catching dust hanging on the wall as deco. We can Facetime or something to try it together.
If it works we have a Schnaps ! I have some delicious Williams Birne.....

Peter

Cyrix200+ wrote on 2021-10-21, 19:00:

ChrisK wrote on 2021-10-21, 18:57:

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just […]
Show full quote

I can't believe it! I spent hours querying the net for this file. And suddely a minute or even less after asking here it is just there. Thank you so much, Cyrix200+!!!

Now reading the info file within it, the ominous F3 key comes into play again 😁
Seems it has to be pressed at POST while a diskette with the update on it is in the drive, then updating the BIOS.
Question is will it make a backup of the old one? Wouldn't be very funny if it breaks the machine after all.
I think I'll need a Schnaps or two before pressing F3 😳

Cheers and good night!

Ha, you're welcome.

Sometimes you just need someone's differently trained Google algorithm to find something... 😀

Seems so. Thanks a lot anyways!

Regarding your cache module "my" google algorithm just lead me to some very similar modules but no clue about the exact type number of yours.
According some old idt databooks modules of 6xxx series are application specific ones.
The closest match regarding the chips used seems to be the IDT7MP613x/615x series which are for the Intel 82420TX (Saturn) chipset. But their presence detect table would encode your module as "no cache present". One odd thing is that there seem to be 17ns SRAMs used which is quite unusual as at least I only know of 10/12/15/20ns types. So this really seems to be something "application specific".
If you don't know where this originally came from the best option probably is recycling the SRAM chips (or just putting it back into your bag of unsorted SIMMs) 😉

@ Peter
Williams Birne you say? Scotty, beam it up...
This is a very kind offer! I'd like to try the flashing procedure together but I can only offer Teams right now. And we would need to find a suitable time which seems to be the most difficulty to me.
Maybe you can make a proposal when it would be comfortable to you!?
I'd also be fine if you try yourself and maybe make some screenshots...

Hi ChrisK,

I tried the bios update today, unfortunately I turned my motherboard in "deco only" now. The update failed. I uploaded a video to YouTube:
https://youtu.be/MP31Kl5w9KI

Hitting F3 at the POST screen enables the flash programming, once booted you need to run "update.bat".
This board does not boot anymore, actually it does nothing.

I dont know why it failed, maybe its a problem with my board, since it acts weird sometimes. The CPU was installed in the wrong orientation once and that also damaged the IDE controller.

No Schnaps for me today.....

Good Luck,
peter

Sad to hear about this Peter!
Would you willing to sell the board to me? I may need clockgen from it...