I would be careful in inserting a 486DX4 100 MHz CPU, since this is a 3.3V CPU and most 486DX2 66 MHz CPU's were still 5V.
Your motherboard does appear to have jumper settings silkscreened on the PCB to change the voltage (refer just below the CPU on the first picture your posted).
If you didn't change those jumper settings, then there is a chance that you could have fried your 486DX4 100 MHz CPU.

Insert the 486DX2 66 MHz CPU again and take a pic of the "POST" screen (where it counts down the memory).
We might be able to identify the motherboard from what is displayed on there.
A full size picture of the entire motherboard might also help, as you can then try and identify it from here: http://arvutimuuseum.ee/th99/#1

jesolo wrote:

I would be careful in inserting a 486DX4 100 MHz CPU, since this is a 3.3V CPU and most 486DX2 66 MHz CPU's were still 5V. Your […]
Show full quote

I would be careful in inserting a 486DX4 100 MHz CPU, since this is a 3.3V CPU and most 486DX2 66 MHz CPU's were still 5V.
Your motherboard does appear to have jumper settings silkscreened on the PCB to change the voltage (refer just below the CPU on the first picture your posted).
If you didn't change those jumper settings, then there is a chance that you could have fried your 486DX4 100 MHz CPU.

Insert the 486DX2 66 MHz CPU again and take a pic of the "POST" screen (where it counts down the memory).
We might be able to identify the motherboard from what is displayed on there.
A full size picture of the entire motherboard might also help, as you can then try and identify it from here: http://arvutimuuseum.ee/th99/#1

Thanks for replying jesolo, yes of course I changed the voltage jumpers before inserting the dx4.

I’ll try to get thise pics and post back.

I do fear naybe the dx4 was dead anyway because it doesn’t even get warm...

Ok, switched jumpers back, re-installed the DX2, boots up fine, and here are the pics:

walterg74 wrote:

I do fear naybe the dx4 was dead anyway because it doesn’t even get warm...

It's also possible that the CPU itself is fine. It may be the motherboard's voltage regulator that's dead - assuming you set the voltage jumpers to 3.3V for the DX4. In 5V setting the board won't use the regulator, it'll simply feed 5V directly from the PSU to the processor. So you may even call it an "escape route" in case of a dead regulator. The other half of this puzzle could be a DX4 Overdrive CPU which has its own regulator and thus runs from 5 Volts.

alvaro84 wrote:

walterg74 wrote:

I do fear naybe the dx4 was dead anyway because it doesn’t even get warm...

It's also possible that the CPU itself is fine. It may be the motherboard's voltage regulator that's dead - assuming you set the voltage jumpers to 3.3V for the DX4. In 5V setting the board won't use the regulator, it'll simply feed 5V directly from the PSU to the processor. So you may even call it an "escape route" in case of a dead regulator. The other half of this puzzle could be a DX4 Overdrive CPU which has its own regulator and thus runs from 5 Volts.

Ok, so I think maybe that could be the answer, and it's the only thing I can think of. Here's why...

- Looking at the string on POST, this seems to be an m912 v1.7 (now why it says v6.1 on the corner beats me, because I haven't found any evidence on that version)

- I double checked the jumpers, and they seemed to be fine and as they should be

- I have another board, a v1.4 which is the one I have configured with the DX2-66. Since it says it too supports the DX4-100, I re-jumpered it for DX4-100 configuration.

- I turned it on and voila, it booted up fine, correctly identified, 100Mhz, etc.

So I think your voltage regulator idea could be right and that's the culprit (and if no Voltage gets to the CPU it would also explain why it just sits there cold, too).

So are these easy to find and replace?

They should be easy to find and replace.
On a side note, it seems your motherboard has fake cache:
http://www.elhvb.com/webhq/models/486vlb3/m912fake.htm
Compared to the one that does have cache memory: http://www.elhvb.com/webhq/models/486vlb3/m912v17.htm

Best way to confirm is to run a cache check utility.

jesolo wrote:

They should be easy to find and replace. On a side note, it seems your motherboard has fake cache: http://www.elhvb.com/webhq/m […]
Show full quote

They should be easy to find and replace.
On a side note, it seems your motherboard has fake cache:
http://www.elhvb.com/webhq/models/486vlb3/m912fake.htm
Compared to the one that does have cache memory: http://www.elhvb.com/webhq/models/486vlb3/m912v17.htm

Best way to confirm is to run a cache check utility.

Just in case, is the regulator the one right by the cpu, the one resting on the heatsink? Do you know how to search for one/part number or something?

Thanks.

PCB is different around the cache area compared to normal 1.7 revision, so 6.1 is probably indicates this changes. Also it looks like you can't add 512kb cache here (no room).
Required VR parts were already discussed here: 486 Motherboard working with 5v CPUs but not with 3v
Personally don't think it's worth to repair, too much hassle. You need to solder a keyboard DIN connector, a new set of cache and two VR parts.