VOGONS


First post, by tour86rocker

User metadata
Rank Newbie
Rank
Newbie

I have a Compaq Deskpro XE 466 computer that I am trying to sell to a guy who wants a 486 66MHz microprocessor in it. It actually came to me with a stronger microprocessor that he doesn't want. I tried to transplant the 66MHz microprocessor into the other box and it booted to Win95 but has had some unusual errors, and eventually became unresponsive. On reboot, it complained of a memory error, which it never has done before. A set of jumpers near the microprocessor socket already had identical settings on both machines so I'm assuming that this is set up fine.

Any ideas what I might be able to try to resolve this? I was going to try another hard drive with a fresh install of Win95, but then the BIOS gave a memory error, which would be unrelated to OS.

Reply 3 of 6, by tour86rocker

User metadata
Rank Newbie
Rank
Newbie
DNSDies wrote on 2021-02-04, 21:56:

Is there a way to find out whether I have "ODP486SX-33/66" or "80486DX2-66"? CHKCPU gave me the following data on it, which I'm guessing is insufficient:

CPU Vendor and Model: 'classic' 486
Internal CPU speed : 67.1 MHz
CPU-ID Vendor string: - No CPUID -
Current CPI mode : Real
Internal (L1) cache : Enabled in Write-Through mode

The most likely case is that this is an Intel i486 DX2 since I think that's what comes with a Compaq Deskpro XE 466 and because the bottom of the chip looks very much like others of its kind. I can't know for sure, because the top has a heat sink and the number on the bottom doesn't turn up in Google. And if it is what I assume, the CPU TYPE CONFIGURATION switches are already correctly set.

Reply 5 of 6, by tour86rocker

User metadata
Rank Newbie
Rank
Newbie
Warlord wrote on 2021-02-05, 08:38:

most likely, memory error could be due to incorrect wait state settings, or l2 cache. or bad ram.

I don't know how to set "wait state settings" but the RAM and L2 cache card haven't caused problems before

Reply 6 of 6, by shamino

User metadata
Rank l33t
Rank
l33t

I don't have much experience with 486s, but are you sure it's getting the correct CPU voltage? I'm guessing the faster CPU was 3.3V, and at least the DX2-66 I had back then was 5V.
Ideally, can you measure the voltage with a meter to confirm what it's really getting? Don't take my word for it needing 5V though - I'm just wondering if that could be the cause of all this.

As far as identifying the CPU:
I was recently reading about the possibility of performing a "CPU ID" (not the CPUID instruction) on 486 CPUs immediately after they've been reset. I'm not sure it can be done in all motherboards though, since the check has to be done at reset.