Normal i80486DX2-66

Normal i80486DX2-66

Hi,

Unless there was a really weird day at the chip packaging part of the facility, i would say it is remarked, I mean it is the top part of the CPU that is visible when installed into the system that is claiming a higher frequency than the bottom part, typically done by people who remarked the CPU and sold it installed in a motherboard or a whole computer, never expecting someone to check the bottom of the IC. It is done rather professionally though, so maybe Intel themselves found out that some i50 grade ones run as i66-es, and forgot to change the bottom marking?

MagefromAntares wrote on 2026-06-06, 13:07:

Hi,

Unless there was a really weird day at the chip packaging part of the facility, i would say it is remarked, I mean it is the top part of the CPU that is visible when installed into the system that is claiming a higher frequency than the bottom part, typically done by people who remarked the CPU and sold it installed in a motherboard or a whole computer, never expecting someone to check the bottom of the IC. It is done rather professionally though, so maybe Intel themselves found out that some i50 grade ones run as i66-es, and forgot to change the bottom marking?

This i80486DX2-66 has passed the test and is very stable at 66 MHz.
All the marks on this CPU don't seem to have been remarked.
It's really weird.

ekkiller wrote on 2026-06-06, 13:22:

It's really weird.

I suspect it's a fake, what CPUID is it? My SX911 reports as below so it's 0435, I half suspect yours will report 0436.

SIV can be downloaded via http://rh-software.com/

red-ray wrote on 2026-06-10, 05:39:

I suspect it's a fake, what CPUID is it? My SX911 reports as below so it's 0435, I half suspect yours will report 0436. […]
Show full quote

ekkiller wrote on 2026-06-06, 13:22:

It's really weird.

I suspect it's a fake, what CPUID is it? My SX911 reports as below so it's 0435, I half suspect yours will report 0436.

SIV can be downloaded via http://rh-software.com/

SIV report shows sx954, confirming this is a remarked i80486DX2-50.

I wonder instead of it being re-marked, if it could simply have been binned as a DX2-50, then somehow ended up in the DX2-66 batch and received the wrong markings but happens to run well enough at 66MHz anyway.

If it was a real 66, it should be able to do 80Mhz rather easily provided you keep it cool. It may not be stable at 80Mhz in absolutely everything, but it should be fine for most things.

A DX/2-50 with writeback cache. I've never seen one of those before.

classic_cola wrote on 2026-06-12, 03:03:

I wonder instead of it being re-marked, if it could simply have been binned as a DX2-50, then somehow ended up in the DX2-66 batch and received the wrong markings but happens to run well enough at 66MHz anyway.

No, it's a fake, it has been remarked '89, '92 when it's '89, '93. Whoever did this was quite stupid as if they had marked it as SX955 it would be more difficult to tell it's a fake, I suspect the fake package is thinner than it should be.

cyclone3d wrote on 2026-06-12, 04:17:

If it was a real 66, it should be able to do 80Mhz rather easily provided you keep it cool. It may not be stable at 80Mhz in absolutely everything, but it should be fine for most things.

No, I have a real SX955 and it will only do 67MHz.

It's as I suspected, CPUID 0436 is a SX954 in WT rather than WB mode. Menu->Hardware->CPU S-Spec->i486 CPU S-Spec shows 2 x SX954 with different CPUIDs

red-ray wrote on 2026-06-12, 06:47:

No, it's a fake, it has been remarked '89, '92 when it's '89, '93. Whoever did this was quite stupid as if they had marked it as SX955 it would be more difficult to tell it's a fake, I suspect the fake package is thinner than it should be.

It can be confirmed that this i80486DX-66 has been polished. Compared with the authentic i80486DX2-50 with a similar number on the bottom surface, it is evident that a layer has been removed from the top surface.