VOGONS


First post, by chublord

User metadata
Rank Member
Rank
Member

Just got a NOS DX4-100 chip but can't get it to post. Anything odd about it or did I just get a bad chip? Tried 3.0V-3.5V, nothing. System boots right up with the old DX4-100 &E chip. System a IBM Valuepoint 466DX2/Dp.

Do some systems not work with the write-back cache (&EW)?

Last edited by chublord on 2020-07-25, 03:54. Edited 1 time in total.

IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2

Reply 1 of 45, by Horun

User metadata
Rank l33t++
Rank
l33t++

Hmm the only ones I have seen have a gold plate on the bottom (not painted), and some numbering on it. Am sure there are other variants but the lack of circled R after Intel in top line makes me suspect as Intel always included that because of registered copyright. Not saying it is fake but have a read here at cpu-world:
real ones: http://www.cpu-world.com/CPUs/80486/Intel-A80486DX4-100.html
about a possible fake: http://www.cpu-world.com/forum/viewtopic.php?t=13445

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 2 of 45, by auron

User metadata
Rank Oldbie
Rank
Oldbie

as seen in this intel document linked in the thread above, the dx2-66 and dx4-100 were indeed affected by the logo change, the SK096 s-spec matches up, and the (200)7 week 26 date code makes sense at a glance as well (deadline for the old chips was march 2007 according to that document, and they were only discontinued in september that year according to wikipedia). a dx2-66 looking just like that can also be seen here: http://www.cpushack.com/2013/01/26/cpu-of-the … tel-486-dx2-66/

can't say for sure that these aren't fakes, as the missing (R) seems odd to me as well, but if they are, the fakers certainly have put a bit of thought into these markings. but was there really a huge market for faking these in 2007 though? i'm assuming this would have been a final run of low-volume replacement parts for the embedded market, though i wonder about the actual prices...

Reply 3 of 45, by chublord

User metadata
Rank Member
Rank
Member

I did notice the die looks too big. Seems odd there would be a market for fakes??? I'm curious what it actually is. I'm tempted to turn it up to 5V and see if it's a phony DX2 or something, of course that would fry it if it's a legit chip.

IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2

Reply 4 of 45, by Horun

User metadata
Rank l33t++
Rank
l33t++

Try DX2-66 settings on the mobo and 3.3v, if it does not boot try 3.45v if board supports it. If it boots then it could be a late SL DX66 or a real DX-100, no way to tell afaik but at least you have a cpu that boots. so not all is lost. Older boards did not rely on a S-SPEC to boot listed in BIOS (unlike later Pentium 2, 3 etc) as long as the Volts and jumpers were set correct. Good luck !

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 5 of 45, by chublord

User metadata
Rank Member
Rank
Member

Just tried it at 3.9V @ 66 MHz, no post. Also tried 50 MHz. Edit: messaged the seller mentioning the chip is either defective or fake. Let's see if he responds.

IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2

Reply 6 of 45, by Socket3

User metadata
Rank Oldbie
Rank
Oldbie

Shouldn't there be some markings on the back of the CPU as well? I can't tell from the pictures, but it's fishy if there are no markings at all... the blank back makes me think of the dummy CPU's some retailers received for presentation and exposition back in the day.

Reply 7 of 45, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

Indeed, this is a weird one. I've seen these on Taobao for a while, and I was always kind of sceptical that these were legit.
As others have pointed out, the die cap area seems too big for a 3.3V intel chip. It seems to have the same proportions as the older .8 micron 5V 486s.
Actually, from the photos it almost looks like the die hasn't even been installed. It looks inverted, but I assume this is an optical illusion.
I suggest seeing if the black covering can be scraped off to reveal gold underneath. If it's a fake, maybe the real markings are covered up.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium

Reply 8 of 45, by imi

User metadata
Rank l33t
Rank
l33t

well the font doesn't look like an intel logo at all at least and the bottom markings are usually lasered and not printed:

edit: oh didn't see the post about the logo change, that's interesting.

Reply 10 of 45, by Horun

User metadata
Rank l33t++
Rank
l33t++
bearking wrote on 2020-07-25, 17:35:

Here is my DX4-100 SK096 with the old logo...
Re: I recently found this hardware, AKA the Dumpster find thread.

That is how the one I personally saw looked and it had gold bottom plate with some numbering, am sure yours does too (though have seen some 486 where the black numbers have faded/missing off the bottom gold plate).

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 11 of 45, by bearking

User metadata
Rank Member
Rank
Member
Horun wrote on 2020-07-25, 18:49:
bearking wrote on 2020-07-25, 17:35:

Here is my DX4-100 SK096 with the old logo...
Re: I recently found this hardware, AKA the Dumpster find thread.

That is how the one I personally saw looked and it had gold bottom plate with some numbering, am sure yours does too (though have seen some 486 where the black numbers have faded/missing off the bottom gold plate).

You made me curiuos, so I popped it out from the socket and took a picture of the bottom of it. Here it is...

Reply 12 of 45, by Horun

User metadata
Rank l33t++
Rank
l33t++
bearking wrote on 2020-07-25, 19:30:
Horun wrote on 2020-07-25, 18:49:
bearking wrote on 2020-07-25, 17:35:

Here is my DX4-100 SK096 with the old logo...
Re: I recently found this hardware, AKA the Dumpster find thread.

That is how the one I personally saw looked and it had gold bottom plate with some numbering, am sure yours does too (though have seen some 486 where the black numbers have faded/missing off the bottom gold plate).

You made me curiuos, so I popped it out from the socket and took a picture of the bottom of it. Here it is...

Looks real ! Nice gold plate and Small die size compared to SX/DX cpu's. I should check my Intel DX4ODPR100, am sure bottom looks just like yours but is in a machine and hate to open it up.

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 13 of 45, by chublord

User metadata
Rank Member
Rank
Member

eBay seller responded:

"First, we build circuit boards for a third party and we purchased several hundred of of these DX4 processors. We completed the project and had 127 left over still in the shrink wrapped package.. All of the board delivered to the third party worked and were satisfactory. We have also sold a number of the excess processors here on eBay and all but you have reported satisfaction with them so there is NO possibility that these processors are not genuine."

Is there any chance I am wrong here??

I am curious to try it at 5V to see if it posts, but I don't want to do that if there's any chance it's a 3 volt chip.

IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2

Reply 14 of 45, by wirerogue

User metadata
Rank Member
Rank
Member
chublord wrote on 2020-07-30, 23:37:

Is there any chance I am wrong here??

it's time to play my favorite ebay game "is this broken or am i just stupid"

Reply 15 of 45, by chublord

User metadata
Rank Member
Rank
Member

I tried it again in my system...and it actually booted once! It showed 100 MHz, but I can't get it to post again. Something doesn't add up.

IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2

Reply 16 of 45, by Horun

User metadata
Rank l33t++
Rank
l33t++

edit: OK so seller says it is not fake. But it could be defective ! They did not deny that. If they bought "several hundred" of those, possible a few are bad and you got one. Also is possible they already had some returns and they shipped you one of them if some are bad and not faked. I think it is a fake DX100 but proving it is different than using Rule #123: never claim a fake back to seller. Just tell them it arrived not working and either want a replacement or refund. added: on items from Ebay I always a VISA card thru paypal not a debit card as it also gives you other protections though could take a few months to get resolved.

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. Stuff: https://archive.org/details/@horun

Reply 17 of 45, by Intel486dx33

User metadata
Rank l33t++
Rank
l33t++
chublord wrote on 2020-07-31, 00:22:

I tried it again in my system...and it actually booted once! It showed 100 MHz, but I can't get it to post again. Something doesn't add up.

1) Make sure you have a good battery ( test with multi-meter )
2) Remove all cards and devices from your motherboard but the video card.
3) press down on cache chips to make sure they are seated correctly.
4) double check jumpers.
5) don't increase your 486dx4-100 CPU voltage past 3v. ( you will damage the chip ).
6) check your motherboard with another CPU to make sure everything is working fine with your motherboard.
7) remove all RAM modules but two parity simms. Make sure you are using the correct ram ( Fast Page mode Parity ) or EDO if your motherboard supports it.
8 ) I believe for DX4 the P24D CPU Selection should be "enabled".

Here are the settings form my Gigabyte motherboard VS-486 for reference.
chrome-extension://oemmndcbldboiebfnladdacbdfmadadm/http://www.motherboards.org/files/manuals/47/486vs8a.pdf

Reply 18 of 45, by amadeus777999

User metadata
Rank Oldbie
Rank
Oldbie

If the cpu does not work in a system that did run with an "identical" one before then your cpu may be faulty. The write back / through mode issue should not stop the cpu from booting - this is a per board problem with some being pickier than others.
Can't say too much about the look of the cpu but the few DX4s that I had all had a golden "belly".

Reply 19 of 45, by Anonymous Coward

User metadata
Rank l33t
Rank
l33t

I have seen DX/2s with a black die cap, but this is the first time I have seen a DX/4 with a black die cap, especially considering how large it is. Come to think of it, it seems to be about the same size as the one used on the 1 micron models. The DX/2s and the DX-50 were 0.8 micron and had a slightly smaller die cap. The DX/4s were .6 micron and were even smaller. If the CPU is legit, I'd be really curious to remove the die cap to get a clear shot of the die.

"Will the highways on the internets become more few?" -Gee Dubya
V'Ger XT|Upgraded AT|Ultimate 386|Super VL/EISA 486|SMP VL/EISA Pentium