VOGONS


Reply 20 of 31, by Rita's Cafe

User metadata
Rank Newbie
Rank
Newbie

I've acquired three DX4-100 from eBay in the past year. All are SK096 write back models. The left one and middle one came from a Chinese seller (sold together). The right one from a seller from USA.

As expected all three perform exactly the same scores on the benchmarks apart from one little oddity . The CPU on the right will not load the Quake benchmark when write back cache is enabled. Every other test demonstrates a performance difference between WB and WT. So I don't doubt that Write Back does work but it will not allow Quake to load. Any theories?
I should point out that all my tests were performed with the CPU overclocked to 120MHz with more than adequate cooling. Chipset is SiS471.

The CPU on the left strangely has a larger die cap than is normal for a DX4. Much like the one in the original post. Is this a late version DX4 but still earlier than the 2006/2007 font change?
When I first tried benchmarking this CPU it would not complete the Doom high settings benchmark. It repeatedly froze somewhere in the middle of the test. I tried write back & write through and both just resulted in a freeze during the test. I just assumed it was probably a remarked DX4-75. After all, I was trying to run it at 120MHz. Curious if these were re-badged I decided to clean the ceramic top of all three CPUs with acetone as well as cleaning the die cap of the left CPU. Absolutely no difference to appearance was observed. However the left CPU now completes the Doom benchmark flawlessly every time.

If I understand the date codes correctly then these are the dates:
LEFT: 1995 or 2005, week 21
MIDDLE: 1999, week 15
RIGHT: 1997, week 51

I'm not really sure what to make of all this.
Are these all "Genuine Intel" CPUs? Is the left CPU a 2005 model with a large die cap? What fabrication process is it, 0.8µm, 0.6µm, something smaller? Has anybody tried to de-lid one of these? And why won't the CPU on the right run Quake?

Attachments

  • bottom.jpg
    Filename
    bottom.jpg
    File size
    142.33 KiB
    Views
    351 views
    File comment
    DX4 100 SK096 (Bottom side)
    File license
    GPL-2.0-or-later
  • top.jpg
    Filename
    top.jpg
    File size
    107.89 KiB
    Views
    351 views
    File comment
    DX4 100 SK096 (Top side)
    File license
    GPL-2.0-or-later

Reply 22 of 31, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie

Funny thought...
What if this is actually a rebranded AMD?

Could also be a rebranded dx4-75 that cuts it at 100 but didn't at the Intel fab. Or one that is actually a 100 tested in the fab but was downrated by Intel to 75 to meet portable demands. Our it was already a 100 but markings were worn out and the owner did an attempt to restore (badly).
You just can't tell like this.

Any way, it looks like it is compromised in a certain way, which makes it a little bit collectable. I'd keep it this way 😀

Stuck at 10MHz...

Reply 23 of 31, by debs3759

User metadata
Rank Oldbie
Rank
Oldbie

I believe it is normal to see different caps. I think they were changed at some point in the production run. I remember reading about it somewhere, but can't remember where.

See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.

Reply 24 of 31, by CoffeeOne

User metadata
Rank Member
Rank
Member
Rita's Cafe wrote on 2022-07-17, 10:23:
I've acquired three DX4-100 from eBay in the past year. All are SK096 write back models. The left one and middle one came from a […]
Show full quote

I've acquired three DX4-100 from eBay in the past year. All are SK096 write back models. The left one and middle one came from a Chinese seller (sold together). The right one from a seller from USA.

As expected all three perform exactly the same scores on the benchmarks apart from one little oddity . The CPU on the right will not load the Quake benchmark when write back cache is enabled. Every other test demonstrates a performance difference between WB and WT. So I don't doubt that Write Back does work but it will not allow Quake to load. Any theories?
I should point out that all my tests were performed with the CPU overclocked to 120MHz with more than adequate cooling. Chipset is SiS471.

The CPU on the left strangely has a larger die cap than is normal for a DX4. Much like the one in the original post. Is this a late version DX4 but still earlier than the 2006/2007 font change?
When I first tried benchmarking this CPU it would not complete the Doom high settings benchmark. It repeatedly froze somewhere in the middle of the test. I tried write back & write through and both just resulted in a freeze during the test. I just assumed it was probably a remarked DX4-75. After all, I was trying to run it at 120MHz. Curious if these were re-badged I decided to clean the ceramic top of all three CPUs with acetone as well as cleaning the die cap of the left CPU. Absolutely no difference to appearance was observed. However the left CPU now completes the Doom benchmark flawlessly every time.

If I understand the date codes correctly then these are the dates:
LEFT: 1995 or 2005, week 21
MIDDLE: 1999, week 15
RIGHT: 1997, week 51

I'm not really sure what to make of all this.
Are these all "Genuine Intel" CPUs? Is the left CPU a 2005 model with a large die cap? What fabrication process is it, 0.8µm, 0.6µm, something smaller? Has anybody tried to de-lid one of these? And why won't the CPU on the right run Quake?

The left one is 100% not an Intel DX4 write back (&EW). The DX4 write-back is a very late 486 with a smaller die. Maybe it is a Cyrix.
The other 2 look good to me.

Reply 25 of 31, by Rita's Cafe

User metadata
Rank Newbie
Rank
Newbie
CoffeeOne wrote on 2022-07-17, 22:39:

The left one is 100% not an Intel DX4 write back (&EW). The DX4 write-back is a very late 486 with a smaller die. Maybe it is a Cyrix.
The other 2 look good to me.

If it's not an Intel DX4 then what is it? It gets the exact same results as an Intel DX4 in the benchmarks. If it were a rebadged AMD or Cyrix then surely the benchmarks would show scores representative of them. See Feipoa's "The Ultimate 486 Benchmark Comparison."

waterbeesje wrote on 2022-07-17, 18:07:
Funny thought... What if this is actually a rebranded AMD? […]
Show full quote

Funny thought...
What if this is actually a rebranded AMD?

Could also be a rebranded dx4-75 that cuts it at 100 but didn't at the Intel fab. Or one that is actually a 100 tested in the fab but was downrated by Intel to 75 to meet portable demands. Our it was already a 100 but markings were worn out and the owner did an attempt to restore (badly).
You just can't tell like this.

Any way, it looks like it is compromised in a certain way, which makes it a little bit collectable. I'd keep it this way 😀

Cleaning the tops of all three CPUs using acetone didn't reveal any evidence of a coverup. I could try scratching away the top layer using something like a chisel. Although all three are from Malaysia, only the right one has laser etched text. The other two are printed. A few more CPU comparisons will need to be done but I feel you could be right about it being a rebadged DX4-75. However I tested all these at 120MHz. That's 1.6x the rated frequency!

debs3759 wrote on 2022-07-17, 19:49:

I believe it is normal to see different caps. I think they were changed at some point in the production run. I remember reading about it somewhere, but can't remember where.

Perhaps they really are all genuine (and the one on the right is just faulty) and Intel did just change to larger die caps later into the 486 life cycle. The one shown in the original post is supposedly a 2007 model and it has the large die cap.

I am a little curious to see if these later models can clock higher the the ones from the 90's. Is there a good source of info regarding DX4-100's typical overclocking capabilities/difficulties? Maybe different plants had different tolerances; Malaysia, Philippines, Costa Rica...

Reply 27 of 31, by waterbeesje

User metadata
Rank Oldbie
Rank
Oldbie

If the CPU had a fine dx4-100 test, but was badges to 75 by Intel in first place, there should be no harm indeed.
If it's actually tested to be a 75 but now cuts it at 100, there might be a small risk for a possible instability under certain circumstances. I wouldn't bother.
In general, lots of Intel CPUs were very concervative positioned.
Lots of Intel CPUs could handle more than they had to. Some DX2-66 run fine at 80MHz and some run fine at 66@3,3v. Lots of dx4-75 can handle 100MHz too and dx4-100 might handle 120 give (no experience). This is in contrast to some earlier AMD or cyrix that already ran very (too?) close to their thermal limits.

Stuck at 10MHz...

Reply 28 of 31, by CoffeeOne

User metadata
Rank Member
Rank
Member
Rita's Cafe wrote on 2022-07-18, 10:02:

....
Perhaps they really are all genuine (and the one on the right is just faulty) and Intel did just change to larger die caps later into the 486 life cycle. The one shown in the original post is supposedly a 2007 model and it has the large die cap.
....

An Intel DX4-100 from 2007? I still don't believe it. Did they produce it that long?

Reply 29 of 31, by debs3759

User metadata
Rank Oldbie
Rank
Oldbie

Yes, 486 had a really long production run for some uses

See my graphics card database at www.gpuzoo.com
Constantly being worked on. Feel free to message me with any corrections or details of cards you would like me to research and add.

Reply 30 of 31, by Sphere478

User metadata
Rank l33t
Rank
l33t

Has anyone identified the sl configuration pin? (If that is even what is happening)

Or did I miss it?

I can incorporate it into the tweaker if we know what pin is being pulled to what.

Which may allow people to get these working on unsupported mobos. Assuming bios support isn’t the problem.

-

Btw, 🤣, you guys went off the deep end with this conspiracy theory. At first I was like, that looks fake as shit! But as soon as someone pointed out (reminded me) that the 486 was in production for so long it made total sense. Yeah, I totally believe these chips are legit now.

And I strongly suspect the ones that aren’t working are because of mobo support. I rather doubt that they are duds.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 31 of 31, by CoffeeOne

User metadata
Rank Member
Rank
Member
Sphere478 wrote on 2022-07-20, 06:32:
Has anyone identified the sl configuration pin? (If that is even what is happening) […]
Show full quote

Has anyone identified the sl configuration pin? (If that is even what is happening)

Or did I miss it?

I can incorporate it into the tweaker if we know what pin is being pulled to what.

Which may allow people to get these working on unsupported mobos. Assuming bios support isn’t the problem.

-

Btw, 🤣, you guys went off the deep end with this conspiracy theory. At first I was like, that looks fake as shit! But as soon as someone pointed out (reminded me) that the 486 was in production for so long it made total sense. Yeah, I totally believe these chips are legit now.

And I strongly suspect the ones that aren’t working are because of mobo support. I rather doubt that they are duds.

Hmm, maybe Intel still secretly produces some 486 chips. Or even 386.
Who knows?