Just updated your spreadsheet with results using my new Intel DX4 chip. I got an overall improvement of 13.34fps, which I think is quite impressive. Intel clearly was faster than Cyrix when it came to the DX4 models.
Especially the improvement in PCPbench are impressive, a whopping 5.4fps increase. That's an increase of 39.4% ! Astonishing in my opinion.

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

I'll test that 😀 The 5x86 has twice as much L1 cache. Too bad I don't own any Intel DX4 chips.

RacoonRider wrote:

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

I'll test that 😀 The 5x86 has twice as much L1 cache. Too bad I don't own any Intel DX4 chips.

Already been proven by feipoa in his Ultimate 486 Benchmark. (People should start referencing this material more often)

Observing just the three DX4-100's (Intel, AMD-WT, and Cyrix), the Intel piece won by a landslide, followed by the AMD and Cyrix parts. This may be understood by the following explanation: The Intel CPU has 16 KB of write-back cache while the Cyrix part only has 8 KB of write-back cache. The AMD falls in last because it has 8 KB of write-through cache. An AMD DX4-100 with 8 KB of write-back cache does exist, but was not available for testing. There also exist 16 KB, write-back versions of the AMD DX4-100 (which were later added to the chart and are referred to as AMD DX4-100-WB) which emerged more than a year after the X5 133. The AMD DX4-100-WB-16KB was not commonly found in consumer-marketed 486 machines, however, for completeness, a 100 MHz down-clocked X5-133 was added to the charts to simulate this CPU. The AMD DX4-100-WT was the most common of the AMD DX4's at 100 MHz. The AMD DX4-100-WB now steps ahead of the Cyrix DX4-100 in ALU performance (it has 8 KB more cache than the Cyrix), however still falls considerably behind the Intel DX4-100. The Intel piece clearly had a performance edge for DX4-100 class processors.

Please note that the Cyrix 5x86 does beat the Intel DX4 clock-for-clock

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

The 5x86 always struck me as being kind of a sluggish chip for its supposed clock speed. Would a slightly overclocked DX4 100 be able to outperform one?

Then again, the one 5x86 system I had a long time ago was rather poorly set up, and when it had a 66MHz DX2, it performed nowhere near as well as my other DX2 system despite having more ram and a newer video card.

mr_bigmouth_502 wrote:

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

The 5x86 always struck me as being kind of a sluggish chip for its supposed clock speed. Would a slightly overclocked DX4 100 be able to outperform one?

According to feipoa's benchmark, no.

This is a Intel DX-120 vs a AMD 5x86 X5-133. While the DX-120 is a 0.1 point quicker in ALU performance, the X5-133 wins with better FPU, so in overall the X5-133 wins

vetz wrote:

RacoonRider wrote:

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

I'll test that 😀 The 5x86 has twice as much L1 cache. Too bad I don't own any Intel DX4 chips.

Already been proven by feipoa in his Ultimate 486 Benchmark. (People should start referencing this material more often)

Okay 😀

I think calling it sluggish is misleading 😀 I don't want to be the reason for spreading false rumours 😊

But I found that compatibility with motherboards and BIOS are better with the IntelDX4 and that the 5x86, while being faster, isn't that much faster and doesn't make games that struggle on a IntelDX4 suddenly fluid.

Edit IntelDX4 100 IntelDX4-100 Integrated FPU 486 (Socket 1-3) 33 MHz 3x 100 MHz 70.1 20.3 40.16 10.6 Biostar 8433UUD-A VER:2

11 Edit Am5x86 133 Am5x86-133 Integrated FPU 486 (Socket 1-3) 33 MHz 4x 133MHz 79.9 20.4 46.28 12.6 Biostar 8433UUD-A VER:2

The scores are for 3dbench2, pcpbench/vgamode, Doom and quake.

RacoonRider wrote:

Mau1wurf1977 wrote:

Yes you can't go wrong with Intel.

I found the IntelDX4 to be faster, clock for clock, compared to the 5x86 P75 as well.

I'll test that 😀 The 5x86 has twice as much L1 cache. Too bad I don't own any Intel DX4 chips.

My DX4 has 16KB WB L1 cache, so no, the 5x86 doesn't have twice as much cache.
I guess that cache is part of the reason for the massive performance boost.

Mau1wurf1977 wrote:

I think calling it sluggish is misleading :) I don't want to be the reason for spreading false rumours :blush: […]
Show full quote

I think calling it sluggish is misleading 😀 I don't want to be the reason for spreading false rumours 😊

But I found that compatibility with motherboards and BIOS are better with the IntelDX4 and that the 5x86, while being faster, isn't that much faster and doesn't make games that struggle on a IntelDX4 suddenly fluid.

Edit IntelDX4 100 IntelDX4-100 Integrated FPU 486 (Socket 1-3) 33 MHz 3x 100 MHz 70.1 20.3 40.16 10.6 Biostar 8433UUD-A VER:2

11 Edit Am5x86 133 Am5x86-133 Integrated FPU 486 (Socket 1-3) 33 MHz 4x 133MHz 79.9 20.4 46.28 12.6 Biostar 8433UUD-A VER:2

The scores are for 3dbench2, pcpbench/vgamode, Doom and quake.

What if you ran that DX4 at a 4x multiplier, or compared the two with a higher FSB setting?

Ignore me guess Quake doesn't run without FPU 😀

mr_bigmouth_502 wrote:

Mau1wurf1977 wrote:

I think calling it sluggish is misleading :) I don't want to be the reason for spreading false rumours :blush: […]
Show full quote

I think calling it sluggish is misleading 😀 I don't want to be the reason for spreading false rumours 😊

But I found that compatibility with motherboards and BIOS are better with the IntelDX4 and that the 5x86, while being faster, isn't that much faster and doesn't make games that struggle on a IntelDX4 suddenly fluid.

Edit IntelDX4 100 IntelDX4-100 Integrated FPU 486 (Socket 1-3) 33 MHz 3x 100 MHz 70.1 20.3 40.16 10.6 Biostar 8433UUD-A VER:2

11 Edit Am5x86 133 Am5x86-133 Integrated FPU 486 (Socket 1-3) 33 MHz 4x 133MHz 79.9 20.4 46.28 12.6 Biostar 8433UUD-A VER:2

The scores are for 3dbench2, pcpbench/vgamode, Doom and quake.

What if you ran that DX4 at a 4x multiplier, or compared the two with a higher FSB setting?

DX4 has 2x multiplier instead of 4x.

dacow wrote:

Ignore me guess Quake doesn't run without FPU 😀

It doesn't. But 486DX do have it integrated. Look in the spreadsheet for benchmarks.

kixs wrote:

It doesn't. But 486DX do have it integrated. Look in the spreadsheet for benchmarks.

Ahh yeah my original post about me adding a 386DX40 into the spreadsheet but I didn't realise that most of the other 386DX's (without co-pro's) had 0's in the Quake column!

Added a bulk of information from earlier runs, fixed validation up to row 450, now the perfectionist inside me is happy.

dacow, AMI BIOS string you posted suggests your 386DX-40 motherboard is made by Soyo. To identify your board look for a small yellow sticker, it should say something like "019R1":
http://lh5.ggpht.com/_bNUAEXCfAto/TIjdrJiEAJI … 24/PICT0060.JPG
That's the way they labeled their boards.

Or look it up in th99: http://museum.ttrk.ee/th99/

I've entered my Cyrix 6x86 P80 benchmark

Original CPU running at 40mhz FSB and 2x multiplier as intended. It's slow!

RacoonRider wrote:

Or look it up in th99: http://museum.ttrk.ee/th99/

I just realised that mobo has been packed back up and shipped back to Badmofo!

Added two 386DX40s and a 486DX2-66 to the database! 😀
Tomorrow I will test 10-15 more combinations, ranging from 386SX-20 to Pentium III 750! 😁

Nice GeorgeMan! That's a proper benchmarking marathon 😀

33 results uploaded!
I saw no point in trying P133-K6/300 on the 2 newer motherboards, if anyone else wants just say it. 😀
It is also pointless for me to bench more PIII/Celeron (366-933MHz) as they are pretty much too fast for these DOS benchies.