"As an aside, is a 286 capable of running Doom?"

- No. Doom3D required 32-bit processor.

For statistic: Performance comparison of CPU: 286-25 vs 386DX-25 vs 386SX-25

Ekb wrote:

"As an aside, is a 286 capable of running Doom?"

- No. Doom3D required 32-bit processor.

For statistic: Performance comparison of CPU: 286-25 vs 386DX-25 vs 386SX-25

Thanks. That means PCPBench and Speedsys also will not work on a 286. Good to know!

BTW, how did you run the Wolf3d benchmark? Is there a modified download somewhere with instructions?

Edit: I've taken some results from your thread and done some more research to find old benchmark results elsewhere. I've updated my chart and added Norton SI6. This chart is simply to give an estimate of how each processor would do in each of these benchmarks. In other words, "typical expected performance for a well-configured machine".

A 286 is just a few % faster(!) than a 386SX at the same clockspeed, so if you have 386SX results, you pretty much know what a 286 would do.

"BTW, how did you run the Wolf3d benchmark? Is there a modified download somewhere with instructions?"

- this link has a "download" ("Скачать") button: Performance comparison of CPU: 286-25 vs 386DX-25 vs 386SX-25

Spaseeba. 😉

EDIT: Here is an updated chart with some Wolf3D results. I took Ekb's results from his thread he linked to, along with my own results.

clueless1 wrote:

Spaseeba. 😉

Vy bol'she nužny boepripasy? 😜

Nyet...ya harasho!

This is what I ended up with. Nowhere near perfect by any stretch, but it will do for ballpark estimating. The bulk of the results are real, taken from my own machines, Phil's VGA Benchmark Database, my Cachbench database, the Ultimate 486/686 Benchmark Comparisons stickied in Marvin, as well as some results taken from various threads on here as well as elsewhere on the web. The results that were still missing I derived from neighboring results.

The purpose of this chart is for anyone who is slowing their machine down to play speed sensitive games. This will help them pinpoint about where their machine is performing with whatever slowdown method they're using. Keep in mind that the results for PCPBench, Speedsys and Doom on the 286s are completely derived, as those benchmarks require a 386 or higher.

Nice! Topbench is also pretty cool if you haven't checked it out. https://dosbenchmark.wordpress.com/

Stiletto wrote:

Nice! Topbench is also pretty cool if you haven't checked it out. https://dosbenchmark.wordpress.com/

Yeah, I have a bunch of Topbench numbers too. 😀 I just left them off because they tend to get really inconsistent with the faster cpu speeds and cache/turbo disabling. It was cool, though, that you could export their database into a modern spreadsheet.

Incidentally, I thought it was pretty cool in a coincidental sort of way that Wolf3D and Norton SI results were so consistent with each other from the 8088 all the way through the 386's. Wolf3D seems to scale accurately nto the 386 range, but beyond that it gets weird. 486s score better than Pentiums, for example. It must have something to do with how its coded (no Pentium optimizations back then?)

So it might make more sense to sort 386s and slower by Wolf3D, and sore 486s and higher by Doom.

clueless1: Excellent! 😀

But we must remember that 3Dbench (ver 2, cracked) do not operate below 286 due to counting errors. Need to use the original 3Dbench for XT, 286, Slow-386.

Кушайте на здоровье! 😀

Ekb wrote:

clueless1: Excellent! 😀

But we must remember that 3Dbench (ver 2, cracked) do not operate below 286 due to counting errors. Need to use the original 3Dbench for XT, 286, Slow-386.

Кушайте на здоровье! 😀

Yes, of course. Thank you!

So here is how those 3Dbench numbers on that chart would work. Someone disables caches on their Pentium II and gets a 3dbench 1.0 score of 1.5. they can then estimate that their computer is operating at approximately a 4Mhz 286 level. Of course, they would probably be running some of the other benchmarks as well to get a more accurate feel of what speed slowed-down machine is running.

In fact, I think those 286-25 scores on this chart are yours. 😊

Благодарим Вас за Вашу работу по сравнению с 286-25 386-25 . Мой русский ужасен ! Я вырос в Америке , так что я потерял большую часть языка после того, как я начал ходить в школу. Я использую Google Translate , чтобы помочь с этим .