Benchmarked my AMD 5x86-P75-133 over clocked at ( 4x40mhz = 160mhz ) computer.
64mb ram
540mb WD HD
256kb cache
Mobo - PCChips 918i ( ALI Chipset ).

Intel486dx33 wrote:

I want to benchmark my AMD 5x86 ( 4x40mhz = 160mhz ) computer.
What are some good benchmark tools for readings ?

speed test
dhrystone

Just curious as "Minimum requirements for Win32 port to run are minimum requirements for Windows 9x and SVGA card". 1st part there, at least
dosbox

Intel486dx33 wrote:

Benchmarked my AMD 5x86-P75-133 over clocked at ( 4x40mhz = 160mhz ) computer. 64mb ram 540mb WD HD 256kb cache Mobo - PCChips 9 […]
Show full quote

Benchmarked my AMD 5x86-P75-133 over clocked at ( 4x40mhz = 160mhz ) computer.
64mb ram
540mb WD HD
256kb cache
Mobo - PCChips 918i ( ALI Chipset ).

Speedtst benchmark = 561721
Dhrystone / seconds = 38558

Done.

More photos

Intel486dx33
You may be the 1st to enter the data about 486 speed of emulation in DOSBox. We need a hero. 😀
speedtest + doom in DOSBox - as described there
mb it's mighty to play XT era games. it's rather curious

I'm pretty sure I have some version of DOSBox (.72, I think) installed on my 486's in NT 4.0. Has it become a thing to benchmark games in DOSBox on a 486? Doom 'eh? I'll try to remember to do this when I pull it out next.

EDIT: Wait a minute. Your PPRO 200 gets 2.9 fps in DOOM 1.9? And we want to bench a 486 in DOOM/DOSBOX? You all have reached an all new level. I don't even know if we can call it a level.

feipoa wrote:

I'm pretty sure I have some version of DOSBox (.72, I think) installed on my 486's in NT 4.0. Has it become a thing to benchmark games in DOSBox on a 486?

Should be used 0.74 (not 0.74-2 or other)
You may use any OS. But NT may to have higher background services load, than Win9x where it may to work faster. Mb there are LiveCD based on Win9x or the possibility to install Win9x.

> Your PPRO 200 gets 2.9 fps in DOOM 1.9? And we want to bench a 486 in DOOM/DOSBOX? You all have reached an all new level. I don't even know if we can call it a level.

Pentium Pro allows up to 2.3 XT level. Fast 486 160 MHz has a chance to give ~1 XT level. With lower performance it would stay acceptable for non-action games, at least. The other testings there are interesting for CPU comparisions.

DonutKing wrote:

Edit2: Now this is interesting, apparently AMD made PGA Socket 3 chips well into the 90's!!
Think of the overclocking potential
http://www.cpu-world.com/CPUs/80486/MANUF-AMD.html

If I'm not mistaken the Am486DX2-66V16BGC about 4th up from the bottom has a datecode of 1998... I'm guessing for embedded/industrial applications because I don't know why you'd buy a 486 desktop in 1998 😜

Walland had complete AST DX2 Win95 systems for $500 very late, 97?
I recall a few AMD 5x86’s coming OEM with Windows 98

Odd times back then you could buy a new 486 or a new 900mhz system

Though I suppose my folks bought a cheap Tandy 1000rlx new in 1993 without a hard drive and Pentiums were on their way at that point.

Now days the difference from high end and low end is more marked by videocards than cpu

I am running win95c and it works okay. I am also using an 8gb. CF card. Worked fine with conventional hard-drive too.
I tried win98 too but it would barely run on this cpu.
Minimum requirements for Win98 is a Pentium class cpu.

Some number of years back, I compared benchmarks on a 486 when using Win98SE vs. Win95c and there was no discernible difference in the results.

I got a Pentium overdrive 83mhz cpu recently
Spec:
POD83 clocked at 83mhz
16mb ram
3Gb hard disk
256kb cache
Mobo - Msi 4144 socket3 ( SIS Chipset )
Windows 95 osr2.5

Results:

More results on Pentium overdrive 83mhz

Speedtst:484043

Dhrystones / second :24096

Other tests:

Tertz wrote:

Intel486dx33 You may be the 1st to enter the data about 486 speed of emulation in DOSBox. We need a hero. :) speedtest + doom in […]
Show full quote

Intel486dx33
You may be the 1st to enter the data about 486 speed of emulation in DOSBox. We need a hero. 😀
speedtest + doom in DOSBox - as described there
mb it's mighty to play XT era games. it's rather curious

Here is the DOOM game test results. What do you think is this score good ?

DOOM - 2134 gametics and 1894 realtics.
Speedsys - 561728 , 208.3x

Why do some boards have slower memory bandwidth but same memory throughput? 😕

It is doubtful that Speedsys's "memory bandwidth" is any reliable indicator of memory performance. Sometimes this value appears quite arbitrary.

Has anyone tested a Cyrix 486 DX2-80?
I'm interested to see where it stacks up against other 486 CPUs, namely the Cyrix DX2-66.
In my build, it seems slower than the DX2-66

I don't have the software or the time to do benchmarking myself.

No, sorry, I haven't tested the Cyrix DX2-80. I would suspect that a properly setup 486 with a 40 MHz FSB (that is cache, RAM, and ISA/PCI bus timings are optimised) should yield a Cyrix DX2-80 as fast or a tad faster than an Intel DX2-66-WT.

I have one dx2 80.. but not have mainboard to test it

I recently spent some time to compare the Asus PVI-486SP3 rev 1.22 vs the Biostar MB-8433UUD-A rev 2 with a Amd 5x86 133.

Asus PVI-486SP3 rev 1.20 BIOS release 0306, 256kb cache 15ns.
Some observations:
I was able to run (at 33mhz fsb) with the fastest settings with 16x2 fpm 70ns memory.
With 40mhz as fsb I wasn't able to do many tests, I suppose the problem is the pci frequency, as the cpu\mem benchmarks ran fine, despite of this to achieve the highest cache and memory settings I had to switch the memory to a single stick of 32mb rated 60ns.
The performance were the very same with 16x2mb or a single 32mb memory stick, so it's the same use 1 or both simm slots.
The motherboard can be picky with memory used, I tried a: 32MB PS/2 FPM SIMM RAM Double Sided 60ns 8Mx32 non-Parity Samsung KM44C4000BK-6 but it gets detected as 8mb, while it works fine with the Biostar mb.

Biostar MB-8433UUD-A rev 2 BIOS release 2014, 256kb cache 10ns.
I was able to run the mb without any issue at 33 and 40mhz at fastest memory and cache timings, perfect stability.
The memory which had troubles on the other mb works flawlessy on this one.
Some jumpers position is very uncomfortable, lucky I use it without a case.

Those are the differences with Speedsys:

Asus PVI-486SP3 33x4:

Asus PVI-486SP3 40x4:

Biostar MB-8433UUD-A 33x4:

Biostar MB-8433UUD-A 40x4:

Overall performance are equal, it seems that the sis has better ide controller, I personally like the Biostar more because it allows more fsb settings.

Galaxian wrote on 2019-07-28, 11:34:

Has anyone tested a Cyrix 486 DX2-80? I'm interested to see where it stacks up against other 486 CPUs, namely the Cyrix DX2-66. […]
Show full quote

Has anyone tested a Cyrix 486 DX2-80?
I'm interested to see where it stacks up against other 486 CPUs, namely the Cyrix DX2-66.
In my build, it seems slower than the DX2-66

I don't have the software or the time to do benchmarking myself.

I have these two that I could setup quickly:
Ti DX2-80
It's ST DX2-80

These are both Cyrix cores if I am not mistaken?