My wife has this hard rule of thumb that postulates: "If we bring one thing at home, another one has to go. We cannot be an old house."
... and she consistently purges the inventory.

The other day i spotted her heading for the trash container carrying two laptop boxes that we kept for a while in case the devices break during the first few weeks of the warranty period.
Their size and proportions caught my eye, so gently inquired if i can use them for something.
This resulted in a high-raised eyebrow ... but i got them.

For a while i had this idea of making a computer case out of cardboard. It was time to give it a try.

Here are the patients before plastic surgery. Some smudges on them that had to be cleaned later.

Put a new blade on the ExactoKnife and cut some openings:

3D printed couple of frames and other reinforcement elements for structural stability (and to hide the sloppy cutting job).
Repurposed designs from the 3D printed PC case from several posts ago.

Usually i assemble idealized versions of retro PCs, but wanted to put together an "average PC citizen" that represents the reality back in year 1993 the way most people experienced it.

Motherboard is of unknown brand, based on SiS461 chipset.
This is my very first retro hardware that i spotted in a local electronics market long ago.
When i saw it there, it gave me such a flashback to times long gone.
Bought it for $2 with the cpu, ram and cache as described below. 😁
It can do 25, 33, 50 MHz only.
Initially it was super slow - limited by rudimentary BIOS that lacks any options.
Few years later it was replaced with a BIOS from one of the ASUS motherboards based on the same chipset, that I had laying around.
This unclogged the pipes and the system started performing really well.
CPU is AMD 486 DX2 66MHz.
8Mb Samsung branded 70ns RAM.
256Kb 20ns SRAM, EtronTech.

Used combo VGA/EIDE adapter - CirrusLogic GD-5428, Promise DC-20230C.
Great fit for a "budget" system.

WD Caviar 2340 HDD - 341Mb, 2700RPM.

Sound card will be ... represented by the good old pc-speaker.

Test assembly. Everything seems to fit nicely.
Notice that i am not using any holder for the HDD.
It is kept in place quite well by the triple folded IDE ribbon cable and the power cable with tick wrapper.
Decided not to overdesign things, following the rule "don't allow perfect to get in the way of good enough".

Final assembly.
Had to swap the passive CPU cooler with active one, otherwise the processor overheats when the box is closed.
Also, had to reinforce the area around the power switch

And how things look on the outside.

The usual set of stats:

And some benchmarks (fps):
Wolf3D: 71.9
Superscape: 45.4
PC Player Benchmarks: 10.1
Doom (low/high resolution): 78 / 27.4

While i ran 3D rendering tests to check stability, with 8Mb of RAM only render times are meaningless - not sharing them here.
Processes are going into swap and scratching the hard drive all the time. Measurements are heavily skewed by disk i/o.

Great idea. Love it. Was thinking doing the same.

Thank you MPE.

Btw, interested in the vlb pc chips video card you got recently.
I skipped on one of these not far ago, but always wondered how they perform ...

DTK PKM-5031Y with Symphony SL82C461, SL82C362, SL82C405 chipset.

This motherboard caught my attention with its performance characteristics.
It came crispy clean. I can tell that it was never used - a new old stock. Plus a real looker.
There was AMD 486DX 40MHz CPU, 8Mb 70ns RAM and 256Kb 20ns SRAM.
Dallas battery was dead. Removed it, installed a socket.
Unusual DTK BIOS ...

https://www.petershipkov.com/temp/retro_pc_im … k_pkm-5031y.jpg

Board has jumpers for 25, 33, 40, 50, 66 MHz. I am not sure what they do given the fact that frequency is controlled by crystal oscillator.
Probably ISA bus divider and cache/memory wait states. Jumper config is the same for 25, 33 and 40 MHz. Couldn't test 50 and 66 MHz for obvious reasons.
System is very stable, with one exception - SpeedSys hangs on "Determining memory" or whatever the message is at startup time. No idea why.

The standard set of stats:

As i mentioned above, while playing with it i noticed that its performance is impressive.
At the same time somebody over the (discord) Ultimate Hardware 2019 channel wondered if it will be faster than SiS 460/461 which was available around the same time and was a solid performer. So decided to give that idea a try.
Started with Asus ISA-486SV2 (SiS460), but then switched to the unnamed SiS461 VLB board (with the same Asus BIOS) from my previous post.
While the DTK mobo is very stable at 80MHz (2x40MHz), to even things out and distil the relative performance difference between Symphony and SiS chipsets i relied on the next setup:
AMD 486DX2 66MHz CPU, 33MHz FSB, latest revision ISA ET4000AX for the DOS graphics tests and latest revision ISA ET4000/W32i for the WinTune2 test, standard ISA IDE controller with CF card.

The standard set of benchmarks. Didn't bother with LightWave3D rendering test.

Results are pretty clear - the DTK motherboard with Symphony chipset is quite ahead of what SiS can produce.
It will be interesting to check how it does against SiS 471 and 496, especially if there is a VLB motherboard instead of mere ISA one ...

pshipkov wrote on 2020-11-17, 01:53:

My wife has this hard rule of thumb that postulates: "If we bring one thing at home, another one has to go. We cannot be an old house."
... and she consistently purges the inventory.

I would get along with her then as I have the same rule. My wife is always buying stuff without the forethought of where the heck is it going to go. She has this notion of "we'll figure that out later", but she never does. And in the end, it is always me that has to figure it out or there would be clutter everywhere. The one-in, one-out policy helps keep the house de-cluttered, but as noted, it is always me doing the decluttering.

Hah. I will make an introduction then.
Welcome to the club. 😉

pshipkov wrote on 2020-11-24, 08:25:

DTK PKM-5031Y with Symphony SL82C461, SL82C362, SL82C405 chipset. […]
Show full quote

DTK PKM-5031Y with Symphony SL82C461, SL82C362, SL82C405 chipset.

This motherboard caught my attention with its performance characteristics.
It came crispy clean. I can tell that it was never used - a new old stock. Plus a real looker.
There was AMD 486DX 40MHz CPU, 8Mb 70ns RAM and 256Kb 20ns SRAM.
Dallas battery was dead. Removed it, installed a socket.
Unusual DTK BIOS ...

PKM-5031Y is actually a PKM-0031Y that comes with a 50MHz CPU.
This board came with my DTK 486 in the early 90s. I think considering it doesn't have local bus slots, it's a pretty decent performer for what it is. I like mine and I still have it. I recommend you upgrade to MR-BIOS, it will make it even better (works great with the Cyrix 5x86).

It's really a shame that Symphony disappeared before they were able to make PCI boards.
However, they did make a 386/486 EISA chipset, but it never shipped. Symphony was purchased by VIA, and I am somewhat certain that the EISA boards with VIA chipsets are actually rebadged Symphonies. If you ever happen to see one I recommend checking it out. If it lives up to the other Symphony products it should be pretty slick.

Yep, the two boards are the same really, just different sockets.
I think there is one more earlier variation of the 386 mobo out there as well.
Agreed, too bad Symphony didn't get to VLB. It would be killer.
Didnt know about them and VIA.

Do you have mr.bios version for the 486 board ?
I don't love Monsignor Bios, but will take it any time over the DTK nonsense 😀

DTK had some nice hardware, but I agree that their BIOSes weren't the best. Does your PKM 5031Y had a DTK BIOS? All the ones I've seen are AMIBIOS.

Symphony actually did make a VLB chipset. DTK did make one, but I've never seen one so they must be pretty rare. Most of the ones I do see were made by Young Micro systems. I once had one, but could never get it to POST so I tossed it and have no idea if it's good or not. They support 1024kb cache though, so that's cool.

PK-X486/87SD - CPU/FPU upgrade module for 286 class computers.

Most such modules tend to be finnicky, so i pulled 5 different 286 motherboards to see which ones will "fit" best.
The "natural selection" process was quick - 3 of them worked out - 2 VLSI ones and a slow Headland.
The Headland was discarded as insufficient for best perf, so VLSI for the win.

The upgrade module had soldered on it 24MHz Texas Instruments 486DLC and 33MHz Cyrix 87DLC FPU.
Apparently it was designed for base frequency of ~12MHz, where after clock doubling its CPU and FPU tick at 24MHz.
Not a bad solution for slow 286 systems, but it was actually much slower than my 25MHz 286 PC rig.
What is needed here is not DLC, but SXL2 CPU.

After 2 hours of work the original CPU was gone and a socket stolen from dead 386 mobo took its place.
Later on the FPU had to be swapped as well, but about that below in the post.
This is how the upgrade module ended up looking:

The actual testing started with this motherboard.
Its layout is slightly different than the one in my 286 computer, but their performance is the same.
https://www.petershipkov.com/temp/retro_pc_im … rd_286_vlsi.jpg

50MHz rated TI 486SXL2 processor with clock doubling was inserted in the newly soldered socket.
It can withstand 60MHz - no problem.
So the base frequency of the system has to be 60/2 = 30; 30/2 = ~15MHz.
Tried 70MHz crystal oscillator, but the monitor didn't light up.
A 66MHz worked fine.

Next up - "cyrix.exe -i1 -r" - instant hang ...
After some time i realized that if the FPU on the upgrade module is enabled via the available switches the above command locks the system.
At the same time FPU was never recognized to start with. Actually i saw it showing up only once and never again.
Decided to swap with a spare 40Mhz rated one.
This was a lot harder than i expected.
The actual swapping of FPU chips was quick - few minutes of heatgun action, but after that the module stopped working.
3 hours later and numerous tweaks around the FPU pins and solder pads underneath - still no go.
At the end i pulled my daughter's microscope, quickly discovered and fixed the problems.
That microscope will be seeing more electronics action in the future for sure.

Running the cyrix.exe command in combination with enabled FPU didn't lock anymore, but the FPU was still unavailable.
Tested with multiple other mobos - same.
I see 3 possibilities:
- this module was designed for specific list/set of 286 motherboards, which i don't have
- there is some specific formula required - a combination of hardware, software, jumpers, switches - unlikely - tried pretty hard
- there is something else damaged on the module
Decided to leave this problem for another day, because i noticed something even more unpleasant:
- write access to CompactFlash cards is flaky with the SXL2 CPU, works fine with mechanical HDDs.
- adding more extension cards such as SoundBlaster, LAN, etc. makes the system unstable.
My general interest in upgrade modules is low and these issues lowered it even further - this module+cpu cannot be used in a fully functional computer, at least not for top performance ...
But still, it was interesting to see how far i can go with this 2/4x86 machine ...

Along the way i noticed some small difference in behavior between the two VLSI moderboards.
The one i started with didn't like the 70MHz crystal oscillator, but when i tested later with my 286 rig, 70MHz was just fine, which is the limit imposed by the cpu.

Will update these with proper screenshots in the coming day or two.

Compared with my 286 and 386 rigs:

2 videos showing test runs:
lm6, superscape, wofl3d: https://www.petershipkov.com/temp/retro_pc_im … x486_87sd_1.mp4
doom: https://www.petershipkov.com/temp/retro_pc_im … x486_87sd_2.mp4

Conclusion:
All in all - an interesting hardware.
CPU intensive tasks benefit quite a bit from it.
At the same time IDE performance suffers from low base frequency.

Apart from the not working FPU, the problems are mostly related to the SXL2 CPU really.
Similar CF write access issues are present in 386 class hardware in combination with SXL2 and SCSI, where the processor is drop-in upgrade.

About the FPU - will try it on few more "classic" 286 motherboards with the hope that the upgrade module was tested on them during its development, to see if something will make it "click".

I am not able to see your photos at the moment (they're blocked), but my experience with the Buffalo brand 286-->486 modules has been pretty positive. Though, admittedly I haven't tested them with too many boards. Mine worked perfectly on the IBM Type 1/2/3/4(XT286) planars, as well as the PCChips M209.

However, I recently got an IODATA 286 to IBM486SLC2 (50MHz), and I haven't yet been able to make it stable. So perhaps you might have better luck with the Buffalo modules...but I guess the prices aren't so great these days, eh? Not sure what they go for on yahoo.co.jp at the moment, but one on eBay sold for $700 not too long ago.

It's been years since I benchmarked my 8MHz AT with the SXL40 running at 48MHz (6x8MHz), but I think my results were similar to yours. The disk and memory was slow, but CPU/FPU was fast. It made Windows 3.x a hell of a lot faster, and I was even able to play Sim City 2000 at sub optimal performance.

pshipkov wrote on 2020-11-17, 01:53:

My wife has this hard rule of thumb that postulates: "If we bring one thing at home, another one has to go. We cannot be an old […]
Show full quote

My wife has this hard rule of thumb that postulates: "If we bring one thing at home, another one has to go. We cannot be an old house."
... and she consistently purges the inventory.

The other day i spotted her heading for the trash container carrying two laptop boxes that we kept for a while in case the devices break during the first few weeks of the warranty period.
Their size and proportions caught my eye, so gently inquired if i can use them for something.
This resulted in a high-raised eyebrow ... but i got them.

Looks very slick... love how the 3d printed parts go so well with the rest of it

@drosse1meyer
Thanks.

@Anonymous Coward
I reworded a bit the conclusion at the end of my previous post.
The problems are mostly related to the SXL2 CPU than the upgrade module itself.
In general, many of these TI/Cyrix upgrade processors tend to cause issues with IDE/SCSI + CF cards.

$700 for upgrade module ?!
Way too much for too little.

There is not much info online about this Chicony TK8880F / 2066A2 motherboard from year 1994, so decided to share some findings.

It is an interesting one - very well preserved, i can tell it saw little use in its time - no scratches, no dust, no wear.
Supports up to 128Mb RAM and 1024Kb of L2 cache.
There are 72 and 32 pin memory slots. Used 72 pin modules only in my tests, but it looks like both types can be mixed.
Clock generator can go up to 60MHz. BIOS is full of options, including a bus divider.
There is no on-board IDE controller which is not a big deal.
AMD DX5 @160MHz CPU is incorrectly reported during post as DX4 @120, but all is fine with it really.

Came with the usual dead Dallas RTC module.
Had to de-solder and socket it.
Used a new approach for the socket - utilized female pins from DIP4 crystal oscillator sockets. 😀

UM8881# chipset is versatile = many jumpers.
Their location is a bit disorganized on this board, which makes the configuration process harder than it should be.

https://www.petershipkov.com/temp/retro_pc_im … ony_tk8880f.jpg

EDIT: Manual has JP42 in position 1-2 for 1024Kb L2 cache. This is incorrect. It has to be 2-3.
Picture was taken before getting the board up and running properly, so that jumper is shown in the wrong position.

The system is stable with all BIOS settings set to their optimum (best performance) values. Same applies for the jumpers configuration.
One exception for PCI video adapters below.

VLB support is less than ideal.
Few VLB VGA adapters didn't lit up, including Ark1000VL which i prefer for DOS PCs.
Diamond Stealth 64 DRAM (S3 Trio64) worked fine. It is basically on par with Ark1000VL in DOS, but much better at accelerated Windows graphics.
Same goes for VLB EIDE controllers - few couldn't boot into DOS. From the ones that worked SIS 83611 was the best. Used it.

PCI support was spotless.
When using PCI video adapters IBC DEVSEL# DECODING needs to be set to MEDIUM.

benchmark results

As can be seen in the charts - not a bad performer at all. Maybe not the best out there, but certainly pretty good.
A bit of a hiccup with VLB perf in Windows, but pretty much on par with Asus PVI in DOS graphics otherwise.
One thing to notice: In DOS graphics Asus PVI does better with VLB video adapters, while Chicony prefers PCI ones.

Red flag:
In my experience 3D rendering is very sensitive to system instabilities compared to any synthetic benchmark out there.
For example, this board can cycle for hours in Check-It memory test, but fails frequently at (LightWave3D) rendering.
Had to significantly constrain the BIOS options to get through successful rendering test.
This basically halved PC's performance. Didn't include the result - it was embarrassing. 😀

CONCLUSION:
Pretty good all round piece of retro hardware.
Favors "wide" over "deep". Meaning:
1. Supports broad range of hardware architectures (pci + vlb, 32/72 pin memory, 1Mb L2 cache, voltages, CPUs, etc).
2. While overall performance is pretty good, there are better options. Some system instabilities show-up when pushed.

Retro computing means different things to different people.
To me the "retro era" ends around year 1996 (give or take a few) with the demise of the 486 platform.
Yet, i have a soft spot for Slot 1/2 CPUs and related hardware that were the thing around year 2000 (again - give or take a few).
So not sure if this passes for "retro", but here ...

Melodramatic insert:
Around the turning of the 21st century i was all over 3D graphics and getting deeper into coding.
I associate the period with growing knowledge - learning new concepts from different apps and trying to replicate them in pixels and/or code.
So, for long time i wanted to build a kick-ass year 2001 (give or take 1) dual CPU workstation for computer graphics and software development, with distant second role - gaming.
At the time video gaming was mostly behind me and only Quake3 could distract me for a short while from my other interests.

With that said, i finally found the time to almost finish this project. Currently at the last step before completing it.

The CG scene in year 2001, for context, briefly:
All high-end graphics applications were ported already to the PC platform and steadily evolving.
Alias|WAVEFRONT's Maya established itself as the premiere package for computer graphics.
Softimage3D was substituted by XSI, which was rebuilt from scratch and needed few more years to smoothen the edges.
More people started grasping the power of proceduralism in SESI's Houdini.
Cinema4D already captured its market niche.
3DS MAX was very popular already, but similar to XSI and Cinema4D - its computation engine was stack based and this was a limiting factor.
LightWave3D was in decline.
Global Illumination renderers were replacing the scanline ones and so on.
Anyway.

Back to present day.
As usual, period correctnes is the norm, this time wider range - mid 2000 to mid 2002.

MOTHERBOARDS, CPUs, MEMORY:

I like all dual Slot 1/2 standard ATX motherboards with AGP slots, but not all of them are good fit for my need here, so most boards were discarded in the pre-qualifications as unsuitable (multiple Supermicro assmeblies, Soltek SL-68A, Gigabyte GA-6BXDU, Tekram P6B40D-A5, various abominations from HP, Dell, Compaq).
3 made it to the final bakeoff:
Supermicro S2DGE
Tyan Tiger-133
Asus P2B-D
Will probably go over some of the contenders at some point later, such as OR820/840 - some interesting findings there worth bringing up and comparing with the 3 above.

As you probably noticed i picked non-SCSI versions of the Supermicro and Asus boards.
I use CF cards for local storage and SCSI-to-IDE with its cables and adapter is way too messy inside the case.
Also, the plan was to go with a modern day (for the time) UltraATA IDE RAID controller which will be much faster than the built-in SCSI.
In short - i like my computers dedicated and optimized for the mission - like an F1 car. Cannot have unused baggage dangling in the trunk.
Or at least, that was the theory ...
Unfortunately it turned out that RAID IDE + CompactFlash cards + NT4 = no go. Works fine with mechanical HDDs.
Luckily normal IDE controllers, such as Promise Ultra133 TX2, are just fine with CF cards. Speedsys screenshots below for the Asus and Tyan motherboards have been taken with that very Promise adapter instead of showing the performance of the on-board ATA-33 or 66 ones.
To continue at the end of the post.

Initially i wanted a "clean" P3 Coppermine slot 1 processors, but then decided to max things out with slotketed Tualatins.

Supermicro S2DGE

Dual Slot 2 motherboard based on Intel's i440GX chipset.
Latest revision, latest BIOS.
2Gb PC100 ECC RAM, Micron.
Two 900MHz Pentium III Xeon processors.
The on-board clock generator maxes out at 100MHz, so no chance of overclocking.
Initially the board operated at 83MHz FSB, which resulted in the CPUs running at 740MHz. This was odd.
Tried everything i can think of to fix that but without success.
Even called Supermicro's support, but while nice on the line, they kind of told me to buzz-off since the ancient relic does not officially support 900MHz Xeons, so not their problem.
Then something unexpected happen. Check this out:
Lots of fans are needed to keep the CPUs cool. One day i was setting them up and was measuring with multimeter some of their power plug pins. One of the probes slipped and i shortened the line by accident. The fans on it stopped running. Lame.
Decided to measure the voltage on another fan power plug as a starting point for fixing things. Slipped again. Shortened it too. Two more fans stopped. Bummer.
2 hours later i was still unable to locate the broken components, so decided to do the easy thing - provided voltage to the pins using wires soldered on the back of the board and secured with molded plastic.
Turned on the board for a quick test and surprise surprise - 900MHz CPUs @ 100MHz FSB.
What The Heck ... 😁
But i take it.
No complains.

Motherboard processors, system memory, basic stats.
This thing is massive !

Tyan Tiger-133 v1.05

VIA Apollo Pro 133a chipset.
I don't think i ever owned motherboard with VIA chipset. Pretty sure this is my first ... and i really like it. Especially that i kept it in its original box and wrapping. It still smells like new hardware - that is 20 years after its birth.
Supports up to 2Gb RAM at 100MHz FSB, or 1.5Gb at 133MHz. Takes single sided memory modules.
Everything works as expected, except the on-board ATA-66 controller. Even it says that ATA-66 device is recognized (IDE to CF adapter + the fastest CF card out there), it maxes out at around 8Mb/s. Works fine with mechanical HDDs. Not a big deal, since CF cards with their low latency >> than mechanical HDDs from that time. Still annoying enough, if not using a fast third party IDE adapter.
According to the schematics of the on-board clock generator, it cannot go above 133MHz, so no luck with overclocking.

Motherboard with Pentium III 1000/256/133 CPUs and 1Gb of PC133 Mosel Vitelic RAM, basic stats:

Asus P2B-D

Revision 1.06, stepping D03, ICS 9250CF-08 clock generator, HIP6019BCB voltage regulator.

I have this guy completely pimped:
- Adjustable VTT Tualatin mod for reduced signal voltage to exactly 1.363V - upper left wires on the picture below. The rubber wrap contains two resistors that add up to the desired Ohms. It was easier to combine 2 elements than searching for a single one with the right resistance + error.
- VIO mod for 3.5V to memory for increased stability with maxed-out RAM and 133MHz or higher FSB.
- 4th jumper mod for enabling the entire table of frequencies supported by the clock generator - upper right wires.
- ACPI fix
- Latest BIOS, patched with Tualatin microcodes.

Intel and Asus say that i440BX chipset maxes out at 768Mb RAM @ 133MHz FSB.
In terms of maximum memory, the board is running completely stable in the next configurations:
1Gb RAM @133/44.33MHz
768Mb @ 140/35MHz
384Mb @150/37.5MHz (all BIOS settings on MAX except SRAM MA WAIT STATE = NORMAL, best is FAST)
256Mb @150/37.5MHz (all BIOS settings on MAX)
Took a knee-tall stack of memory modules to find the right ones.

The onboard ATA-33 controller maxes-out at around 10Mb/s. Not very impressive. Promise Ultra133 took care of business.

Motherboard, two Pentium III Tualatin 1400/512/133 CPUs (SL6BY) in Asus S370-DL rev. 1.02 slotkets, 1Gb RAM - Elpida PC133.
Cooler Master heatsinks. Compact, but surprisingly dense and heavy. The CPU modules feel like golden bricks in the hand.
Fans are like mini-helicopters, but reasonably quiet.

1470 / 140

1575 / 150

AUDIO + LAN:

As mentioned above - i am going for an F1-like system.
The only allowed luxury is a sound card, such as this Creative Labs Sound Blaster Audigy EAX:

As well as a standard LAN adapter:

GRAPHICS ADAPTERS:

Back in year 2001 the situation with professional graphics adapters was a bit blurry.
So, decided to check the latest AGP ones, that are compatible with the mobos:

A non-period correct fancy SDI version of Quadro FX4000 (the PRO equivalent of GeForce 6800 Ultra) was only used as an indicator for the graphics performance limit, considering what is possible with the available AGP slot version on the motherboards (card's AGP connector is modified with an extra notch to fit the slot).

Wildcat II 5000

Quadro 2 Pro (ELSA Gloria III) - overclocked to 350/550

Oxygen GVX1

Oxygen GVX1 Pro

Oxygen GVX210 + box.

Fire GL2 (Diamond). Added a fan. No idea why these cards were sold with passive cooling only. I can fry bacon on it when under load.

OPERATING SYSTEM:

Windows NT4 + SP6a (128-bit) + DX5 + IE6 (excluding its ActiveDesktop - no CPU cycles can be allocated to unsubstantial stuff)

I really liked Windows NT4. It did right everything i cared about:
- ~90Mb size on disk out of the box, ~250Mb with VC++ 6
- small memory footprint - 12Mb out of the box, ~20Mb with VC++
- availability of good video drivers with solid OpenGL support
- rock solid, simple to operate and snappy to use
- completely portable - install and setup everything properly. Boot into DOS/Win9x/Linux/etc. - copy the files to a backup location. The restororation to fully operational NT4 state is a matter of simply copying them back. Assuming boot sector/loader are intact - the OS will just work, including on a different hardware. In this case it may ask for some drivers and stuff, but that's it. This is something i relied a lot on back in the day.

I resisted upgrading to Windows 2000, XP and such for as long as i could. They pushed me into the Linux world where i stayed until recently, but i admit Windows 10 is good and I am kind of back in MS land.

BENCHMARKS:

The systems were tested in-depth, but most of the details will be out of context here. Will keep things on a high-level, short and sweet.

For professional use 3 things matter:

1. C/C++ compilation times.
Used PovRay 3.5 source code. It takes about 2-3 minutes to compile, which is ideal for testing.
While the code is pretty much completely self-contained, i had to patch it in few places to work around a dependency on some MS library that i couldn't obtain. Compile times may be skewed a bit because of that.

2. OpenGL viewport performance.
Used a tank asset that i have. It has dense enough (for the time period of interest) mesh geometry.
vertices: 120,395, edges: 236262, faces: 122,698

3. 3D rendering performance.
Used two scenes:
For scanline rendering i recreated a shot from StarWars Ep.4. Have to say that it felt great making pixels again. I had a star destroyer model laying around. Took me 2-3 hours to get the shot completed. One thing to mention - all textures are render time procedurals, no bitmaps.
The original shot has this blue tint on the destroyer that always bugged me, corrected it in my interpretation. 😉

For global illumination rendering (which was getting very popular at the time) i used the tank scene.
Interior environment, one ambient (fill) light, entire lighting comes from emissive surfaces. Worst case scenario in many ways.

Added and the previously used LightWave3D test - it was interesting to see the performance improvement compared to my AMD 486DX5 @160MHz system from page 1:

4. And one gaming test - Quake3's default timedemo. 1024x768, all settings on max.

RESULTS:

Systems configuration, for clarity:

Supermicro S2DGE:
900MHz CPUs
100MHz FSB
33MHz PCI
2Gb RAM

Tyan Tiger-133:
1400MHz CPUs
133MHz FSB
33MHz PCI
1Gb RAM

Asus P2B-D:
1470MHz CPUs
140MHz FSB
35MHz PCI
768Mb RAM

Asus P2B-D, as a peak performance reference (gray bars):
1575MHz CPUs
150MHz FSB
37.5MHz PCI
384Mb RAM (all BIOS settings on max except SRAM MA WAIT STATE = NORMAL, used for the MentalRay offline rendering test)
256Mb RAM (all BIOS settings on max, used for all other tests of this configuration)

Code compilation. CPU frequency is everything that matters.

OpenGL. The important professional test is "shaded" mode. Empty cells indicate system instability.
Notice Quadro's best Quake3 number. 😁
i440BX + high frequencies = the deal, but i440GX + Xeons are "dangerously" close.

3D rendering. High frequency is everything that matters.
Another set of tests not shown here confirm that VIA's chipset in Tyan Tiger is clock-to-clock slower than i440BX. It kind of can be seen here as well.

Added and some DOS tests with the Quadro2 Pro, as it is the fastest from the pack in DOS. Again, wanted to see black on white the perf improvement over the 486 class hardware from previous tests.
i440BX + highest clock frequency for the win.

RAID:

As i mentioned at the beginning of this post, the plan was to use top-dog period correct IDE RAID controllers - for compactness of the physical setup and for practicality, considering CompactFlash cards.
This failed.

RAID controllers work just fine in DOS or Windows 9x, etc., but hardware cannot be recognized by the corresponding drivers when in NT4.
This is unexpected and the reason is software. Bunch of different IDE RAID controllers with CF cards cannot be recognized on bunch of different motherboards, but only under NT4. 😖

Some of the tested controllers:

3Ware Escalade 7000 - great controller - gets me to ATA-100 (in DOS)

FastTrak Tx2000 - pretty good too. Sees the CF device as ATA-66 and maxes out at 24Mb/s (in DOS).

SIL0680 - another good one - maxes out at around 21-24Mb/s (in DOS).

Instead, this Promise Ultra133 TX2 IDE controller was used for the Asus and Tyan motherboards:

FINAL WORDS:

The goal is to build the most optimized, best performing Slot 1/2 based workstation.

Looking at the results, pros/cons, etc., it is clear that Asus P2B is the top performance dog, but is RAM limited and with on-board ATA-33 IDE only.
Supermicro's S2DGE is slower than P2B, but supports 2Gb RAM and does much better with its ATA-66 IDE.
Using third party ATA-133 IDE controller levels out the field and on-board IDEs are not a factor.
Tyan Tiger-133, despite being a really great motherboard on its own, is somewhat in the shadow of the other two.
So it looks like it will be the P2B or S2DGE that will end up in the computer case.

About the video adapters:
Back in the day OpenGL graphics were a lot more glitchy than we know them today. I am using the latest available driver for each card. Apparently they worked out most of the issues. At least i didn't encounter any. So, deciding factor is the performance alone.
Looking at the numbers - it will be Fire GL2 or Quadro2 Pro.
Lit+texture test is unbearably slow for all adapters which makes it kind of irrelevant - this mode is practically unusable with this class hardware.
Quadro2 kills it in Quake3 and DOS graphics, but is slightly slower in the OpenGL "shaded" test than FGL2.
Both Q2Pro and FGL2 are on par in the "wireframe" test.

In addition - Quadro FX4000 (GeForce 6800 Ultra) is not period correct, but seems to be the last generation card that works on these motherboards and is supposed to define the practical performance limit - so, i ran tests with it as well. Results were similar to Quadro2 Pro and Fire GL2, which indicates that performance is CPU bound - host cannot saturate the GPUs. Didn't expect that, but the outcome makes sense. Anyway.

So, as usual, there is not a single ideal option from the considered components, but at least tinned the candidates down to 2 motherboards and 2 video cards.

-------------------

Rather boring video showing a test run with the Asus P2B-D board - Maya, Softimage, VC++, Quake3: https://www.petershipkov.com/temp/retro_pc_im … est_run_140.mp4

As i mentioned early, the board runs completely stable at 1575MHz CPUs, 150MHz FSB, 37.5MHz PCI, 256Mb RAM only, 2-2-2 wait states. In this configuration we get a kick-ass gaming PC.
Here is a quick Quake3 roll with the FX4000 GPU:
https://www.petershipkov.com/temp/retro_pc_im … est_run_150.mp4
262 fps <- this !
EDIT: Initially this number was 258 fps, but performance improved after some changes explained here.

pshipkov wrote on 2020-07-19, 06:28:

When not using the software driver for the SIS controller, both buffered and linear read speeds are at ~4100 Kb/s. When use the driver, the buffered read goes through the roof, but the liner speed increased only by 700Kb/s = ~4800 Kb/s.

Can you share that sis83C411 driver? I haven't been able to find anything. SIS doesn't have it on their site. I'd really appreciate it, I have one on the way.

Sure, np.

Add
LH=C:\DRV\SIS_EIDE.COM /1
to autoexec.bat

I looked and purchased set of two 3ware 2 pata raid card just exactly like yours. This has 2 chips that is memory cache chips meant well designed feature.

Thanks for trying them out for us!

Cheers,

3ware controllers are great. For some reason they are not very popular around here, which results in great price on eBay and thats fine by me 😉

Updated the Pentium3 motherboards post above with additional info/images about:
- using Promise Ultra133 TX2 IDE controller for the Asus and Tyan motherboards instead of their on-board ones (no-go RAID situation)
- AGP modded GeForce 6800 Ultra (represented by Quadro FX4000) as an indicator for graphics performance limit
- speedsys screenshot for Asus P2B-D running at 1575/150

---------------------------------------------

Suntac 286H motherboard.

A value version of sort - without integrated IDE and FDD controller.
Came with 16MHz soldered processor which was replaced by 25MHz one. Socketed the crystal oscillator.
BIOS provides set of interesting options, such as synchronous or asynchronous bus. Best performance is achieved with the synchronous option, obviously.
Maxes out at 40MHz, past that things get unstable.
Problems with using CompactFlash cards in general, regardless of overclock or not.

The usual set of stats and benchmarks. It was easier to snap photos of the screen than plug a mechanical HDD and transfer the screenshots over.

benchmark results

Typical Suntac board - slow and clunky, but with a charm of its own, considering the option for integrated IDE/FDD controller, which was unusual on this class hardware.