The BL3 stalled midsummer as I couldn't get hold of that one other guy on the whole internet who has one, abandoned his former socials and I haven't set myself up on Mastadon yet. Was hoping for clues about the IDE. Nervous about trying trial and error with the jumpers, because the ones I have found references for are every which way, horizontal and vertical, so apart from making for way too many permutations, not sure if damage possible. Also rabbitholed on the Zenith box with cyrix dlc in it which I was gonna run comparison. That turned out more broke than I want to spend time on. So that was all on hold anyway with homestead repairs getting priority and christmas and family stuff, where I am just now getting a bit of retro time. It was all set up on wobbly folding tables and wasn't great to work on like that, so put it away temporarily and got to setting up a bigger bench. Just done that, then promptly buried the bench in other stuff today, whoops... yeah it's intended to be high on the list to meddle with again soon.

pshipkov wrote on 2023-02-20, 22:28:

I won't be surprised if am the Young Micro Systems most avid end user to date with many of their 386/486 motherboard models coll […]
Show full quote

I won't be surprised if am the Young Micro Systems most avid end user to date with many of their 386/486 motherboard models collecting dust nearby.
The interest is entirely based on my curiosity about Symphony Labs silicon which YMS used in their products. Much less about the company YMS itself.
Unfortunately none of their assemblies turned-out exciting so far. They don't respond well to tight BIOS timings and overclocking.
Until now. Well, mostly.

Young Micro Systems VS486G-3VL (VEGA 486 Green PC)
Based on Symphony Labs Wagner (SL82C491, SL82C492).

It is the most feature complete 486 VLB motherboard from the YMS series.
There is even a voltage regulator on board. Crazy ! : )

In crisp condition.

--- Am5x86 at 160MHz (4x40)

All BIOS settings on max, except CACHE BURST READ CYCLES 2:4 = X-2-2-2 (best is X-1-1-1)
Motherboard is not picky about RAM or level 2 cache chips.
Ark1000VL video cards works well with tight wait states (noting it since this is not always the case).
There is not much more to say really. Things work as expected, except SpeedSys refusing to cooperate and Quake 1 test showing some funny result.

Looks like skewed game clock at play. Otherwise the test completes just fine and the game is playable entirely.

Coretest reports 12207 Kb/sec with Promise EIDE 2300Plus controller (turbo mode + speed 8 ) + CF card.

benchmark results

One sentence conclusion - above average performer.

--- Am5x86 at 200MHz (4x50)

There are 6 jumpers for setting up base frequency, but no configuration results in 60/66MHz. Going with 4x50.

Had to slow things down for complete stability.
CACHE SHORT WRITE CYCLE = NORMAL (best is SHORT)
BURST WRITE CYCLES 2:4 = X-2-2-2 (best is X-1-1-1)
BURST READ CYCLES 2:4 = X-2-2-2 (best is X-1-1-1)
BURST WRITE CYCLE 1 = 3-X-X-X (best is 2-X-X-X)
BURST READ CYCLE 1 = 3-X-X-X (best is 2-X-X-X)
AT BUST CLOCK SELECT = CLKIN/4 (best is /3)

Won't do a dedicated chart as usual since i will be consolidating all 200MHz results into one place.
Posting in textual form temporarily:
Wolf3D: 146.1 fps (Ark1000VL)
PC Player: 30.7 fps (Ark1000VL)
Doom: 60 fps (S3 Trio64, unstable with Ark1000VL)
Quake 1: skewed clock
WinTune 2: 13137 KPixels/sec (S3 Trio64)
3D Studio R3: 113 seconds (lower is better)
LightWave3D: 1089 seconds (18:09, lower is better)

Coretest reports 13523 Kb/sec with Promise EIDE 2300Plus controller (turbo mode + speed 8 ) + CF card.

Not bad at all really.

--- Intel Overdrive P24T at 100MHz (2.5x40), POD100

System is inherently unstable with the modified P24T processor.
Tried hard to improve on that but without success.

benchmark results

---

Despite the POD100 issues, this is the best Symphony Wagner based VLB motherboard i have seen so far.
No hustle, overclocks pretty well, supports 3.45V CPUs.

I think I got the same board now.

A bit dirty, but otherwise in a very good condition.
It allows for 1MB L2 cache, the Dallas chip is socketed, there is even a mouse header (but I am a bit confused that it is 10pin).
Not tested yet, but I will do it soon.

EDIT: I will make a shoot-out against the SV2GX4. I guess the SV2GX4 will win, I really like the 159 in Wolf3D and the 66.7 in Doom 😁. On the other hand this board seems to support 128MB, which is somehow nerdy and I like it. We will see.

Any indications it was in a Dell? First half of 486es Dell seemed to like bus mice and some of the boards from them had onboard busmouse header.

BitWrangler wrote on 2024-02-27, 18:17:

Any indications it was in a Dell? First half of 486es Dell seemed to like bus mice and some of the boards from them had onboard busmouse header.

I have no clue about the origin.
So 10 pin is a bus mouse header? I guess will stick to a serial mouse then ....

It's a possibility to bear in mind, can't say for sure.

@CoffeeOne
Cool. My favorite Symphony Haydn 486 board.
Looking forward to see what you derive from yours.
Will be interesting if it matches my early findings.

These are too rare to be Dell related.
I cannot remember how big Dell was at the time - let's say 1995.
Inclined to think these boards are some industrial equipment, but who knows.

pshipkov wrote on 2024-02-27, 21:37:

@CoffeeOne Cool. My favorite Symphony Haydn 486 board. Looking forward to see what you derive from yours. Will be interesting if […]
Show full quote

@CoffeeOne
Cool. My favorite Symphony Haydn 486 board.
Looking forward to see what you derive from yours.
Will be interesting if it matches my early findings.

These are too rare to be Dell related.
I cannot remember how big Dell was at the time - let's say 1995.
Inclined to think these boards are some industrial equipment, but who knows.

Sorry for nitpicking, but it is Wagner.

confusingly https://theretroweb.com/motherboards/s/young- … ga-486-green-pc doesnt acknowledge 10 pin header and claims its a ps2

The Bios string of my Wagner board is 40-1800-006259-00111111-121593-SYM WAG-H. Retroweb lists one with the same number but 2800 instead of 1800. Is it relevant?

EDIT: So the board works in principle, of course the Dallas is empty. I ordered the Necroware thing and will continue testing when I got it.

MORE EDIT: This 1800 seems to be the version number. So I probably should upgrade to the 2800.

486 class motherboards with AMD Am5x86 processors running at 180 and 200 MHz

Numerous motherboards have undergone evaluation.
The charts present a select few from a larger list that demonstrate relative stability at 180/200MHz.
The gray bars represent those that haven't achieved full stability, having failed at least one test.
The blue bars denote motherboards that are completely stable. In other words - these are the most interesting ones.
Deep blue bars indicate peak performance of a fully stable system, but pushed to a maximum overclock that may compromise stability.

This data effectively highlights the top-performing motherboards in the 486 class.

(It looks like i lost data for Shuttle HOT-433 rev 4 and Pc-Chips M918i at 200MHz, will re-run and update later)

------------ 180MHz (3x60) ------------

------------
------------ 200MHz (4x50, 3x66) ------------
------------

Awesome, thanks for sharing your careful compiled results once again.

Damn, I was excited about finding a Gigabyte GA486AM/S board the other day, guess I just gotta throw it in the trash now 🤣 /sarcasm

I am happy that VL/I-486SV2GX4 @160MHz + Ark 1000 is still ahead in Wolf 3D, beating the fastest 200MHz machine. OK, it is only Wolf3D everywhere else clearly slower 😀

@BitWrangler
Good sarcasm.
How you throw the trash we play with to the trash.
But more seriously - these Gigabyte GA486AM/S mobos look powerful on picture, but are actually not very good.
At 160MHz (4x40) they are in the upper echelon, but takes effort to get them there.
180MHz is possible, but slow and flaky.
200MHz is just not happening.

@CoffeeOne
Asus VLI + Ark1000VL is the VGA blaster, but i think you are talking about something i am not aware, or forgot about.
What is your top Wolf3D score ? The best i know is 153.3 fps. Did you get better ? Fully stable ?
Also, what are your best scores in the other tests ?

pshipkov wrote on 2024-03-04, 08:02:

@BitWrangler Good sarcasm. How you throw the trash we play with to the trash. But more seriously - these Gigabyte GA486AM/S mobo […]
Show full quote

@BitWrangler
Good sarcasm.
How you throw the trash we play with to the trash.
But more seriously - these Gigabyte GA486AM/S mobos look powerful on picture, but are actually not very good.
At 160MHz (4x40) they are in the upper echelon, but takes effort to get them there.
180MHz is possible, but slow and flaky.
200MHz is just not happening.

@CoffeeOne
Asus VLI + Ark1000VL is the VGA blaster, but i think you are talking about something i am not aware, or forgot about.
What is your top Wolf3D score ? The best i know is 153.3 fps. Did you get better ? Fully stable ?
Also, what are your best scores in the other tests ?

Re: Where to obtain 110 MHz RAMDAC for ARK1000VL graphics card?
Ark 1000 clocked to 80MHz (before it was 60MHz).
Values did not change much except Wolf 3D. Massive increase in speedsys video throughput.
I will make a new table of all Dos benchmarks for VLI 160, Ark1000 @ 80MHz here soon.

EDIT: Yes, fully stable. Windows 98SE approved.

Nice and fairly complete charts again. The first thing that will probably cross the minds of other's when viewing the charts, is 'how can i replicate this?' - e.g. was pshipkov only using 256K WB on that UUD, how much RAM, what CPU voltage, and was a peltier needed to be stable inside the case. It's probably too late now, but it would be helpful to include under each heading, in smaller font, something like (256k-i-wt, 2-1-1-1, 32M EDO, 1/0 ws, 4V, no peltier). I see you already have the video card listed on the chart, which is helpful.

Wouldn't the ARK1000VL be faster on the M919 compared to MGA? When looking at your scores, I get 77.3 fps in DOOM using ARK1000VL, while your tabulated score is only 75.4 fps.

Is the M918 run cacheless?

It would be nice if you were able to include more GLQuake scores for comparison. This means you'd need a Voodoo2, since this is more compatible across socket 3 motherboards compared to the Voodoo3. For reference, my M919-180 gets 30.0 fps in GLQuake.

What's the DOOM score for the Soyo 4SA2? The chart says not tested.

From the results, it looks like the M919 w/1024K is probably the best overall board for stable 180 MHz operation. I was able to run my M919 in a standard AT mini tower with air only cooling at 4 V. The challenge I ran into was finding memory modules stable enough to run at 2-1-1-1, 1/0 ws, yet while having more than 64 MB EDO DRAM. I eventually found stable modules in sizes 32 MB total, 48 MB total, 96 MB total, and 128 MB total. Do not use 64 MB total or you will have a performance hit, especially noticeable in Quake.

pshipkov wrote on 2023-02-25, 22:47:

Will add to the 386 story 2-3 more motherboards based on less popular chipsets, for completeness. […]
Show full quote

Will add to the 386 story 2-3 more motherboards based on less popular chipsets, for completeness.

First in line is Matra 486SLC2 VESA based on EFAR Microsystems 82EC392, 82EC495 (relabeled OPTi 82C392, 82C495B) chipset, hidden under cool stickers with the word "PATH" written on them.

Still in great condition despite some scratches on the VLB and ISA slots which indicate some rough handling here and there.
Warranty labels still in place. Traces of small battery leak that was taken care of before causing real damage.

One day will spend the time to find what "PATH" stands for. Anyway.
The motherboard is interesting for several reasons.
The obvious ones - a mixture of rare IBM BL2 (SLC2) processor soldered on-board, 256Kb level 2 cache (unusual for 386SX class system), and a VLB interface.
EFAR chipsets are seen predominantly in cost reduced assemblies, or late hybrid 386/486/VLB ones.
The first type are perf and scalability crippled in various ways, the second type come with 25/33/40MHz clock generators instead of crystal oscillators which limits overclocking options.
This motherboard comes with up to 80MHz clock generator, but also has wiring for crystal oscillator which opens the path to overclocking and discovering where the performance peak really is.
Removed the clockgen from its socket. Soldered 4 pins. Ready to roll.

It will be great if one day i manage to obtain one of these late EFAR based mobos to test with BL3, SXL2 and standard 386DX processors. Ok, back to tpoic.

It was established in the past that the fastest 386SX class system is Alaris Leopard (OPTi 82C295) with the same IBM BL2 CPU, so i was really curious how the EFAR chipset will stack against it and handle the unusual combination of hardware this Matra motherboard is.

Only 25 and 33 MHz base frequencies are supported via jumpers, but i was able to find a 40MHz configuration. Unfortunately between warm resets the frequency drops down to 20MHz from time to time.
This was another incentive to move to crystal oscillator.
The system takes 80MHz crystal with easy, but struggled with 90MHz. No lights at 100MHz.
While my hopes for higher overclock faded quickly, at least 80MHz became reliable between resets.
All jumper configurations on max.
The AWARD 4.51 BIOS resembles 486 layout and type of options. There are quite a few timing settings.
All parameters on max except:
CACHE DATA BUFFER OUTPUT = DISABLED (best is ENABLED)
DRAM QUICK READ MODE = DISABLED (best is ENABLED)
DRAM QUICK WRITE MODE = DISABLED (best is ENABLED)

Similar to Alaris Leopard this motherboard was not happy with most VLB EIDE controllers - their DOS drivers hang during initialization.
DTC 2278E worked fine but had to be slowed down a lot which kind of defeated the point of using an EIDE controller. If nothing else it proved that 2278E is one of the most compatible guys around.

Ark1000VL works fine.
One unexpected outcome - the S3 Trio64 VLB driver hangs Windows. This is an omnipresent occurrence regardless of jumper/BIOS settings.
Happens even with the slowest possible base frequency and most conservative wait states.
Used Orchid Kelvin 64-VLB 2Mb (Cirrus Logic GD-5434) instead - slower than Trio64 in Windows, but still pretty good.
(Will double check that stuff at some point later)

benchmark results

Matra / EFAR 82EC392, 82EC495 is in the top performance tier for 386 class hardware and on par with Alaris Leopard / OPTi 82C295.
It is same or faster in some of the DOS interactive graphics tests, slightly slower in the FPU intensive ones.
Really liked this motherboard. Not an entirely trouble-free experience, but somehow it grew on me a lot.
Maybe because it is a great/better alternative than Alaris Leopard, which until now was in a class of its own in the area of all things 386SX.
Also, this was my first time with EFAR silicon that i really wanted be good, and it did.

Excited to say I found one of these today. Facebook find. Owner left it behind in a barn. The people who bought the property were selling off stuff. Found an Adam computer, and told them after buying that to keep an eye out for other computer stuff. A week or two later and this with one of those classic at cases. The kind that everybody goes Gaga over with the LEDs and chunky front switch on the mini tower. I had no idea what this IBM CPU was when I first popped a little cover off of it.

@TimWolf
Good story.
Is the motherboard working ? These are quite rare. I have seen only 2 of them showing up online, mine including. Yours is the third one.
What else was in the PC case, since you mentioned it ?

@feipoa
Can add extra bits of info to the left-side descriptions - amount of L2 cache and RAM.
Not sure if adding there cache/ram timings will be useful - can get overcrowded, but will experiment with the formatting to see what can be done.
I rely extensively on the directory in the first post of this thread. So if i want to lookup one of the combined charts, i usually have 1-2 tabs open.
One for the chart and another for the post of the hardware in question.
In the individual posts i tried to establish some schema. For example, for 486 hardware you will see 4x40, 3x60, 4x50, 3x66, POD100 sections.
Timings and other details are listed there. Just the other day i was combing the UUD posts for more consistency.
But i understand that this is not straightforward. Fact is that this blog format does not fit the mechanics of a web forum.

Biostar 8433UUD 3.1, 200MHz (4x50), 256kb 12ns L2 cache, 32Mb, fastest timings, 1:1 CPU:PCI, 5V to CPU + air cooling
Biostar 8433UUD 3.1, 200MHz (3x60), 256kb 12ns L2 cache, 128Mb, fastest timings, 1:1 CPU:PCI, 3.45V to CPU air cooling

PC-Chips M918i
Running without L2 cache. Otherwise things get real slow, as we found out a while ago.

PC-Chips M919
I think you upgraded your RAM modules with 10 or 8 nanosecond chips which allows you to go slightly tighter on the wait states.
Can you list your results here, with details added ?
Would like to update the charts with them.
64Mb is no-no for this board. Weird stuff.
But you are right, this board with 1Mb L2 cache module is probably the nicest/easiest/fastest assembly for 3x60.

(Btw, i have to pick another top score from @CoffeeOne - Wolf3D with VLI+Ark1000VL @160MHz)

Soyo 4SA2 @180/200MHz was tested by @Chadti99.
I need to run the missing tests for it. That will be for later.

About more GLQuake tests.
Voodoo cards are not my thing, so i find it hard to motivate myself to deal with them.
UMC based boards being prickly about V3 and later models only makes it worse.
: P

Yeah I know, it can get messy, which is why I'd stick with the core hardware values and stick to a single line in small font. Any more and it would get cumbersome to read through.

The 8 and 10 ns SRAM modules did not improve wait states at 180 Mhz on the M919. Still can only do 2-1-1-1, 1/0ws w/EDO 32 MB or larger. I think 16 MB could do 0/0 ws, but this is also what you uncovered with the 12 ns modules. I was only intending to point out that your numbers could improve if you used the ARK1000VL on the M919. If you run MCLK on ARK1000VL at 80 MHz, you can further get some Wolfy3D improvement.

For my M919 v3.4BF, the details are:
Am5x86 QFP, 3x60, 4 V. no peltier
1x128 MB EDO, 1/0 ws
1024K double banked, WT, 2-1-1-1
ARK1000VL, 80 MHz MCLK
60 MHz FSB, 60 MHz to VL bus, 30 MHz to PCI bus

Start EDIT:
DOOM: 77.3 fps (fullscreen, but with utility bar)
DOOM: 75.22 fps (fullscreen, no utility bar)
Quake: 21.1 fps (previously I had 21.8 fps. I'm not sure what caused the drop in framerate)
GLQuake: 30.0 fps
PCPBench /VGAMODE (320x200): 32.0
WolfyD: 152.4 fps
WinTune 2.0 (Win95c):
12149 Kpix/s at 1280x1024x256c, 60 Hz
9644 Kpix/s at 800x600x65k, 60 Hz
3272 Kpix/s at 800x600x16M, 60 Hz

When did everyone start benching Doom without the utility bar? Is it just phispkov, or was it due to Phil's Bench Pack?
End EDIT

Another comment concerning the M919 at 180 MHz, it was difficult to find a sound card and graphics card combination which did not cause sound crackling, but I eventually found one. The M919 compared with the LSD at 180 MHz, I could only get the M919 to be fully stable on air cooling (no peltier) once inside a standard AT case.

Added on 31 Oct 2025:
The use of the ARK1000VL graphics card has an impact on operation at 180 MHz. If I use a Matrox G200 graphics card, I am able to run 180 MHz at 3.15 V, 2-1-2, 0/0 ws, with 8 ns 1024K megacache and a TSOP 64 MB EDO stick. If I swap the graphics for the ARK1000VL, I can no longer run at 180 MHz at any voltage (tested up to 4.4 V) and any wait states. I have been able to find other CPUs which can do 180 MHz at 4.4 V with the ARK1000VL and 2-1-2, 1/0 ws, but not with my special 3.15 V CPU.

pshipkov wrote on 2024-03-04, 08:02:

@BitWrangler Good sarcasm. How you throw the trash we play with to the trash. But more seriously - these Gigabyte GA486AM/S mobo […]
Show full quote

@BitWrangler
Good sarcasm.
How you throw the trash we play with to the trash.
But more seriously - these Gigabyte GA486AM/S mobos look powerful on picture, but are actually not very good.
At 160MHz (4x40) they are in the upper echelon, but takes effort to get them there.
180MHz is possible, but slow and flaky.
200MHz is just not happening.

Yes, mine will probably just get used for something "normal" like a DX4-120. That PC Player benchmark though, I have seen elsewhere it is always putting hardware combinations near the top that don't seem so great on anything else. I am in two minds of "That means it's unreliable for characterizing real world performance" or alternatively "It DOES show real world performance on everything not written by ID software." since Wolf3D, Doom, Quake are all products of the same minds more or less.

pshipkov wrote on 2024-02-27, 21:37:

<Chart> GLQuake (fps) […]
Show full quote

<Chart>
GLQuake (fps)

Microstar MS-4144 (4x50), MGA = 35.1
Luckystar LS-486E rev:D (3x66), Voodoo3 = 35
Biostar MB-8433UUD-A (4x50), MGA G200 = 30.5
<Chart>

At what resolution is the Matrox G200 running GLQuake and with which driver revision? 35.1 fps seems way too fast for a Matrox, especially with the Voodoo3 at 35 fps.