Found the board on Stason.org:

http://stason.org/TULARC/pc/motherboards/F/FO … ml#.Up71S8QW3l0

The brand is FORCOM TECHNOLOGY CORPORATION.

Today I peeled the sticker off the chipset because I was curious what it is.

It is....

Some chipsets contain a very small amount of internal cache for low cost systems. On the MX chipset, 8kb. On the SiS Rabbit chipset, there is a miniscule 128byte cache of some sort. I assume this makes a difference. Not sure about the SARC chipset though.

SARC, please tell what does this mean / stand for?

Who knows. It's hard enough trying to get information for the mainstream chipsets, much less the smaller guys. SARC seems decent, but usually found on low end 386 and 486 boards.

A couple of days ago I bought the same board with 486slc-33, rev. 2.6. It came today 😁

I quickly set it up and done some benchmarks. But to my surprise it was a bit slower then I expected - I used to have 486slc-33 back in the days and still have banchmarks of it - PMIPS to be precise.

Well, what can you do... 😒

I did all the standard benchmarks and then something interesting, yet unexplained happened, it begun to run faster, but only until reboot 😕

Setup configuration:

486SLC-33MHz, 1KB cache enabled
16MB 70ns
CL-5429 2MB ISA
HDD Controller ISA

These are results with a clean boot:

3dbench 1.0c... 11.6
PCPlayer... 2.8
DOOM... 15994
Checkit... 5666
Speedsys... CPU: 13.91, MEM avg: 7.74
Vspeed... 2.95
PMIPS... 8.2
Cachechk... 1: 45, 2...16: 127
F1GP... 258%
Sysinfo... 54.3

And these are results when it runs faster:

3dbench 1.0c... 15.8
PCPlayer... 4.0
DOOM... 10851
Checkit... 9337
Speedsys... CPU: 16.81, MEM avg: 17.25MB/s
Vspeed... 3.32
PMIPS... 9.3
Cachechk... 1: 41, 2...16: 64
F1GP... 184%
Sysinfo... 54.3

This latter results are what they should be in the first place.

After digging what actually made it run faster... it was only after I run SYSINFO. I guess I'll add sysinfo /tsr in autoexec.bat

This board has jumpers for 33/40MHz operation but it isn't stable at 40 even with additional cooling. At 40MHz it runs a bit faster then 386DX-40 - 3DBench is 18.1

I also did the benchmarks with 1KB cache disabled and they are relly low. Without SYSINFO it runs like a 286-10MHz and with SYSINFO it runs like 386SX-25.

I wish BIOS had more advanced options.

And what does SYSINFO actually do to make it run like it is supposed to?

Pictures:

🤣 that's very odd.

Yes, to say the least 😲

And there are no BIOS options nor jumpers... just the SYSINFO that does the trick 😒

I tinkered a bit with this board again... and it seems all it needs is a jumper on the turbo switch (as many boards do but the difference between ON/OFF with this one is minimal, so it wasn't that obvious). But it's still weird that sysinfo would somehow turn the turbo on.

Also tried to run it at 40Mhz again. It works fine with 60ns simms and additional cooling of the CPU - it gets really hot.

At 33MHz with turbo on it's exactly the same speed as my 486slc-33 from Acer back in 1993. Nice flash back 😀

I always love reading about this board. I like it a lot, it's so minimalistic and really reliable too.

Mau1wurf1977 wrote:

Ok took screenshots for both systems: […]
Show full quote

Ok took screenshots for both systems:

386DX-40 system with UMC chipset and 256KB cache disabled:

]

Which benchmark program is this?

386_junkie wrote:

Which benchmark program is this?

That's Norton System Information. It's included in this benchmark pack with lots of others if you want to check it out: http://www.philscomputerlab.com/dos-benchmark-pack.html

@Mau1wurf1977:
Do I get any noticeable advantage if I buy a 80387 NPU to fit in the socket?
Do you think this will work? http://www.ebay.com/itm/1PCS-Arithmetic-Proce … cgAAOSw7FRWUx5c
It is only 25MHz, but the CPU speed is 33MHz

It might be okay. 1992 is kind of borderline for overclocking potential. Why not just get a usli or iit 40mhz part, or a 33mhz Cyril fasmath?

http://m.ebay.com/itm/Coprozessor-80387-33-MH … a%257Ciid%253A2

Anonymous Coward wrote:

It might be okay. 1992 is kind of borderline for overclocking potential. Why not just get a usli or iit 40mhz part, or a 33mhz Cyril fasmath?

http://m.ebay.com/itm/Coprozessor-80387-33-MH … a%257Ciid%253A2

I searched only of 80387SX and found the 25MHz version. Didn't know there were others. I see if I can get it cheap in the auction. The IIT seems more expensive that I can afford right now for a hobby. But I will keep an eye on them, thanks. So it means that I have to match the frequency of the CPU.

You don't need to match the speed, but it should be at least as fast as the cpu, unless you run the fpu in Async mode(intel only) which is much slower. Btw, for 386 generation Cyrix fpus are fastest.

Sorry to bump this thread, but I managed to score one of these from the recycler with a Cyrix Cx486SLC/2-50 processor onboard. Amazingly, even though it fell off a forklift onto the ground, it still boots and runs.

With the turbo on, it feels rather nice for having no L2 cache, definitely like what I'd expect a 386SX-33 to feel like. It came with an Oak OTI-067 VGA card, which originally had 256K but I added an additional 256K to boost it to 512K; despite this, my Trident 8900CL ISA card is definitely faster on it.

I also own this board with a ti486SXL2-50 on board. The board doesn't post because of substantial damage to the area around the battery.
Is it possible to make some detailed pictures of this area and upload them here? Then I can try to figure out which paths need repair 😐

This board would be pretty slick with the SXLC2. I used to own this board, and was planning to swap the 386SX-40 for an SXLC-40, but sold the board before I got around to it.

Mau1wurf1977 wrote:

Motherboard on Stason.org: http://stason.org/TULARC/pc/motherboards/F/FO … ml#.UpPYwcQSbzY

And I found a PDF version and attached it to this message!

On my motherboard the jumpers for clock speed are hard-wired or don't exist. Might be a newer revision with the jumpers.

I found another version of that manual, almost identical, except that the cover page lists 486SLC and the last page lists info about jumper that disabled the battery during shipping.