VOGONS


3 (+3 more) retro battle stations

Topic actions

Reply 240 of 376, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

Love the fact that you overclocked 3 out of 4 CPUs to run at same speed rather than underclocked the 1200 😀

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 242 of 376, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie
pshipkov wrote on 2021-06-06, 00:46:

Downclocking?

Sir, that goes strictly against my sense of decency. 😀

That's a quote worthy of a signature 🤣

Kudos on getting the Coppermines running at 1200, it's a bit pass their design ..

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 243 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

Yeah, the statement came out poetic and everything ( :} ), but joking aside - after reading your comment i was a bit stupefied, because it really didn't cross my mind that i can just underclock one of the CPUs.

---

Not far ago a 386 motherboard (unknown brand, model 386-VC-H) built around the VIA VT82C495/VT82C481 chipset got my attention.
The used 486 class chipset itself was worth checking from close-up, but there was something else that i wanted to understand - a second 80MHz crystal oscillator placed next to the FPU socket. That was unusual. Most often reference clock oscillators are 14.something MHz, but 80 ?!
Both are socketed, which implies they are allowed/supposed to be changed by design.
An asynchronous 386 FPU "hack" - does not sound plausible, but couldn't find any information online about this piece of hardware, so decided to spend my own time to examine it.

Great layout in great condition.
All jumpers are documented on the PCB.
Takes DLC upgrades - used 50MHz rated SXL2 CPU.
Up to 286Kb L2 cache - used 15ns UMC ones, because it didn't like 12ns ISSI.
Tested with 16Mb 50ns RAM, but it looks like it can go up to 64Mb, or even more.
STB Nitro ().

motherboard_386_via.jpg

Examination started without FPU - wanted to see how far i can go on the overclock.
Easily reached 60MHz in DOS interactive graphics. Board does not light past that (I know the CPU can take 65Mhz).
At 60 only very few video cards complied. Ended up using this 2Mb STB Nitro (Cirrus Logic CL-GD5434).
All BIOS settings on max, except DRAM wait = 1 (up from 0) and CACHE WRITE WAIT = 1 (up from 0).
Windows starts and works just fine, can complete WinTune2 benchmarks.
So far so good.

Time to move onto more rigorous testing (which requires FPU).
None of the Cyrix FasMath FPUs i have worked at 60MHz, but 40MHz rated ULSI took it for the team.
This was the most interesting part of the examination of the mobo - what is the purpose of the second crystal oscillator ?
Initially i quickly played with different oscillators - from 60 to 120 MHz and managed to fool myself that the FPU works asynchronously.
I was wowed for a moment, but after the initial excitement passed i quickly realized that it was just my unstructured approach to testing.
While a bit disappointed and back to Earth's orbit i was still ok with eventually having a system running natively at 60MHz ...
... until i ran the 3D rendering tests.
While there were no crashes, the results were all garbled pixels - which was indication that all is not well with the system.
Lowered the BIOS settings to their most inflated wait states, but nothing changed.
Had to step down to 55MHz to get the expected results from the test computations.

SpeedSys hangs with the SXL2 CPU, so sharing the performance metrics only.
As you can see this board is significantly clock-to-clock slower than the DTK Symphony one that i use as the highest 386 watermark.
Even at 55MHz VIA is slower than the 50MHz Symphony. Difference is smaller when running VIA at 60MHz, but is (border line) not fully stable.
FPU running at 55MHz helps the PC Player Benchmarks and 3D Studio offline rendering tests, which are FPU heavy. Interestingly enough - LightWave3D which is the difficult tests to pass and very FPU intensive, didn't care much about the faster FPU.
benchmarks_386_via.png

I didn't get async 386 FPU system (which i kind of expected), but the board is great - fully stable at 60MHz for everything but the most challenging tests. Even then it does not crash, just produces incorrect results.
It is not a clock-to-clock performance beast, but the extra MHz compensate for that.

At the end of the entire exercise I am still unsure what is the role of the second oscillator. The board works with and without it.

retro bits and bytes

Reply 244 of 376, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

Would a Weitek maybe be able to do something with the extra oscillator?

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 245 of 376, by megatron-uk

User metadata
Rank Oldbie
Rank
Oldbie

That's a really nice, clean and well laid out board. Also black simm sockets make it 3 times as cool!

My collection database and technical wiki:
https://www.target-earth.net

Reply 246 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

@H3nrik V!
Weitek FPUs are nowhere to be seen these days, so reading about how they work - looks like a synchronous operation ...
I may be on a whichhunt here, but something around that second oscillator does not add up.
Also, notice the green jumper - it controls FPU DLC setting. But similar to present or not second oscillator, it has no effect regardless of its state.

@megatron-uk
I don't easily handout "excellent board" badges, but this one gets it.

retro bits and bytes

Reply 247 of 376, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

It does seem that the Weitek has some extra clock signals than the 387 https://www.yumpu.com/en/document/read/484868 … eitek-bitsavers and the socket does say "WT3167", so might be something?

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 248 of 376, by feipoa

User metadata
Rank l33t++
Rank
l33t++
feipoa wrote on 2021-03-15, 03:09:

I should have a collection of these IDE drivers for Win3.1 and/or Win95. I'll look for them for you when I continue my Linux installation battle. Next up is trying to get my no-name USB-to-Wifi drivers working on my garage/mechanic station. I didn't run ethernet to the garage and have been regretting it ever since.
...
EDIT: This is what I have (attached). Please note, I do not maintain archive file names (the *.rar or *.zip name). I rename them so I can actually find them, so for anyone else finding this, you may already have these archives but with a different naming convention.

I wanted to update this. I tried the v3.2 Windows 95 driver on my Biostar MB-8433UUD. There is an installer file in the Win95 directory. You install, reboot, and find your system unresponsive once Windows 95 loads up. There's a new driver listed in the Device Manager for the IDE controller, UM86 something, but the system comes to a halt before I can check their properties. Anyone have any luck getting these drivers working in Win95b or c?

Back to using the Windows 95c supplied driver, you can click on CD-ROM or the HDD device in Device Manager and put the check box in "DMA", but it will be gone upon reboot.

What kind of transfer speeds were you getting on your MB-8433UUD with the built-in IDE PIO 3/4 controller? Can PIO-4 really only do 4.5 MB/s max on these things? I remember reading on another PCI 486 board that you were getting 10 MB/s? Is the IDE controller inside the UM8886BF defective, or just junky?

Ultimate 486 Benchmark | Ultimate 686 Benchmark | Cyrix 5x86 Enhancements | 486 Overkill Graphics | Worlds Fastest 486

Reply 250 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

you dont need driver for 496/7
you get ~10k in dos and windows 3.x
windows 95 and layer supports it as well.

---

the umc disk controller is ok, but needs drivers for both dos and windows.

---

cannot remember seeing crashes with the driver @feipoa proveded. i use it with win95 osr2. it just worked.
can double check during the weekend ...

retro bits and bytes

Reply 251 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

without drivers the umc controller gets around 4-5 mb/s. with drivers: 7-8.
sis 496/7 gets 7-10 mb/s without the need for drivers in pretty much every board based on this chipset and cpu running at 160mhz - tested on msi 4144, asus pvi, luckystar c2 and d variants, zida tomato, tms pci400, with PVI being slightly ahead of the pack.
numbers go higher at 180/200mhz with the late isa/pci models.

but on a 486 class hardware the best so far i get is with holtek and sis vlb eide controllers. there is a post in this thread with the metrics.

Last edited by pshipkov on 2021-06-18, 14:09. Edited 3 times in total.

retro bits and bytes

Reply 252 of 376, by feipoa

User metadata
Rank l33t++
Rank
l33t++
pshipkov wrote on 2021-06-17, 22:00:
the umc disk controller is ok, but needs drivers for both dos and windows. --- cannot remember seeing crashes with the driver @f […]
Show full quote

the umc disk controller is ok, but needs drivers for both dos and windows.
---
cannot remember seeing crashes with the driver @feipoa proveded. i use it with win95 osr2. it just worked.
can double check during the weekend ...
---
without drivers the umc controller gets around 4-5 mb/s. with drivers 8-9.

Which driver version did you try in Win95 OSR2? And what exact file and procedure did you follow to install it?

Looking at the v3.1 driver package, I think the Windows driver is for Win3.11. The v3.2 package has a Win95 driver, but it hangs up the system. Good thing I made a clone of the HDD before I attempted this. Are you using the v3.1 driver in Win95 and doing a manual install via the Device Manager?

Ultimate 486 Benchmark | Ultimate 686 Benchmark | Cyrix 5x86 Enhancements | 486 Overkill Graphics | Worlds Fastest 486

Reply 254 of 376, by maxtherabbit

User metadata
Rank Oldbie
Rank
Oldbie
pshipkov wrote on 2021-06-17, 22:00:

you dont need driver for 496/7
you get ~10k in dos and windows 3.x
windows 95 and layer supports it as well.

I do not. Tried it with a variety of CF cards and spinning rust. At PCI bus speeds ranging from 25-50MHz. Best I've seen in windows 9x is about 5MB/sec

Reply 255 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

that is unusually low.
at least i never encountered sis49x board with such low disk metrics, assuming decent bios settings, CF, or HDD.
i saw you posting recently some quake 1 numbers in another thread that were also too low.
what brand/model is your hardware ?

retro bits and bytes

Reply 257 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

i have EF, C2 and D.
While EF is slower than C2/D, i bet C is more or less the same as C2/D, which are top performers.
i posted some numbers for EF - even they seem to be higher than what you report.
something is off.

Last edited by pshipkov on 2021-06-18, 01:08. Edited 1 time in total.

retro bits and bytes

Reply 258 of 376, by maxtherabbit

User metadata
Rank Oldbie
Rank
Oldbie
pshipkov wrote on 2021-06-18, 00:42:
i have EF, C2 and D. While EF is slower than C2/D, i bet C is more or less the same as C2/D, which are top performers. i think i […]
Show full quote

i have EF, C2 and D.
While EF is slower than C2/D, i bet C is more or less the same as C2/D, which are top performers.
i think i posted some numbers for EF - even they that is faster than what you report.
something is off.

I'm willing to accept my IDE numbers may be off, but I thought my quake performance (16.7fps at 3x50MHz on the Am5x86) was reasonable, no?

BTW in Speedsys I get about 8-9MB/sec

Reply 259 of 376, by pshipkov

User metadata
Rank Oldbie
Rank
Oldbie

With C2 or D revisions i see around 18fps in Quake 1 at 4x40MHz.
Q1 is hard cpu bound, so given some decent vga, processor speed is the dominant factor.
dont think i ever checked 3x50 on these late isa/pci boards. so maybe the perf difference we are seeing comes from there.

retro bits and bytes