Better replace that VARTA battery - - soon.

majestyk wrote on 2024-04-04, 15:37:

Better replace that VARTA battery - - soon.

Indeed, first thing I'll do tomorrow. I still haven't connected the motherboard to a PSU to check if it works. It came from the mail this morning.

It's a 486S, maybe has a separate FPU underneath.

BitWrangler wrote on 2024-04-04, 16:55:

It's a 486S, maybe has a separate FPU underneath.

Good guess! I didn't think about that. The 487S is supposed to be quite rare. That would be a great find, for the price I paid...

487S is not that rare, they occasionally turn up by the trayful. Not so common on an interposer though, but still not what I consider rare. Guess it depends where you hang out and buy 😀

MSxyz wrote on 2024-04-04, 17:24:

BitWrangler wrote on 2024-04-04, 16:55:

It's a 486S, maybe has a separate FPU underneath.

Good guess! I didn't think about that. The 487S is supposed to be quite rare. That would be a great find, for the price I paid...

We should make an interposer project for this.

386 and 486 version. If memory is serving many boards didn’t have math co pro slots so this could be handy.

Of course we know that on intel 486 the 487 is just a 486 dx 🤣. You can turn a 486 dx into a 487 btw with a pin mod

Sphere478 wrote on 2024-04-04, 21:46:

We should make an interposer project for this.

I don't have the skill or time for this but would buy one if there was one that worked for UMC processors since they never shipped one with a FPU and the other FPU don't seem to work with the UMC.

demiurge wrote on 2024-04-04, 22:00:

Sphere478 wrote on 2024-04-04, 21:46:

We should make an interposer project for this.

I don't have the skill or time for this but would buy one if there was one that worked for UMC processors since they never shipped one with a FPU and the other FPU don't seem to work with the UMC.

You’re thinking 486 168 pga variant?
Got datasheet for chip in question?

How’s chip availability?

The way these projects work usually is someone designs the pcb, someone tests it, then you order it from a pcb company and assemble it yourself.

Most people around here don’t put forth the margins that would be required for someone to want to make them and sell them for a device this simple many could make it themselves
. The sound cards and video cards sell but even there I wonder how much those guys make. I publish all my work to public domain anyway so building it yourself would always be an option if I did it.

Anyway, I might be willing to play with it. Convince me there is enough interest and that the chips are available and it’s practical. And we would also need to examine an interposer (possibly destructively) (will see maybe not, looks pretty simple )

Sphere478 wrote on 2024-04-04, 22:55:

looks pretty simple )

I had ChatGPT lay one out, only took 7 layers and two folds in spacetime. ... ... kidding.

BitWrangler wrote on 2024-04-04, 23:06:

Sphere478 wrote on 2024-04-04, 22:55:

looks pretty simple )

I had ChatGPT lay one out, only took 7 layers and two folds in spacetime. ... ... kidding.

Lmfao!!

But for real, this is probably 4-6 layer. I have a hunch 4 will work.

Good news and bad news for me... 😀

The "interposer" is indeed a Cyrix 487S FPU. So the CPU is piggy backed on top of the FPU.

The bad news is that the system is a bit underperforming. It may be due to the old MB (A Soyo SY-025D1) with an early vlb chipset, but maybe there is more... Even when using the fastest BIOS settings, I've run a few synthetic tests and this CPU seems to be barely faster than a 486DLC (which runs off a 386 bus and not on top of the better performing 486 bus). I've a hunch the internal cache also work in Write Through mode because the motherboard predates the arrival of 486 with WB caches. At any rate, there is also another odd fact... CacheCHK doesn't seem to recognize the 2KB cache at all (see pic). Before working further on this setup, I need to dig up some old utilities for the Cyrix CPUs. The one I used for the 486DLC doesn't appear to be working with the 486S.

The last line says "clocked at 486 45.5MHz. Cache ENABLED."
I take that to mean the internal cache is in fact enabled.

Cyrix had writeback cache in some of their CPUs about a year before intel on the desktop, so that doesn't necessarily mean the cache must be in write through mode. The date code on the chipset is the second half of 1993, which lines up perfectly with the release of the 486S. They most likely worked with a few chipset vendors to make sure the features unique to their CPU were supported. If you want to know for sure, you can dig out the datasheet for the SiS461.
The system running about the same speed as a 486DLC kind of makes sense, as it basically is a 486DLC with an extra 1kb internal cache. They were definitely a notch below proper 486s, which is why this product was not very successful.

Can you post the graphical chart from SpeedSys 4.78 please?

Anonymous Coward wrote on 2024-04-05, 15:04:

The system running about the same speed as a 486DLC kind of makes sense, as it basically is a 486DLC with an extra 1kb internal cache. They were definitely a notch below proper 486s, which is why this product was not very successful.

You are right. I dug through all my documentation and datasheets and it appears that the 486S has the same core as the 486DLC with a different bus and cache. The later Cyrix 486DX is an improved design that is closer, performance wise, to Intel and AMD 486s.

However, I was also right that the internal cache was working in write-through mode. I found a small utility that lets you switch the internal cache of all Cyrix 486s between WT and WB mode: https://github.com/karcherm/cx486wb

By default, after boot up, the processor cache is set in WT mode. Switching it to WB yields a modest 5% increase in real world applications (I tried with Wolf3D and Doom).

Disruptor wrote on 2024-04-05, 16:00:

Can you post the graphical chart from SpeedSys 4.78 please?

Unfortunately I put the motherboard back in storage after removing the CPU+FPU from the socket. I want to mount it in a faster motherboard because this one had both memory and bus bottlenecks. In some tests, I was measuring lower performance than what a 486DLC scored on my Panda 386V VLB motherboard. In a couple of week, I should be able to go back to this project and I'll post screencaps from various tests.

On some CPUs, you might be able to force write back through pin mod. Pulling high/low check your data sheet

K6-2/3+ Success on Asus P5A 1.06 - simple mod

Socket 1/2/3 Voltage Interposer Tweaker (Alpha)

Sphere478 wrote on 2024-04-05, 18:42:

On some CPUs, you might be able to force write back through pin mod. Pulling high/low check your data sheet

The selection of WT/WB via a configuration pin applies to late AMD and Intel 486 processors. Cyrix 486 processors (including bastards like the 486SLC) have been software configurable from the start. If a BIOS allows selection of L1 WB/WT, it is usually just use if a Cyrix processor is installed, although some quirky BIOSes use that BIOS setting also to correctly configure the chipset to WB or WT mode of AMD/Intel processors (instead of the "correct" way of determining the correct chipset configuration from the software readable CPU ID).

Moving the 486S to another board is probably not the best idea. From what I understand not many motherboards actually support it.

I would imagine it tends to want a 386/486 chipset as it has a foot up each ass, excuse my Red Forman.

Quite a lot of 486 boards use chipsets that also support the 386. The bigger issue would probably be the pinout, which I believe is slightly non-standard. Then you'll need a BIOS that can correctly identify the CPU and enable the cache at boot. I've never owned a 486S. Feipoa has one, and I think he's tried it on boards that don't officially support it. I forget how that worked out.

But seeing as how you already have the board that probably came with your 486S, you should keep all that stuff together. Having the board, the CPU and interposer makes it somewhat collectable as far as I'm concerned.