dml attempts a Raven

[PhilC]

@dml you need to partition it Fat16. I have 8 partitions on a 4GB card formatted using windows.


BTW,if you and @agranlund could send me your address details, then I'll send you some parts to make up an Excelgraph board. You'll need to try and locate the chip yourself though as I've not had any luck with this.

[dml]

@dml you need to partition it Fat16. I have 8 partitions on a 4GB card formatted using windows.

Yeah, was gradually realising this is necessary, or at least more likely to work - trying to get my hands on an old style IDE adapter for PC to try this again.

BTW,if you and @agranlund could send me your address details, then I'll send you some parts to make up an Excelgraph board. You'll need to try and locate the chip yourself though as I've not had any luck with this.

:thumbsup: :thumbsup: :thumbsup:

[dml]

While I wait for more extremely mundane but important bits and pieces to very slowly turn up by post, I took some pics from the cards I will be playing with, assuming I don't blow something up first.

Here is the Diamond Hercules with the sticker peeled back to reveal the only W32i chip I have at the moment. I still need to test this card in the Raven.

w32i.jpg

And here are some of the verrry interrresting cards waiting for attention:

full-length-cards.jpg

The bottom two are effectively full-length TIGA cards - top one with a TMS34010 and the bottom one has a TMS34020.

The top board is a MVA/256c card and chances are it is ok. The bottom board is a MiroGRAPH and my concern with this one is the crazy number of GALs there. If any one of those has a single bit expired, the board is probably scrap for part-harvesting. If it works though, it could be fun to poke around with.

I have been playing with the idea of desoldering the GALs, trying to read them and program fresh copies then reinstall with sockets. However chances are the GALs are all locked since the card was a commercial product and they probably didn't want to encourage clones. If that's the case, it's probably quite far along the road to self destruction (~35 years!), if it even still works o_o. At least the TMS chip and the RAMDAC & RAM could be reused...


The board on the right is some kind of industrial control board. I think I had identified it as something to do with dental equipment but not sure now. In any case, it's not really a video card even if it does appear to have a TMS34010 on it and a VGA port. That's probably just for monitoring & feedback on what the card was supposed to be controlling. There is a lot of isolation on there and at least 3x DC convertors...

[agranlund]

Oooh, Tiga cards!

Have you managed to find any low-level information on these?
Or would it be a case of disassembling and trying to make sense of tigacd.exe? :)


I have a similar (same?) model Diamond Hercules card which works very well in my machine.

[dml]

Have you managed to find any low-level information on these?
Or would it be a case of disassembling and trying to make sense of tigacd.exe? :)

I have some scraps of info but there is likely to be some distance between general TIGA specs and a specific card implementation so I am certain there will be some RE involved even if that means taking the driver apart to scrape register ports. It's probably not going to be a quick thing.

There is a third (and final) TIGA card - I'll add a photo of that when I can. I'll probably focus on trying to save the 34020 card first (or at least figure out of the GALs are protected or not) since the '020 is a more fancy chip. It is clocked a bit slower though at 32, versus 50 on the other one.

I have a similar (same?) model Diamond Hercules card which works very well in my machine.

:thumbsup: that's good to know!

I'll be playing with this as soon as my storage solution is working. Hopefully sometime at the weekend or next week.

[frank.lukas]

The MATRIX TC1006/TC1208 VME Bus Graphiccard has a TMS 34020 GPU ...

There are NVDI Matgraph or NVDI 5 with install option Matrix TC as driver.

TC1208E.jpg

[PhilC]

@dml have you tried a usb to ide44 adaptor? I either use one of those or use a CF card IDE setup

What's so interesting about a Tiga card then? Is it something I should try and find?

[dml]

The MATRIX TC1006/TC1208 VME Bus Graphiccard has a TMS 34020 GPU ...

There are NVDI Matgraph or NVDI 5 with install option Matrix TC as driver.

That's a very nice card. But I expect absolutely unobtainable now (!?). The driver might be interesting to look at indeed...

[dml]

@dml have you tried a usb to ide44 adaptor? I either use one of those or use a CF card IDE setup

That is one of the things I am waiting for :) I can go to the CF option though as a last resort... I suppose I could prep that from Windows with the stuff I do have.

What's so interesting about a Tiga card then? Is it something I should try and find?

TMS340x0 are programmable GPUs with instruction set & some info available for them - since TIGA (Texas Instruments Graphics somethingorother/Adapter?) was meant to be a sort of open standard ('open' can mean different things esp. after multiple decades and stuff getting lost).

Remember the Elmtech PARSEC board for ST back in the day? More or less the same thing, but on an ISA slot, behind various perplexing puzzles most likely.

(I didn't know the Matrix card above was another variant - curious!)

[dml]

This evening I built the Raven board into its new case & ATX PSU.

Something I noticed quite quickly - the machine crashes quite rapidly if the PS2 mouse is plugged in. Sometimes it crashes when activating the EmuTOS dropdown menus, sometimes when opening the info dialog. Sometimes it crashes with a LineF panic on the boot screen.

Without the PS2 mouse present, it doesn't seem to do this (or it's a lucky coincidence). I'll do a bit more testing later to make sure the two things are related.

The CPU is still running at 80MHz and I'm still using the FN33 68150, with a FN40 waiting to be installed (not sure about the 1508 - I had to swap it once already - will check it later).