Matrox E-VDP

[Malc]

fa62 05    **

Your unipak has rev 05 firmware, Rev 09 rom images are in the files section of the Data I/O group on Yahoo, Worth joining it's free to join.

70b3 64k 29b v01

Your 29B unit has 64k memory and v01 firmware, You might be able to upgrade the memory to 128k depending on your memory board, to go higher you'd need a new memory board 1 Mb max. The v01 firmware is upgradable to v6, I think rom images are in the files section too. I would upgrade the unipak 2 and see how it goes for you, You can always upgrade the rest later if need be.

 

[NeXT]

Okay I'll start working on that on the side but it will take a few weeks as I ahve a number of other projects on the go that take priority. Sourcing the memory might be a strange task but I'm sure we're on the right track for that now.

Anyways I still can't find any faults even after replacing the missing tants and reseating multiple chips. Aside from cleaning the case up and refurbishing the LD player which has a broken door hinge and still needs two belts there's nothing else I can do. unless I find another board. For reference, here's the board I am now seeking with much better quality photos.

 

[bear]

I would start by replacing the capacitors you removed, before assuming anything more serious is wrong.

That D27513 has a 4-page address space, not a linear one like the D27512 you were trying to treat it like. It is probably fine, or at least what you observed is probably not a sign of any real problem.

 

[NeXT]

I would start by replacing the capacitors you removed, before assuming anything more serious is wrong.

Excluding the shorted caps on the backplane and one on the DRT, this has been done and so far nothing has been unhappy with them being in place again. For the others I'll have to source replacements but as per usual I can only determine half of their ratings. They're either marked 15V or 173D but their rating stamp is either 9032 or 226M which I think are 32pf and 26pf.

We could also technically be dealing with even more shorted tants in a section of the video board and SBC that I missed. There's something between a rail and ground that read 54 ohms but it's not enough for the short to blow open and not enough to pull the voltage down and because it's the +5 rail it could be anything, if there's a short at all. It could always be another component functioning properly in the circuit like that block inverter.

That D27513 has a 4-page address space, not a linear one like the D27512 you were trying to treat it like.

I noticed that in the datasheet and a random thread on CPU-world forums. I was expecting it to have some sort of problem trying to dump in the wrong mode but I wasn't expecting something around half the contents of the chip to accurately dump. Anyways, if you believe that the chip and its contents are fine for now I take my mind off it and try to focus on what else might be the issue.

Edited: I found another blown component. L3 on the DTR board appears to be a 1/2w resistor with a thin gold stripe, two thin black stripes and a thick silver stripe. It's a little crispy with cracks in the shell but reads 8.3 megs. Under load one side attaches to the 12V rail but the other side reads -4.1v and pin 4 on a nearby LF347N seems to link to it. Again, no idea what's up.

 

[bear]

Whoa there, slow down a little.

1) 9032 looks like a date code.
2) 226M is 22 uF! That's six orders of magnitude off (22000000 pF).
3) 54 ohms isn't a short. At 5V, it's a load sinking ~0.1 A. If it's across 12V, then ~0.25A.
4) L3 is an inductor. It should have an almost zero resistance, but you can't measure it while it's in-circuit. It's unlikely there's anything wrong with it. The casing might crack from age or heat, but the physical properties of the coil of wire inside is what matters. If it fails electrically, it will fail open. Otherwise, if it's physically in one piece, leave it alone.

 

[NeXT]

Awesome. So at least we got a bunch of questions answered.
Okay, so I totally did not realize that there was something printed on both sides of the tant.. >_>

In reality, all of the tants are identically marked:

173D+, 226M
15v+, 9032

Don't panic too much. I'm not going to start pulling components off the board until I'm told so at this point.

 

[NeXT]

Managed to get into the BIOS. Again, we're currently running with the VGO board removed. You press and hold the DEL key during POST and it gives you an absurdly basic BIOS (set date, set time, ask if you have an internal disk, ask if you have an external disk, then tells you what floppy drives are installed and if you want to save and reboot).

Once that was set the two floppy drives worked and the machine would boot from the hard drive. Now of course I need a new 3.6v lithium cell and I'd sub in one of my many mac 3.6v PRAM batteries but there is a voltage present on the computer's battery terminals when the machine is plugged in so me thinks I just need a diode to prevent it from blowing the new battery up?

The belts in the laserdisc player have also been replaced.

 

[bear]

new 3.6v lithium cell and I'd sub in one of my many mac 3.6v PRAM batteries but there is a voltage present on the computer's battery terminals when the machine is plugged in so me thinks I just need a diode to prevent it from blowing the new battery up?

Because it's not a lithium cell and is in fact a rechargeable NiMH cell? Yes.

 

[NeXT]

Thought so. I had to deal with one of these buggers once before on a portable Imgdisk rig I built for someone.

 

[triton199]

Hello, First post here and naturally I'm bumping a thread back from the dead, but this time I promise it's with good reason. I found this thread while google searching for more information about the E-VDP and trying to dig up an article that I know used to be on the Matrox website years ago. no luck there yet unfortunately but I should get to the point.

I also own an E-VDP! so that makes at least 3 that do (or at least did as of 5+ years ago) exist.

from what i remember reading the initial (and probably only) order by the military was 12,500 units, unknown how many were delivered. i bought it a number of years ago at an auction with the original military box, styrofoam and manual. the box and styrofoam are in pretty abysmal shape but the unit did power on and boot to bios at the time. last time i tried to power it on i think it blew caps in the power supply so i'll be opening it up to poke around and clean it up sometime soon. lmk if anybody wants more pictures lol

 

[NeXT]

Wooow, you are indeed the second known machine pictured in recent history. Sorry I did not see this post sooner.
Do me a big favor and scan that instruction manual. Likewise once the power supply is up I really want to know how your video card works. I've still been unable to get it to POST.
Please verify the condition of the tantalums on the ISA backplane. I had multiple blow up before it settled down and there were no shorted voltage rails. Also, PLEASE backup your hard drive. There are unobtanium drivers that should be present that relate to the EIDS laserdisc standard and the cards.

 

[triton199]

I will absolutely work towards getting you scans of the manual and an image of the hard drive as soon as i can. It may be a while because life gets in the way sometimes, but i'd love to get this thing booting and operational again. I've been doing some digging on background info and documentation on the thing and will work on posting some sources for interesting info. based on my limited research so far there many have been as few as a couple thousand units made, or some number between 12,500 and 47,500 units depending on what sources you can still dig up. My ultimate goal, and I may have to recruit some help to get it to this point, would be to get to the point where i can figure out how to run my Space Ace arcade game laserdisc in the machine. may be a pipe dream but in theory its possible because the LD player that this machine uses is one of the few that is known to be compatible with dragons lair and space ace. Honestly I'll be happy to get it to boot again! will post again as soon as I have more to share.