• Please review our updated Terms and Rules here

Clearpoint QED1 4MB Qbus PMI Memory

I'd say go ahead and ask Ron at VARx. He's been very supportive of the hobbyist community.

Note that he moved some years ago and has downsized his inventory a bit, but he still has quite a few rarities.
We have also had good experiences dealing with VARx, both in professional and hobbyist capacities.
 
I reached out to Ron at VARx, but he does not have that documentation. Bummer. This really shouldn't be so difficult to find.
 
Just to wrap this up, I was definitely experiencing cache failure on my KDJ11-B: I found the dead cache chip and replaced it. Just finished 10 passes of `VMJAB0` on the Clearpoint QED1 with no errors or weird stoppages, green light on the whole time!
 
sorry, real life had me, this works as 4mb pmi in an 11/83
 

Attachments

  • IMG_2292.png
    IMG_2292.png
    4.7 MB · Views: 16
  • IMG_2293.png
    IMG_2293.png
    4.3 MB · Views: 16
Last edited:
My board went to yellow again on Friday (at VCF West), then early this afternoon started failing POST (though LED still shows yellow). Switching the jumper to 2MB got it back to green - apparently the new failure(s) happened in the 2MB it doesn't use when it's configured for 2MB - and made it through the rest of VCF West, but I have some work to do when I get home...
 
I'm late to this, but I just got around to first, testing all the ram chips (all good, one was misinserted), and then trying my QED1 in an 11/73. Plugged into the second slot, after the CPU. No LEDS lit on the QED1 and the boot diags report no memory. Anyone else have one of these act this way?
 
Did you pull all the PMI jumpers? What architecture is the 11/73 backplane? Q22 / Q22, Q22 / CD, other? I had a misinserted DRAM as well, plus two questionable chips.
 
Did you pull all the PMI jumpers? What architecture is the 11/73 backplane? Q22 / Q22, Q22 / CD, other? I had a misinserted DRAM as well, plus two questionable chips.
Yep, no PMI jummpers (it came without any, so I didn't have to remove them). It's a BA23, so first three slots are Q22/CD and the rest are serpentine Q22.
 
Just for laughs it may be worth inserting all the jumpers on the left side and plugging it in above the CPU. Maybe it will work as PMI memory.
 
OK, there's another one of these boards up on ebay at the moment, and it comes with the manual. As of right now, lots of views and 11 watchers, so I don't think I'm starting a bidding war with this post... I hope whoever buys it will scan the manual and make it available. Or that someone can finagle a copy or scan of the manual from the seller. I haven't contacted the seller, and I won't be bidding on the board.

There is a clue in the listing as to why there are 148 memory chips - it says (probably extracted from the manual) that "the QED-1 had 64-bit words with 10-bit error correction." (64+10)*2 = 148, which matches the number of chips on the board. The board self-identifies as Parity, not ECC (with whatever pronouns are appropriate for that, I suppose), which probably simplifies the interface or avoids infringing on a patent or something...
 
The seller is VERY well known in the RSX community.
He has also forced the IA to block the contents of his old website.
I assume he's retired and thinning the herd.
 
That IS interesting Steve and explains why my attempts to map the board to figure out bad chips drove me bonkers. I figured it was doing some kind of ECC code, but could not figure out what.

64 bit reads is interesting as it can probably buffer 4 16 bit words and feed them quickly to the 11/73 CPU over PMI.

What was wrong with the seller? Jerk? Seems to deliver the product.
 
Ok. Well, at least the stuff is getting into peoples' hands. PMI memory is a lot nicer than non-PMI.
 
OK, there's another one of these boards up on ebay at the moment, and it comes with the manual. As of right now, lots of views and 11 watchers, so I don't think I'm starting a bidding war with this post... I hope whoever buys it will scan the manual and make it available. Or that someone can finagle a copy or scan of the manual from the seller. I haven't contacted the seller, and I won't be bidding on the board.

There is a clue in the listing as to why there are 148 memory chips - it says (probably extracted from the manual) that "the QED-1 had 64-bit words with 10-bit error correction." (64+10)*2 = 148, which matches the number of chips on the board. The board self-identifies as Parity, not ECC (with whatever pronouns are appropriate for that, I suppose), which probably simplifies the interface or avoids infringing on a patent or something...

I'll pitch in if you want to win it, so that we can have the manual scanned!
 
Yeah whoever goes after it is fine, I'll pitch in!

If anyone else in this thread is thinking of driving the price up because they really really want one, please don't and I'll sell you a tested/working PMI board for less :D
 
Back
Top