• Please review our updated Terms and Rules here

PDP 11/45, Part 5

Hi All;

Dave, Thanks for the update information..

"" Do you have any idea what your RAC card revision is Marty (or I suppose the next best thing is the part number on the E89 ROM). E89 (Rev 'H') is supposed to be '23-C02A2' (as is Rev. 'F'). ""

Mine show '23-C02A2' just like Yours.. I think my Rev on the Card is (I think) REV 'C' .. As mine says ""M8103C P2, but it could be 'P' ..
What is written on the Back of Your Card ??

For the Most part my Proms are not TI, I have three TI, three National, and the rest that I have no idea of who they were made by, It kind of looks like 'IIIIIL' ..

Also Proms E99 and E100 are both Good.. Now to the Next set..

"" I think you may have come across a microcode 'fix' here! "" OK ..

At Location '056, (once we get everything read) I think we should compare what the Micro-code does in each of the scenarios and make a flow chart for each one and try to see what changed between the two Versions..

Also Proms E105 and E107 are also Good.. Now to the Next set..

This also means that If the rest of the Proms are OK, that there is something else wrong with my RAC Board..

E106 and E108 are Good.. Now to the Next set..

I found a Bad Prom, I saw it change right in front of my Eyes from the Correct Code to the Incorrect Code.. E115 is Bad and will need to be replaced.. The first four Addresses are no good.. And it's a TI Prom !!

E114 is Good, I checked it.. Next is the Last two proms..

E113 and E116 are Good to go..

Dave, Your Hex file on Address '027 You have C8008 and it should be C9808, just in case, there is any confusion..

So the only really Bad Prom is E115..

I was able to verify most of the E115, with cool spray, except for the first 16 addresses of '200 to '217 which has a problem that no amount of cool spray will seem to correct..


THANK YOU Marty
 
Last edited:
The fact that the numbers are the same on E89 implies the contents should be the same...

TI and National Semiconductor did each other's chips (I assume there was a tie-up at some point in time).

Now if the unknown devices were 'ILL' I could understand it :)!

Same numbers on ROMS should equal same contents - so why are we not seeing that? Let me sleep on that overnight...

Dave
 
I have also checked my microcode myself and I have found nothing else other than you guys have spotted:

At address '027 change field DRK from [BLANK] to '3'.
At address '061 change field CCL from [BLANK] to '4'.
At address '122 change field BRK from [BLANK] to '1'.
At address '123 change field BRK from [BLANK] to '1'.

I will do this when I get back from my business trip and re-issue the spreadsheet and PDF files.

I will also remove a few extraneous zeros that have crept in by mistake - I prefer [BLANK] instead of zeros...

Hopefully (by then) Marty should have a set of ROM images that I can import and compare to my magic numbers.

I will have a look at the other ROMs next (KT11-C, instruction decode and ALU) and add these as separate tabs in Excel.

FYI Marty: My bare UA11 and 2*KM11 PCB's have just turned up from Shiresoft. I shall order some components shortly. I have a few extender cards and prototype boards on order from Douglas Electronics. Hopefully they should arrive shortly as well.

Dave
 
Hi All;

Don, I am Sending You the Files for E89, E91 and for E115..

Would You be willing to Program these three for me for Now, and If the rest of the Proms Check out, I will then Send You out the Complete set..

But, I don't want to waste any more than necessary of the Proms that I would Like to have Programmed..

I can then see if that fixes the problem or if there are more things to check out on my RAC Board
and I can compare E89 and E91, using both with the maybe 'fix and what You would program as is Listed on the Source Code Listing..
I could compare the two and see what they do that is different from each other..

THANK YOU, so Very Much for doing this and making the offer..

THANK YOU Marty
 
Last edited:
Hi All;

It looks like my system has become Infected, and I am going to have to Re-install Windows 7, so I don't know when I will be back..

Win 7 has been Installed, but some of my other programs have not, as of yet..

THANK YOU Marty
 
Last edited:
Marty,

I have downloaded the latest binary ROM files you sent me and compared them to my spreadsheet. I get two deltas:

Address '027. I get:C9808BC031004CCF. You get:C8008BC031004CCF.
Address '061. I get:07C1000006C100AC. You get:0781000006C100AC.

It's a bit late in the UK to have a look at these now - I will only make more errors than I solve! I will have a look tomorrow.

Dave
 
Hi All;

Dave, You are correct on Both of them, I will check the two files, and most likely it is a typing mistake on my part..

I can't get the Hex Editor Program that I had bought to load, it say it can't find a dll file, which I have, and so it won't load..

So Dave, You might have to make the change Yourself..

THANK YOU Marty
 
Last edited:
Hi Marty,

Just for my information - are these new BIN files produced from the microcode fields or from your 11/45 ROMs?

If from the microcode, I can generate BIN, HEX or any file format you like directly from my EXCEL spreadsheet and (baring programmatic errors on my behalf) they should be 100% consistent with the source. Effectively, I am using EXCEL as a microcode 'assembler' if you like.

The important thing from my perspective is that the contents of the fields that I entered by hand from the 11/45 engineering drawings are correct. I have checked them through myself and (with both of our eyes looking at them) I think we are there now. If we need to use the 11/45 microcode for anything in the future (e.g. blowing PROMs) then it would make sense to spit the files out of my EXCEL spreadsheet in the format we want them so we know they are consistent rather than performing activities by hand on them.

Yes, DLL's were always a pain in Windoze. I seem to remember you needed to ensure that the DLL's were in a directory that was referenced by the PATH environment variable. Since I went to an iMac most of those horrible thoughts don't exist anymore (although I am still using a Virtual XP box under OSX for odd bits of software that I don't have for the iMac that I still use).

Dave
 
Hi All;

Dave, Thanks for Your Response..

"" Just for my information - are these new BIN files produced from the microcode fields or from your 11/45 ROMs? ""

Both, as stated before somewhere lately, I compared the files that I could with what I have in the Notebook, all of it was OK, that I was able to check, except for the differences that You thought might be a Fix..
And for a Part of E115 that is Unreadable..

The Exxx_256x4.bin files are supposedly extractions from Out of the Exx.bin files..

"" If from the microcode, I can generate BIN, HEX or any file format you like directly from my EXCEL spreadsheet and (baring programmatic errors on my behalf) they should be 100% consistent with the source. Effectively, I am using EXCEL as a microcode 'assembler' if you like. ""

If You Look at the _256x4 format, You'll see that it is slightly different from what You might generate or extract straight from the files..

"" The important thing from my perspective is that the contents of the fields that I entered by hand from the 11/45 engineering drawings are correct. I have checked them through myself and (with both of our eyes looking at them) I think we are there now. ""
Yes, I assume that they are correct and that we need to find out where the error crept in on the above code that You found..

"" If we need to use the 11/45 microcode for anything in the future (e.g. blowing PROMs) then it would make sense to spit the files out of my EXCEL spreadsheet in the format we want them so we know they are consistent rather than performing activities by hand on them. ""

That is what has been attempted with my sending You the Exxx_256x4.bin for You to check..

"" Yes, DLL's were always a pain in Windoze. I seem to remember you needed to ensure that the DLL's were in a directory that was referenced by the PATH environment variable. ""
But, the Path variable I don't have, I get a straight error message, but I don't know in their code where it is expecting the DLL to be located..

THANK YOU Marty
 
Hi All;

M-Thompson, Thank You for the reference to this Forum.. And also tell Him Welcome Aboard..

I am No Expert, but we do have a few experts..

I would say that Dave and Fritz are our Resident Experts..

I hope He keeps us in touch with what all He is doing.. Another 11/45 Posting, would be Just fine with me..

THANK YOU Marty
 
Experts - we all have 11/45's in various states of working with aspirations to get them working. If that qualifies us as experts...

I'm with Marty - the more the merrier.

Dave
 
Hi All;

I tried one of the proms that I got from Don, (Thank You, Don), and that didn't fix it..

On Monday morning or so I will put all of the Proms in the M8123 and see what the results might be, if I can it it to work or partially to work in the M8123, then I know that there are more problems in the M8103.. If not then I suspect the Proms are / might not be right..

I tried the M8123 and it did the same thing, a no go.. I will need to check out all of the new proms in my Prom reader, just to be sure..

THANK YOU Marty
 
Last edited:
Hi All;

E115, Prom checked out OK..

Next, is E89 and E91, to be checked out..

Ok, E89 and E91 are Ok.. Thanks, Don..

I think I am going to go a slightly different direction for a while.. I am going to get out one of my NoteBooks and work on the mapping of the MotherBoard, I had started that awhile back and then I got side tracked.. And never got back to it..

THANK YOU Marty
 
Last edited:
Hi All;

I ran across one of the sites that I had visited some years ago, while copying files from my Jump sticks to my PC, now that everything seems to be in order, and seemingly fixed..

It is about a guy making a Transistor equivalent of the 4004 and His progress, the Last time I visited it, it looked like it had gone Dark.. But, now he has started it up again.. So here is the link..

http://insanity4004.blogspot.com/2012/06/worlds-first-microprocessor.html

And my thinking is If He can do this, I should be able to make my PDP 11/45 into a working machine, maybe.. Anyway this further inspired me, to keep working on the 11/45..

THANK YOU Marty
 
How are you getting on Marty - it's all gone quiet...

Anyhow - I have now entered all of the 11/45 ROMs into my EXCEL spreadsheet and created columns in HEX for (a) PROM programming and (b) importing into LOGISIM.

So, I have the following in separate worksheets:

MICROCODE ROMS
ALU CONTROL ROM
INSTRUCTION DECODE ROM
CONDITION CODE CONTROL ROM
M9301-YA BOOTSTRAP PROMS

Apart from the MICROCODE worksheet - they are all 'unchecked' at the moment.

I am off to the DEC Legacy Event in the UK next weekend - so I will hopefully check them out whilst I am there plus solder the components onto my KM11 and UA11 PCBs (assuming they arrive from Mouser) and do some more work on the 11/45 backplane slot pinouts. A busy weekend ahead of me...

Dave
 
Hi All;

Dave, Thanks for the news update..

I am waiting to hear from Don, about if He received my updated files and will be making the Proms for me..

And once I get them back, and even now trying to decide what to do next.. Alot depends on whether the new Proms are able to make the 11/45 work.. If not my thoughts are about making a wire-wrap board to Debug the Microcode board (M8103)..

"" and do some more work on the 11/45 backplane slot pinouts. ""

Many of us will be anxious to see that and if what You have agrees with what I have..

THANK YOU Marty
 
I received the files and have converted them from 128x8 binary images to 256x4 binary for programming.
Here is the list of files with the hex 16b hex checksum inserted in the name (ie, E89 checksum is 0x0880).
I will program them in the next day or so and ship them out Tuesday (Monday is a US holiday).

Don

Code:
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E89-0880_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E90-07C6_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E91-025B_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E92-0261_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E97-028A_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E98-061A_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E99-02FD_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E100-0425_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E105-0258_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E106-0419_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E107-00C6_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E108-01C7_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E113-0119_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E114-027E_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E115-02C1_256x4.bin
-rw-r--r-- 1 Don None 256 Oct  7 11:41 E116-032F_256x4.bin
 
Last edited:
Hi All;

THANK YOU Very Much, Don..

Al Kossow, I have sent You the Files to Put on Your Site..

I have taken the Wire-Wrap Board that I had intended to use for the PDP 8i Micro-coded board, and made a few changes to add another two 28 pin sockets and a couple of 20 pin sockets, so that I can (for now) use it to debug the PDP 11/45, M8103 with lots of Led's and using the EEproms instead of the proms that were used in the M8103..
And doing a comparison, as it cycles between the two boards, at least that is the Idea.. More, later..

THANK YOU Marty
 
Last edited:
I will add the same checksum to my EXCEL spreadsheet now I have something to compare them against.

I assume you are just using a simple 16-bit unsigned sum?

Dave
 
Back
Top