PGram+ Card --- Loading up the DSR

[Gary from OPA]

Well, I got my PGram+ card working, replaced the battery and such, but of course the DSR is ram-based, so now its dead.

 

I went thru the floppies i have and sadly none of them have the copy of the DSR on it to load onto the card.

 

I see this link: https://ftp.whtech.com/utilities/Pgramy2k.ark

on WHTECH that should be it, right?

 

Now, I am just waiting on my 25 foot rs232 null modem cable to i can connect my ti99 to my pc, as that is for now the only way i will have to transfer over files, that should arrive from amazon by sunday.

 

Anything else am i missing? Does anyone have a diskette image of the original pgram disks?

 

Thanks.

[+arcadeshopper]

yes that's it

 

should get a tipi

[Gary from OPA]

38 minutes ago, arcadeshopper said:

yes that's it

 

should get a tipi

Yes, in the cards once I got enough extra funds as there a few other things I need from your shop along with the tipi, plus I need to get the raspberry itself to go with it, too bad the one I ordered before from you is stuck down in DR.

[Gary from OPA]

Well, back to this problem.

 

I finally have a way to transfer files from my PC to my TI99 and got everything working, but well PGRAM card the archives are damaged, the pgram2k.ark at 49 sectors unpacks fine, that is good. but that is only the DSR, you need the loader UTIL1 and that is in the pgsrc2k.ark which is 36 sectors, i downloaded it before from whtech, and it catalogs the archive file, but it only extracts the first few files and then bang, locks up the archiver big time, even if you try to extract just the UTIL1 it does the same.

 

So I thought, the file is too small no way its only 36 sectors, so went back to whtech and thought I would redownload.

 

guess what its even worse, the files i downloaded before, but ark are even smaller and totally corrupted now.

 

so that i am screwed, anyone have a working archive?

[atrax27407]

See if this helps.

PGRAM-Card-001.zip

[Gary from OPA]

19 hours ago, atrax27407 said:

See if this helps.

PGRAM-Card-001.zip 9.87 MB · 7 downloads

i going to try this soon, finally awake on this "family day" holiday here in canada, and having a coffee. i will let you know if it helps.

 

i still want to get the source code as well, but so far all the archives are bad. also my card is just the normal pgram not the plus version. i going to look at upgrading it at some point as well.

 

EDIT: hmm, it seems most of the source is in the files you sent me thanks. -- I wonder if there is info on how to do the pgram to pgram+ mod, there a picture on the conversion, alot of wires to add, and 3 chips and the bigger ram, but i don't see the step-by-step way on how to do it yourself.

Edited February 19 by Gary from OPA
more info

[Gary from OPA]

I been searching but I can't find any utility on the PC that can convert BACK things to the Gram/Kracker format that PGRAM and other gram devices used to use on the real-iron TI99 hardware. I lost all my archives that were in GK format on 5.25" floppies, so now since I want to properly load modules on my TI99 using the PGRAM code, I need the old format, and everything archived online now is sadly in the flashrom or finalrom or rpk mame format, which is great for those using finalrom (I have one as well), but useless for those that want to use the original gram stuff.

 

there has to be utility that can convert BACK, I just at a lost finding it. well i guess I could rewrite the Pgram loader to handle the newer format, that might be quicker way to do it.

[apersson850]

A bit funny that in the zip-file posted by atrax27407 there's a document from my friend Jan Alexandersson. He has given up on his TI 99/4A due to old age (his, not the computers). He donated his stuff to somebody who's still playing with these things, so I have his P-GRAM card now...

Not in use, though, as my box is full with cards already.

Edited February 19 by apersson850

[Gary from OPA]

1 hour ago, apersson850 said:

A bit funny that in the zip-file posted by atrax27407 there's a document from my friend Jan Alexandersson. He has given up on his TI 99/4A due to old age (his, not the computers). He donated his stuff to somebody who's still playing with these things, so I have his P-GRAM card now...

Not in use, though, as my box is full with cards already.

at the rate it is going the ti99 will outlive us humans, now if only we could download ourself into the PEB as a card. -- future project to work on someday before my time comes. Best wishes to your friend in his golden years. I know the feeling i have two peb's fully stuffed now, and still have more cards to install, so hopefully soon I can grab another flexcable card, and the peb joiner, and still filling up a second set of cards in my other peb's.

[Gary from OPA]

trying to look now at how i can mod the pgram to handle more than two rom banks, so i can use the RXB2024 on it, and other modules that have more than 16k of rom.

 

did anyone ever in the past do a mod on pgram to handle 24k or 32k of rom at least.

 

should be possible to wire in 378 decoder and another memory chip for more rom banks, and then upload the source to handle it.

 

i wonder if possible also to add support for the dataobitics >0400 cru base paging of >6000 space.

 

@RXB- what way is your 3 way rom pages working, how do you swap pages, -- i need to read up on 378 banking expansion.

Edited February 20 by Gary from OPA
add tag to rich

[RXB]

36 minutes ago, Gary from OPA said:

trying to look now at how i can mod the pgram to handle more than two rom banks, so i can use the RXB2024 on it, and other modules that have more than 16k of rom.

 

did anyone ever in the past do a mod on pgram to handle 24k or 32k of rom at least.

 

should be possible to wire in 378 decoder and another memory chip for more rom banks, and then upload the source to handle it.

 

i wonder if possible also to add support for the dataobitics >0400 cru base paging of >6000 space.

 

@RXB- what way is your 3 way rom pages working, how do you swap pages, -- i need to read up on 378 banking expansion.

Basically does kind of like the XB module does and writes to >6000 or >6002 for ROM 0 and 1

In Classic99 or FinalGROM I write to >6004 to get ROM 2

 

I was told it works on the HSGPL Card too. (Up to 4 ROMs)

[Gary from OPA]

15 minutes ago, RXB said:

Basically does kind of like the XB module does and writes to >6000 or >6002 for ROM 0 and 1

In Classic99 or FinalGROM I write to >6004 to get ROM 2

 

I was told it works on the HSGPL Card too. (Up to 4 ROMs)

Thanks. Do you think you ever go to more than 4 8k roms for RXB?

[+Ksarul]

Note that the card doesn't have the switching circuitry for more than two RAM banks, but that it may be possible to add them with some potentially extensive surgery. I was given permission to make a new run of the cards, but I haven't had time to finish the new layout yet. I'll have to take a close look at it to see if it can be done without a major board redesign (and if it goes to a 378, it would actually be possible to expand the ROM side to 512K, or 64 Banks--that would make it fully compatible with a lot of the homebrew cartridge images out there).

 

 

[Gary from OPA]

10 minutes ago, Ksarul said:

Note that the card doesn't have the switching circuitry for more than two RAM banks, but that it may be possible to add them with some potentially extensive surgery. I was given permission to make a new run of the cards, but I haven't had time to finish the new layout yet. I'll have to take a close look at it to see if it can be done without a major board redesign (and if it goes to a 378, it would actually be possible to expand the ROM side to 512K, or 64 Banks--that would make it fully compatible with a lot of the homebrew cartridge images out there).

 

 

Yes it would. I have to mod first from my PGRAM to pgram+ I guess you be doing the plus version. You wouldn't happen to have the original on paper mod notes that did the conversion from regular to plus. The good thing is the board has a space in the top right corner to add some chips to do mods.

[RXB]

1 hour ago, Gary from OPA said:

Thanks. Do you think you ever go to more than 4 8k roms for RXB?

Yea down the line if I finish what I am doing will be forced to keep adding more ROM banks as I turn GPL into Assembly.

[apersson850]

9 hours ago, Gary from OPA said:

I wonder if possible also to add support for the dataobitics >0400 cru base paging of >6000 space.

Interesting to see that they selected the same CRU base as I did for banking my internal 64 K expansion. At that time I wasn't aware of anybody else using CRU >0400.

[Gary from OPA]

2 hours ago, apersson850 said:

Interesting to see that they selected the same CRU base as I did for banking my internal 64 K expansion. At that time I wasn't aware of anybody else using CRU >0400.

Yes, if you ever get hold of their dataobitics superspace ii or another one of their cartridges that had like 32k or 64k in I think console writer ii does it would be interesting to see the effect of it using 0400 along with your 64k mod.

 

If I remember right I think I used 0B00 for my pop-cart to control banking. There is not many devices that used below non dsr space. I forgot what Ubergrom uses I think it had cru base as well.

[apersson850]

>0B00 is probably more safe, since it's a more odd address to decode. >0400, >0800 and >0C00 are the first candidates, since >0000 is occupied in the console and >1000 is the beginning of DSR space.

[Gary from OPA]

I did some surgery on my PGRAM card today. I only had the original model, not the PGRAM+ model, which allows for 4 banks of GROM.

 

After looking at various photos, and schematics, and manuals, I finally figured out how to do the upgrade, it is dirt simple, since the original notes on the upgrade seem to been long gone.

 

STEP #1 - Remove the middle 8k ram chip

STEP #2 - Transfer the left 32k ram chip into the now empty 8k middle position, but when doing so bend out pin #26 the old CS2 on 8k, but now will be A13 once we run some wires.

STEP #3 - Install the 128k chip into the left 32k socket, you will have of course 4 pins hanging over, bend those out so you can solder some wires.

 

STEP #4 - Run the A13 wire to pin 26 on the middle now 32k chip, and continue onto pin 31 on the new 128k chip in the left socket.

STEP #5 - Run the A12 wire to pin 1 on the middle now 32k chip, and continue onto pin 2 on the new 128k in the left socket.

STEP #6 - Connect pin 32 of the new 128k chip to VBattery

 

That's it folks, you have now upgraded your PGRAM to PGRAM+, now all you need to do is load the PG+ DSR files, which I finally found as well, so you can select which bank and load your GROMs up.

 

To find the spots for A13 and A12 that is easy they are accessible between the middle and right sockets, the two traces coming out from pin 7 and 8 from the right 32k chip, which if you notice from the photos there is line of 8 via's, A12 is the 5th via down counting from the top, and A13 is the 6th via down counting from the top, see my attached photo of the completed upgrade.

 

Now comes the next hard part, turning the PGRAM+ into PGRAM-SuperSizeMe expanding the banks from 4 to 8, and also expanding the rom banks from the limit of only 2 to having more pages, that will a bunch more extra wiring and installing some chips as well, which can be done, as in the top-right corner of the peb card, there is room for some extra 74ls chips, i will work on designing up that mod next.

 

Edited July 27 by Gary from OPA
trying to add a video to my post...

[+jedimatt42]

It looks like you pin 1 of the 128k chip is now floating. I would expect that to be an address line and need to be held in a single state for stability. ( Edit: I'm wrong see following posts )

[Gary from OPA]

37 minutes ago, jedimatt42 said:

It looks like you pin 1 of the 128k chip is now floating. I would expect that to be an address line and need to be held in a single state for stability.

Nope, it's not used on the 128k chip, here is the pinouts for 8k / 32k / 128k. Don't have any 512k chips they hard to get these days.

[+Ksarul]

2 hours ago, Gary from OPA said:

Nope, it's not used on the 128k chip, here is the pinouts for 8k / 32k / 128k. Don't have any 512k chips they hard to get these days.

Here's a link to the 512K x 8 SRAM chips (Mouser). Both are Alliance parts. I use these on the HRD4000B and the SAMS boards I build.

[Gary from OPA]

45 minutes ago, Ksarul said:

Here's a link to the 512K x 8 SRAM chips (Mouser). Both are Alliance parts. I use these on the HRD4000B and the SAMS boards I build.

thanks i will try to get some in once i have some extra funds, i need to order $150 worth from mouser to get free shipping.

 

I am also working on my SuperSizeMe mod for the pgram, once I have documented, and updated the DSR from the source code, I will relay the info to you, as you might be able to include it on the upcoming new version of PGRAM you slowly have been working on.

[Gary from OPA]

Spent the day trying to get the pgram card to work when you have two PEBs connected together.

 

Sadly, there seems to be too much lag to get stable grom read data when running two PEBs together.

 

The data is not always the same. Even reducing the number of cards and devices connected, still no luck.

 

I tried different firehoses as well. Everything else is ok. All the other cards work. But sadly, not the pgram.

 

I wonder if there a way to fix this issue or speed up access.

 

I guess no one here has much experience in running two PEBs together with one card being a gram device.

 

If so let me know.

 

Thanks.

[Gary from OPA]

Today. After reading alot of old threads about pgram and how bad the timing can be, with the recommendation that faster sram helps, I replaced the middle 32k gram which was a 120ns chip with a newer 128k 55ns matching the already new 128k 55ns on the left socket.

 

And it helped there was last days errors when running the pgram card in a two PEB joined together setup.

 

But still not stable enough to use that way. So faster ram helps, but not enough. I am thinking next to replace the 3 244's and the 245 on the pgram with the HCT or ALS versions. And see if that is enough of a bump to get the speed up.

 

But I going to study the schematic more and see what else can be improved in timing, maybe the 161 counters are not quick enough as well.

 

I wish I had another gram device that operates in the peb like a mechatronics gram krate or a hsgpl to see if it has the same timing issues.

 

I need to find those schematics and compare them to see if there is less decode logic on those designs.