Smart electronic dudes here

[Gazoo]

I think he's trying to get an alternate version of the ROMs to run with the XB GROMs, but that requires that the original ROMs are disabled. Since he's going to be providing an alternative to the disabled ROMs, that eliminates your legitimate programming issue, Rich. But as the real issue was to disable the ROMs so that he could perform the test, the programming issue is actually moot.

 

Please look at those purported XB source disks, Gazoo! You have my curiosity piqued too!

 

 

Found the source disks for the Groms, and a bunch of other good stuff, still digging. Tomorrow is another day.

 

Gazoo

[+Ksarul]

Thanks for hunting, Gazoo! Finding documentation and code thought to be lost is really exciting for me!

[+InsaneMultitasker]

 

Yea, I've found a couple of floppies recently labeled XB rom source code, but haven' tried to read 'em yet.

Too many actually important irons in the fire!

 

Go away, pest!

 

Gazoo

Digging through files is somewhat contagious. So... just to be annoying, I uploaded a few ZIP files to the off-topic thread. <grin>

[+Ksarul]

There is a strong possibility that the TI Writer version of the 1979 GPL Manual you have is originally from me. I was one of two people that went to the trouble of typing the entire thing into TI Writer back in the day because most of the copies in circulation were pretty poor photocopies. Mine circulated mostly in Europe, although I also let copies of it loose in the US. The other one originated somewhere in the US. I once knew who had typed that one in, but the name escapes me now.

 

One of these days I'll have to finish typing in the source for the BASIC Support Module, as the copy of that in my collection is not good enough to make a reasonably legible PDF.

[RXB]

There is a strong possibility that the TI Writer version of the 1979 GPL Manual you have is originally from me. I was one of two people that went to the trouble of typing the entire thing into TI Writer back in the day because most of the copies in circulation were pretty poor photocopies. Mine circulated mostly in Europe, although I also let copies of it loose in the US. The other one originated somewhere in the US. I once knew who had typed that one in, but the name escapes me now.

 

One of these days I'll have to finish typing in the source for the BASIC Support Module, as the copy of that in my collection is not good enough to make a reasonably legible PDF.

My copy is the entire manual also too.

 

I was the first person to buy them from Jim Leasure as he charged me $300 or that and the XB source code books included.

 

Very few people knew how to write GPL of what it was. Let alone be interested in learning that language.

Edited July 3, 2014 by RXB

[Gazoo]

It was actually much easier to do this than I expected. All that's required is to take the widget apart and temporarily tack on 2 jumper wires.

The widget's switch actually is 2 pole switch, one switches the Grom line and the other switches the Rom line. No hacking up of carts.

The pic below shows the 2 jumpers, they make the 3rd socket the Grom-only socket and the 2nd socket the Rom-only socket.

 

 

My test worked as expected.

 

First plugging in only the XB cart to socket 3 showed XB on the menu, since the Grom was recognized. As expected, selecting XB resulted in a colorful screen since there was no Rom present.

 

Adding the 512k Rom cart to the second socket caused everything to work normally. XB ran and loaded programs as expected.

The 512k Rom contains the 16k of XB Rom repeated throughout the entire 512k space.

 

The next step will be to put only the 16k of XB Rom at the beginning of the 512k Rom space and see if that works.

 

If that works, the idea is to put a menu program in the last 8k of the 512k Rom and see if I get both choices on the TI menu screen, since the 512k Rom apparently powers up in the last 8K.

 

Stay tuned.

 

Gazoo

 

[Gazoo]

Well things sort of worked. With the 16k XB Roms only at the beginning of the 512k, XB didn't work. I guess XB doesn't write to c>6000 before accessing Rom.

But I was able to put a header and short program in the last 8k of the 512k Rom which shows up on the TI menu. The short program executes from >8300 and writes to >6000 to bank in the first Rom bank, then quits. The header in Rom no longer shows up on the TI menu screen after the reset, but XB works correctly then.

 

The end result is that a menu program could be written for this setup, but the user would have to choose it from the TI menu screen and NOT choose XB. XB could then be part of the menu program, setting the correct Rom bank and then using a GPLLNK to enter it (XB).

 

Not sure if I'm going to pursue this or not, as there doesn't seem to be a way to eliminate the XB from the TI menu on startup and have everything work.

 

Gazoo

Edited July 3, 2014 by Gazoo

[RXB]

Sorry to hear that. Did you find the XB ROM source as a re-write of them could also solve your problem.

 

When XB starts up it could have a mod so you can select your ROM instead.

Elimination of VGWRITE and VGREAD are not really needed in XB ROMs and could be taken out as superfluous.

Others include GREAD and GWRITE as these are only there by accident as the notes showed the Assembly team never talked to the GPL team much.

GPL has MOVE and is the same speed as these routines in the ROMs, funny thing is the ROM ones use more memory to access then the GPL MOVE.

 

When I had the XB ROM source I had planned on fixing the problems in these XB ROMs, but since my Hard Drive died I got screwed.

 

Gazoo please let me know if you find them. This would be great for your project or others too.

Edited July 3, 2014 by RXB