Running a soft cartridge on a real TI

[awesp]

Dear all,

 

Would there be a way of running the programs stored on a cartridge image file (.bin) on a real TI? In other words, has anyone written a utility that will allow a TI with a memory upgrade to load a soft cartridge into RAM and run it?

 

Any ideas, or suggestions will be greatly appreciated,

 

Adam

[+Ksarul]

There are several possible answers to this problem. A program called YLOAD is one method of getting a soft copy of a cartridge to work, but it is not a universal solution. The biggest issue with TI cartridges is that some of them are actually larger than the total available memorty on the machine. A cartridge can have up to 5 GROM chips with 6K each (30K) and normally, up to two ROM chips with 8K each, but TI switched only half of each chip in their bank switching routine (4K at a time), so the ROM side generally topped out at 12K. Many cartridges use less than this maximum though, so YLOAD will generally work with them.

 

There are a couple of other routines as well, I just don't recall their names at the moment (others will chime in here). The other option you have is to get a GRAM device (GRAM Kracker, P-GRAM, GRAMulator, HSGPL, Mechatronics GRAM-Karte, Wiesbaden Supermodul, and a few others) to provide a simulated cartridge space that doesn't eat up the system memory. These options run almost all TI cartridges, although there are often issues with cartridges for the MBX, as they hijack the power-up routine of the console, and not all GRAM devices will allow that to happen.

[marc.hull]

Dear all,

 

Would there be a way of running the programs stored on a cartridge image file (.bin) on a real TI? In other words, has anyone written a utility that will allow a TI with a memory upgrade to load a soft cartridge into RAM and run it?

 

Any ideas, or suggestions will be greatly appreciated,

 

Adam

 

Some (not sure how many) carts with ROM and GROM exist in a runnable state in files that can be loaded into the Mexp.

 

Alternatively the 32K16 board expansion available from SWIM and designed by the SNUG group (or Michael Becker, not really sure which) allows memory to be mapped into cart space. You could dump an 8K ROM to disk and load it into this space. Probably a longer way around than you are looking for.

[RXB]

There are several possible answers to this problem. A program called YLOAD is one method of getting a soft copy of a cartridge to work, but it is not a universal solution. The biggest issue with TI cartridges is that some of them are actually larger than the total available memorty on the machine. A cartridge can have up to 5 GROM chips with 6K each (30K) and normally, up to two ROM chips with 8K each, but TI switched only half of each chip in their bank switching routine (4K at a time), so the ROM side generally topped out at 12K. Many cartridges use less than this maximum though, so YLOAD will generally work with them.

 

There are a couple of other routines as well, I just don't recall their names at the moment (others will chime in here). The other option you have is to get a GRAM device (GRAM Kracker, P-GRAM, GRAMulator, HSGPL, Mechatronics GRAM-Karte, Wiesbaden Supermodul, and a few others) to provide a simulated cartridge space that doesn't eat up the system memory. These options run almost all TI cartridges, although there are often issues with cartridges for the MBX, as they hijack the power-up routine of the console, and not all GRAM devices will allow that to happen.

 

The program in question is already done. GRAM PACKER stores EA5 or EA3 or XB programs in GRAM then runs them from any Cart. EA5, EA4 or XB and even BASIC.

 

But running it from a PGRAM was a problem as you had to run a EA3 program to unprotect the GRAM before it would write. You could read fine, but storing them was a pain.

 

GRAM PACKER worked better from a GRAMULATOR or PGRAM or HSGPL card according to Bud Mills.

Edited July 16, 2013 by RXB

[RobertLM78]

I'm also interested in something like this - I'd like to write over the GROMs on an extra cart with a cart in the form of a .TICart file (if that's possible ). I transferred the (SMU) Circuit Analysis disks to the TI today only to find I need a command module to run any of the programs .

[RXB]

I never made a copy of GRAMPACKER so it should still be on my TI hard drive somewhere. It should be on the FTP site.

[Manic1975]

There are several possible answers to this problem. A program called YLOAD is one method of getting a soft copy of a cartridge to work, but it is not a universal solution. The biggest issue with TI cartridges is that some of them are actually larger than the total available memorty on the machine. A cartridge can have up to 5 GROM chips with 6K each (30K) and normally, up to two ROM chips with 8K each, but TI switched only half of each chip in their bank switching routine (4K at a time), so the ROM side generally topped out at 12K. Many cartridges use less than this maximum though, so YLOAD will generally work with them.

 

There are a couple of other routines as well, I just don't recall their names at the moment (others will chime in here). The other option you have is to get a GRAM device (GRAM Kracker, P-GRAM, GRAMulator, HSGPL, Mechatronics GRAM-Karte, Wiesbaden Supermodul, and a few others) to provide a simulated cartridge space that doesn't eat up the system memory. These options run almost all TI cartridges, although there are often issues with cartridges for the MBX, as they hijack the power-up routine of the console, and not all GRAM devices will allow that to happen.

 

Please can you put YLOAD for download?

[RXB]

MBX just uses the Power Up header in GROM like RXB did, but people complained that it always went to the RXB load routine, so I put in a optional XB program to load it or not.

You can take out the power up routine in the MBX and make it behave normally but it has to execute the Power Up first before it does the normal stuff.

 

The GRAMULATOR would run ANY cartridge including the MBX.

 

By the way TI carts from TI used 6K chips instead of 8K chips so that is why they are only 32K in size while Miller Graphics set the standard of 40K per cart by using 8K for the GRAMKRACKER.

This fixed two issues the GRAM device had:

1. When fetching data from a TI cart it hit the end of 6K (6144 bytes) then start fetching data from the next area, thus skipping the 2K area not there. (Bad and odd design concept.)

2. When Miller Graphics made carts 8K (8192 bytes) the boundaries were the same as RAM or VDP and went up from 32K GRAM to 40K GRAM, so had more memory then the 32K in the TI RAM.

 

A bonus was the TI can access 640K or 16 banks of 40K Carts thus this was at the time the most memory you would get access to in a TI for quite awhile (7 years).

 

By the way I wrote a SWGR routine that would switch GRAM banks and RTGR back to the one calling it, but used VDP address >3FE0 to do this (16 bytes) as a buffer to switch banks.

Edited July 16, 2013 by RXB

[awesp]

Thanks Ksarul, marc.hull, RXB and RobertLM78 for your prompt suggestions and comments.

 

Manic1975, I have a copy of YLOAD, but am not sure where I got it from. Maybe I could email it to you?

[+Ksarul]

I'm not sure I actually have a copy of YLOAD, Manic1975--I use the various GRAM devices I listed above on my systems--either those or my pizza-box GSIM device from TI, which is the original GRAM device (they also built EGROM boxes and cartridges, and the CEC9914 cartridge emulators, all of which let regular ROMs work like GROM chips using the support circuitry included in the devices). I have at least one of each of these items. . .all in different hardware configurations, which is why I have more than one of some of them.