+acadiel Posted December 6, 2017 Share Posted December 6, 2017 Atrax reminded me of the 64K cartridge with EA and Memo Processor in it. I don't remember how I dumped it, but pictures of it are here: http://hexbus.com/TI-99_4A_Home_Computer_Page/Compact_Computer_40.html#25 http://hexbus.com/TI-99_4A_Home_Computer_Page/Compact_Computer_40.html#26 I think there was a bank switch or something that I did before I dumped each 32K memory space. cc40_ea_bank_0-5000-cfff cc40_ea_bank_1-5000-cfff The only thing I can guess is that when Mechatronics built this cart, they did something with the paging bit (we'd have to check with JGardner on VCFed) and that RUN MP switches it to the 32K of the cart with Memo Processor and RUN ALDS switches it to the 32K of the cart with Editor Assembler. I *think* how I dumped it is that I switched each 32K to the appropriate side by running the command for MP or EA and then exited back to BASIC and did a PEEK. It stayed in the bank of the last run program. Also, maybe check with HAP who wrote the MESS emulation for it, because it's supposed to work as a 64K bank switched cart there. Ksarul is in possession of this cartridge now, and can map where the A15 line went to. Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 6, 2017 Share Posted December 6, 2017 Hey, is anyone from SNUG reading this thread? I'd like to see if we might be able to get some of Michael Becker's Hexbus schematics and such over to Jim Brain. Display adapter, etc. A lot of that information is unobtainable. Quote Link to comment Share on other sites More sharing options...
brain Posted December 6, 2017 Share Posted December 6, 2017 Atrax reminded me of the 64K cartridge with EA and Memo Processor in it. I don't remember how I dumped it, but pictures of it are here: http://hexbus.com/TI-99_4A_Home_Computer_Page/Compact_Computer_40.html#25 http://hexbus.com/TI-99_4A_Home_Computer_Page/Compact_Computer_40.html#26 I think there was a bank switch or something that I did before I dumped each 32K memory space. cc40_ea_bank_0-5000-cfff cc40_ea_bank_1-5000-cfff The only thing I can guess is that when Mechatronics built this cart, they did something with the paging bit (we'd have to check with JGardner on VCFed) and that RUN MP switches it to the 32K of the cart with Memo Processor and RUN ALDS switches it to the 32K of the cart with Editor Assembler. I *think* how I dumped it is that I switched each 32K to the appropriate side by running the command for MP or EA and then exited back to BASIC and did a PEEK. It stayed in the bank of the last run program. Also, maybe check with HAP who wrote the MESS emulation for it, because it's supposed to work as a 64K bank switched cart there. Ksarul is in possession of this cartridge now, and can map where the A15 line went to. PORT25.2. I can see the pad connected to something (3rd edge pad from bottom, in top view). I still am not sure if that was TI-sanctioned, or just something this company did. It'd be nice to know. Jim Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 6, 2017 Share Posted December 6, 2017 PORT25.2. I can see the pad connected to something (3rd edge pad from bottom, in top view). I still am not sure if that was TI-sanctioned, or just something this company did. It'd be nice to know. Jim Hopefully pins 3 and 17, we can figure out how the system toggles them to support paging up to 32K x 4. That would make for some additional fun stuff if we can combine cartridges and the ROM actually scans for them! Sent from my Moto Z (2) using Tapatalk Quote Link to comment Share on other sites More sharing options...
brain Posted December 6, 2017 Share Posted December 6, 2017 I am not concerned with paging. I can support it just fine with the new cart design I have. The question is who can write code to support these things? :-) Jim Quote Link to comment Share on other sites More sharing options...
+helocast Posted December 6, 2017 Share Posted December 6, 2017 Mostly just playing around tonight with the ram cart stuffed with a Dallas DS1250Y (its what I had available from new 99/4A mobo design/project - $7 on eBay's not going to break the bank). Starting with one of my CC40s that I already added 16KB internally > FRE(0)=18009 Issuing CALL ADDMEM to acquire the 32KB cart > FRE(0)=34414 @ each dipswitch bank setting Poke and Peek in a short program @ start of cart space and @ end of cart space to see if values were retained. Thanks again, Jim! 1 Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 6, 2017 Share Posted December 6, 2017 Mostly just playing around tonight with the ram cart stuffed with a Dallas DS1250Y (its what I had available from new 99/4A mobo design/project - $7 on eBay's not going to break the bank). Starting with one of my CC40s that I already added 16KB internally > FRE(0)=18009 Issuing CALL ADDMEM to acquire the 32KB cart > FRE(0)=34414 @ each dipswitch bank setting Poke and Peek in a short program @ start of cart space and @ end of cart space to see if values were retained. Thanks again, Jim! That’s my experience as well with my cart. You only get 16K added, for a 34K system. I wonder if it's just using >5000->8FFF at that point and the system ROM just ignores the rest (>9000->CFFF), so you can still poke to it to your heart's content. Also, good call on the DS with the built in battery. Will try that as well with my prototypes. 2 Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 6, 2017 Share Posted December 6, 2017 Impatiently waits for RAM and EEPROMs to come in :-) 2 Quote Link to comment Share on other sites More sharing options...
sparkdrummer Posted December 7, 2017 Author Share Posted December 7, 2017 Is this latest and greatest CC40 stuff going to be available in a fully assembled form? I'd love to have one of the carts and the device that allows loading and saving files. I've wanted these things for a long time, that's why I started this thread. Quote Link to comment Share on other sites More sharing options...
brain Posted December 7, 2017 Share Posted December 7, 2017 Is this latest and greatest CC40 stuff going to be available in a fully assembled form? I'd love to have one of the carts and the device that allows loading and saving files. I've wanted these things for a long time, that's why I started this thread. The answer will be yes, but what form the products take is still in flux. The SD drive needs some more work, and a production resizing to be of most benefit, and the RAM/ROM carts need to be tested (I already know there is more functionality that should be added) before we decide on final form. JIm 3 Quote Link to comment Share on other sites More sharing options...
brain Posted December 7, 2017 Share Posted December 7, 2017 If folks missed out on the first board run, I am happy to open a second one for the initial three boards. Let me know if there is sufficient interest. The new boards will differ: unneeded diodes removed from ROM design 6K8 resistor on RAM board rerouted (not sure it is needed, but I'll leave the footprint there) 2 missing lines on the extender were added option to use 27c040 or sst39sf040 ROMs I may make one more change Route 2 "banking" signals to ROM cart (using 2 lowest bank address pins) 3 Quote Link to comment Share on other sites More sharing options...
sparkdrummer Posted December 7, 2017 Author Share Posted December 7, 2017 brain (Jim) - This is absolutely great, I can see many CC40's being put into use. Quote Link to comment Share on other sites More sharing options...
ElectricLab Posted December 7, 2017 Share Posted December 7, 2017 If folks missed out on the first board run, I am happy to open a second one for the initial three boards. Let me know if there is sufficient interest. The new boards will differ: unneeded diodes removed from ROM design 6K8 resistor on RAM board rerouted (not sure it is needed, but I'll leave the footprint there) 2 missing lines on the extender were added option to use 27c040 or sst39sf040 ROMs I may make one more change Route 2 "banking" signals to ROM cart (using 2 lowest bank address pins) I'm a little late to the game, but I'd be interested in one. I picked up a CC40 last year and would love to put it to use! Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 7, 2017 Share Posted December 7, 2017 If folks missed out on the first board run, I am happy to open a second one for the initial three boards. Let me know if there is sufficient interest. The new boards will differ: unneeded diodes removed from ROM design 6K8 resistor on RAM board rerouted (not sure it is needed, but I'll leave the footprint there) 2 missing lines on the extender were added option to use 27c040 or sst39sf040 ROMs I may make one more change Route 2 "banking" signals to ROM cart (using 2 lowest bank address pins) I'll test the new ones too if you like. Just got RAM chips in and will be testing those carts later on. I also ordered some of the DS backed up chips on eBay to try those out too. Ksarul, think it's time I help you dump the CC-40+ bios now. Sent from my Moto Z (2) using Tapatalk 1 Quote Link to comment Share on other sites More sharing options...
TI998owner Posted December 7, 2017 Share Posted December 7, 2017 Hi Jim, , The new Rom/ Ram board looks very interesting. Also I would be interested in the second board Order. I will PM you later on amount I would Like. , Need a little help today, From Jon Guidry of HEXBUS.com , I have been working on a Editor /Assembler cart copy board of the Mechatronics E/a cart on your site. Until the files are available for Jim ROM board. , Need to know the correct pin connections for PIN # 20 and connections for PIN # 22, where they connect to The Cart edge connector and signal name for both. Also if you could check the resistor value. To see If 3K3 is correct value. Could not tell from picture. , Thanks Jim and Jon if you are reading this Lee TI998owner Quote Link to comment Share on other sites More sharing options...
brain Posted December 8, 2017 Share Posted December 8, 2017 Hmm, Jon is acadiel, is he not? Pin 20 is Vplug (or Vswitched, on some schematics) Pin 22 is A6 3K3 would be typical on both other carts. Jim Quote Link to comment Share on other sites More sharing options...
brain Posted December 8, 2017 Share Posted December 8, 2017 I'm a little late to the game, but I'd be interested in one. I picked up a CC40 last year and would love to put it to use! I turned the order form back on, so folks can pre-order. http://store.go4retro.com/ti-compact-computer-40-prototype-boards/ Jim 1 Quote Link to comment Share on other sites More sharing options...
TI998owner Posted December 8, 2017 Share Posted December 8, 2017 Hi Jim, The pin 22 is listed as GVPP on the 27C512 The pin 20 is listed as E ( with line over e) on the 27C512 , The problem is the picture of the Mechatronics E/A cart Is not clear enough to trace each of these pin,s. Hopeful Jon could check the cart. , Yes , Jon is Acadiel Thanks Lee TI998owner Quote Link to comment Share on other sites More sharing options...
brain Posted December 8, 2017 Share Posted December 8, 2017 I routed 25.2 and 25.3 to A15 and A16, configurable with a jumper I also added support for 28 pin EPROMs 1 Quote Link to comment Share on other sites More sharing options...
brain Posted December 8, 2017 Share Posted December 8, 2017 (edited) Hi Jim, The pin 22 is listed as GVPP on the 27C512 The pin 20 is listed as E ( with line over e) on the 27C512 , The problem is the picture of the Mechatronics E/A cart Is not clear enough to trace each of these pin,s. Hopeful Jon could check the cart. , Yes , Jon is Acadiel Thanks Lee TI998owner Ah, you wanted to know the pins on the EPROM. Here they are: Pin 20 is !OE (and is connected to the !CS lines Pin 22 is !CS, and is connected to GND The layout is the same as mine, with the following changes P25.2 (Cc40 pin 17 is connected to A15) Pin 28 is connected to Vcc. If you install the 28 pin EPROM at the pins 3-30 of the 32 pin socket and wire pin 3 (of 32) to P25.2 (cc40 pin 17) and pin 30 (of 32) to pin 32 (of 32), you have it. Jim Edited December 8, 2017 by brain Quote Link to comment Share on other sites More sharing options...
TI998owner Posted December 8, 2017 Share Posted December 8, 2017 Hi Jim, I will make the changes and test the EPROM again Bob, sent me Thank you for the help Lee TI998owner Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 9, 2017 Share Posted December 9, 2017 I got my SRAMs and EPROMs in today, so will be testing the EPROM image this weekend. 2 Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 9, 2017 Share Posted December 9, 2017 I turned the order form back on, so folks can pre-order. http://store.go4retro.com/ti-compact-computer-40-prototype-boards/ Jim Are these the rev 2 ones? Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 9, 2017 Share Posted December 9, 2017 Making good progress. RAM: Memo Processor PoC: 3 Quote Link to comment Share on other sites More sharing options...
+acadiel Posted December 9, 2017 Share Posted December 9, 2017 For posterity, the reason dumped cart images differ depending on method is due to the weird way TI put the ROMs in the memory map. Obviously, dumping a ROM via programmer/reader will yield the correct ROM image. If you dump from BASIC or assembly, look at the memory map for guidance. The 32kB external space starts at >5000. For purposes of this discussion, you can ignore the bottom 3 hex digits, so we're starting at "page" 5 in memory, or 0101 in binary 0101 :memory page 5, and we consider this page 0 of the image 0110 :6/1 0111 :7/2 1000 :8/3 1001 :9/4 1010 :A/5 1011 :B/6 1100 :C/7 Now, understand that the top address bit (bit a15) is not connected to the ROM. Thus, on the above addresses, skip the first digit: 101 :5/0 110 :6/1 111 :7/2 000 :page 8 in cc40 memory map and page 3 in our image memory map , but to the EPROM, it looks like page 0 001 :1/4 010 :2/5 011 :3/6 100 :4/7 Thus ROM image address >0000 will sit at position >3000 in the MEMORY map relative to that start of the image, which is >5000, or >8000 in the cc40 memory map Thus, to convert an image read by a program to one read by a ROM reader, you have to do the following calculation: ROM dump address = MEMORY dump address - 0x3000 mod 0x8000 or ROM dump address = MEMORY dump address + 0x5000 mod 0x8000 or MEMORY dump address = ROM dump address + 0x3000 mod 0x8000 Since 0x3000 = 3 * 4096, or 12K, that's the 12K offset Acadiel noticed in his memory dump data. Not sure if TI used this as a weak copy protection scheme (can't think of any better reason to do this at present), or something else, but it just makes things ever so slightly more weird in the memory map. A visual aid (as I cut the 12K off the top of my memory dumps and paste them at the end so they work): 1 Quote Link to comment Share on other sites More sharing options...
Like us on Facebook
Follow us on Twitter
AtariAge on Instagram
Watch us on YouTube
Chat on Discord
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.