Game-Tech.US Posted November 5, 2010 Share Posted November 5, 2010 I started a thread earlier wanting to know which ic in the gemini was the TIA, pretty sure which one it is, but its not the same pinout as an atari TIA. At least I dont think so after soldering up the longhorn board to the pins as if it were a 2600 TIA, no video or audio through the mod except hum and nothing on r/f either. I found no datasheet for the 73192, but this was interesting. These guys are dumping original hardware for emulation, I think that would mean some idea of the ic's pinout, am I wrong? One of the members has a gemini, but has yet to dump it apparently and its been over 3 years... Anyone know if a patent search would turn up pinouts? Unless someone has any info I will have to start looking at 2600 schematics and trying to figure out which pin is which on the gemini, but i'm not in a rush to start that project as I have others to finish first. Quote Link to comment Share on other sites More sharing options...
+5-11under Posted November 5, 2010 Share Posted November 5, 2010 You could also locate a trio of resistors with values that are double/half of each other. That'll get you about 75% there. Quote Link to comment Share on other sites More sharing options...
Rybags Posted November 6, 2010 Share Posted November 6, 2010 That thread talks about "dumping", which you only do with ROMs. The 2600 and Gemini in default configuration don't have any ROMs built into them, and in any case a game dump or BIOS ROM source code wouln't give much clue towards pinouts. Like I said before, the best bet is to check where the traces go - find the resistor ladder and you should find the luma lines. Find connectivity between the address/data bus lines between the IC and 6507 and that's most of the chip done. Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 Right now I'm tracing the pins to their 6507 counterparts. That's working so far, but I have just one question... how do I know for sure that the 6507 pins match up to the ones in an official 2600? If they're not, this information is going to be pretty worthless. Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 Here's a partial pinout (exactly half the pins) of the 73192, the Coleco Gemini TIA chip. I'm not sure how to get the rest, but it's a good start. Quote Link to comment Share on other sites More sharing options...
SoundGammon Posted August 6, 2013 Share Posted August 6, 2013 Would the ColecoVision 2600 adapter have the same chip(s)? Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 Would the ColecoVision 2600 adapter have the same chip(s)? That's the assumption. I presume that the chip in the expansion unit was also the foundation of the Gemini. Unfortunately, I don't have pinouts for THAT either, so... A quick update: putting one end of a multimeter on the color adjustment wheel and the other on pin 10 of Coleco's TIA knock-off caused a beep, so pin 10 is probably CADJ. This is only guesswork, but I assume that means pin 11 is the oscillator, same as it is on Atari's official chip. I've been examining Coleco's layout, and if I were a betting man, I would put money on the three LUM pins being arranged in numerical order. Coleco's engineers went to great pains to keep the A and D pins arranged together, while on the official chip, they're a bit more loose about it. Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 An update to the pinout is provided below. Diligent research and testing has brought the total of pins identified to seventy-five percent. Regrettably, most of the video and audio pins still remain a mystery. However, I did do some voltage tests on pins four through eleven, and came up with these results: PIN 4: Pitfall, Omega Race, and Breakout all registered as 4.5V. PIN 5: Registered 380mv, 245mv, and 173mv, with Pitfall getting the highest rating. PIN 6: Pitfall got 1.03V and Breakout got 1V. Omega Race got 182mv. PIN 7: Pitfall got 1.44V and Breakout got 1V. Omega Race got 183mv. PIN 8: Pitfall got 18mv, Omega Race got 16.2mv, and Breakout got -24mv (these may have all been in the negatives; I should retest to confirm this) PIN 9: Pitfall got 3.5V, Omega Race got 4.76V, and Breakout got 2.87V. PIN 10 (already confirmed as CADJ): All games received 1.8V. PIN 11: Pitfall got 348mv, while Omega Race and Breakout received much higher ratings; 2.52V and 2.44V respectively. Now I'm not an engineer, BUT if I were to take an educated guess, I would say that pins 6 and 7 were LUM 1 and LUM 2, given Pitfall's high voltage. Pin 5 may be LUM0 but the voltages aren't as extreme here. Pin 4 held steady for all the games, singling it out as either BLK (vertical blank) or CSYNC (composite sync). Pin 11 is a mystery; Pitfall got barely any voltage while the other two games crested at 2.5 volts. I'd like to say it was COLOR, but why would a game with so much of it output such little voltage? I wish I better understood how the machine works; it would go a long way toward filling those gaps. So hey, is there anyone out there who could help me solve these mysteries? Come on, throw me a bone here! All I've got left are educated guesses and trial and error! Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 Reverse engineering the chip (that's such a cool term) by checking out the voltages on an Atari 2600 Jr. and comparing them to those on the Coleco Gemini. At this point I'm reasonably certain that pin 4 is CSYNC and that pins 5 through 7 are the three LUMs, likely in numerical order. I suspect pin 11 is OSC and pin 9 is COLOR. I've got no idea what pin 8 is; you'd think it was BLK but the voltages are extremely low on the Gemini. If my hunch is correct, the LUMs and CSYNC are the only ones I'll need to hook up for a black and white display. I can fake it from there. ...uh, why is nobody else interested in this? Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 Update: Pin 4 is DEFINITELY color. Pin 5 is probably CSYNC, as I had a wire with a resistor connected to it and the resulting picture was halfway down the screen and wavy on the bottom half. That suggests a lot of interference. So, for those keeping score: PIN 4: COLOR PIN 5: (likely) CSYNC PIN 6: (likely) LUM 0 PIN 7: (likely) LUM 1 PIN 8: (likely) LUM 2 Quote Link to comment Share on other sites More sharing options...
+Mitch Posted August 6, 2013 Share Posted August 6, 2013 Check this page: http://xi6.com/files/docs/ Mitch Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 6, 2013 Share Posted August 6, 2013 (edited) I'm thinking pin 7 is CSYNC now, but the picture is still off-center and wavy at the bottom. I'm really not sure how to fix the problem, and any help you could provide would be appreciated. Please? EDIT: Oh, thanks for this information! It should be really helpful! Now I need to see how accurate my diagram was... Edited August 6, 2013 by Jess Ragan Quote Link to comment Share on other sites More sharing options...
CPUWIZ Posted August 6, 2013 Share Posted August 6, 2013 Did you look at the PDF? Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 7, 2013 Share Posted August 7, 2013 (edited) Hm. These don't match up. That's weird, because I traced most of the pins on the 6507 to the ones on the TIA, and many are different. Are you sure this is correct? I had a wire connected to pin 4 on the Gemini's TIA and it gave me color output; when I removed it, it switched to monochrome. The PDF has pin 4 set to Phase 0. I don't mean to sound ungrateful or anything, but... are you sure about this? Edited August 7, 2013 by Jess Ragan Quote Link to comment Share on other sites More sharing options...
CPUWIZ Posted August 7, 2013 Share Posted August 7, 2013 No, I am not, but I trust Mitch's advice. Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 7, 2013 Share Posted August 7, 2013 I do appreciate Mitch's help, but the diagram in the PDF doesn't match up with the TIA chip in my Coleco Gemini. There are several key differences: * The two AUD pins have been moved from 12 and 13 on the bottom of the pin, to 27 and 28 on the top. * The D pins are arranged in numerical order, with D0 at pin 12 and running from D7 at pin 19. D6 and D7 are no longer at pins 33 and 34. * Pin 2, formerly CSYNC, has been replaced with Phase 0. I know where most of the pins lie because I used a multimeter to trace them back to the 6507 chip, which Coleco thankfully did not change. I was able to find about half the pins in this manner; the others I needed trial and error or other methods to find. Were there Coleco TIA chips with the same pinout as the chips in Atari's consoles? Possibly. But they weren't in my system. Anyway! Here's the work I've done on the system. It's sloppy and there are wires everywhere, but it does work as shown. There's color and a stable picture and even sound! The picture is a little too bright (especially in the early stages of the mod, shown in the second picture), but swapping the 500 Ohm and 1000 Ohm resistors helped a lot. Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 7, 2013 Share Posted August 7, 2013 Contrast is kind of poor with this mod. Pitfall is fine, but Crystal Castles is impossible when you're hunting for dark blue gems on a grey background. Hopefully different resistor values will do the trick, because the contrast dial on the monitor isn't cutting it. Quote Link to comment Share on other sites More sharing options...
+Mitch Posted August 7, 2013 Share Posted August 7, 2013 Those aren't my docs I just found them browsing the web. Sorry it didn't contain helpful info. Mitch Quote Link to comment Share on other sites More sharing options...
Jess Ragan Posted August 7, 2013 Share Posted August 7, 2013 No problem, you posted what you thought would be useful. As mentioned earlier, I did manage to get this working, so it's all good. Anyway! Here's the progress on my mod. It's a lot tidier than it was yesterday, but the contrast still sucks. I understand now why the original mod asked for potentiometers instead of static resistors, because it makes it a lot easier to adjust the resistance and get the picture just right. I think I've got a couple lying around somewhere, and if worse comes to worse, I'll use those instead of the two resistors I've got in there now. Also, one of the pins on the first joystick shorted out, making it impossible to press fire. I had to resolder the bottom four pins to get that input back. I guess that's the price you pay for owning a Coleco Gemini, huh? Quote Link to comment Share on other sites More sharing options...
+-^CrossBow^- Posted June 25, 2020 Share Posted June 25, 2020 I don't normally necro bump old posts, but I felt this was pretty valuable and was recently asked about a Gemini to possible look at that has been modded previously. So..these pinouts at console5's Wiki seem to match what @Jess Ragan found on his initial reverse engineering. https://console5.com/wiki/E4002 Quote Link to comment Share on other sites More sharing options...
ChildOfCv Posted June 25, 2020 Share Posted June 25, 2020 4 hours ago, -^CrossBow^- said: I don't normally necro bump old posts, but I felt this was pretty valuable and was recently asked about a Gemini to possible look at that has been modded previously. So..these pinouts at console5's Wiki seem to match what @Jess Ragan found on his initial reverse engineering. https://console5.com/wiki/E4002 That was likely the pinout I used while adding the EM1 to my schematics. At least, it matches what I have. I forgot that Coleco made a standalone clone, but it seems to have the same 3 chips as the expansion. Quote Link to comment Share on other sites More sharing options...
Marcão Posted November 17, 2021 Share Posted November 17, 2021 Complete Pinout Coleco Gemini: Quote Link to comment Share on other sites More sharing options...
SoundGammon Posted November 18, 2021 Share Posted November 18, 2021 Anyone ever put a UAV in one? Quote Link to comment Share on other sites More sharing options...
Fabrice Labrousse Posted November 18, 2022 Share Posted November 18, 2022 Did Anyone tried an adaptator to change the Coloco VIC to an ATARI TIA ? 1 Quote Link to comment Share on other sites More sharing options...
pcrock Posted December 22, 2022 Share Posted December 22, 2022 (edited) On 11/18/2022 at 7:59 AM, Fabrice Labrousse said: Did Anyone tried an adaptator to change the Coloco VIC to an ATARI TIA ? A friend of mine did it. He made a small PC board to do it. Now he uses 6526N ou KSC131 tô replace the hard to find ic E4002 Edited December 22, 2022 by pcrock Quote Link to comment Share on other sites More sharing options...
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.