lomdar67 Posted May 24, 2021 Share Posted May 24, 2021 (edited) I have an Atari Jaguar in nearly mint condition. It's a Rev H board and it works fine with my original cartridges. The cartridge slot is fine, not bent pins. After waiting for quit some time a got my GameDrive from DragonBox, it's a Rev 1.2. I've installed the latest GameDrive firmware and I can browse to the contents on the sd-card. But when I try to load any Rom, I get the "Loading ROM..." popup and nothing else happens. The progress bar does not move forward and the console freezes. I tried several sd-cards (I only have SanDisk at the moment) with different sizes. I also tried Roms, Homebrew Roms and jcd files. Nothing is loading. I would appreciate any help to find the problem. Best regards, Michael Edited May 24, 2021 by lomdar67 Quote Link to comment Share on other sites More sharing options...
SainT Posted May 24, 2021 Share Posted May 24, 2021 This is a common problem when one of the interrupt pins on the cartridge port is bad. Have a very close look at your cartridge port for any damaged pins or corrosion. There have even been issues with corroded / broken traces on the console PCB, so while the cart port looked fine there were issues elsewhere. Unfortunately the GameDrive cart uses several more pins on the cartridge port than a normal game cartridge, so while original game cartridges may work fine, it does not guarantee the console is in perfect working condition. Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted May 24, 2021 Author Share Posted May 24, 2021 Thanks for the reply. PCB seems fine to me. Can you be more specific which pins/traces to check? Best regrads, Michael Quote Link to comment Share on other sites More sharing options...
SainT Posted May 24, 2021 Share Posted May 24, 2021 Check EINT0 and EINT1. Oddly enough looking closely at the cart port there does appear to be white residue on the connector plastic and pins around where those pins are located. Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted May 24, 2021 Author Share Posted May 24, 2021 14 minutes ago, SainT said: Check EINT0 and EINT1. Oddly enough looking closely at the cart port there does appear to be white residue on the connector plastic and pins around where those pins are located. I checked all connections with the GameDrive plugged in, to the back of the PCB where the connector is soldered in. So connection from the GameDrive to the back of the PCB are all fine. If there is a problem with a interrupt line it must be on the PCB. I checked EINT0 and IENT1 twice, there is a connection from the GameDrive to the PCB (can you please confirm that the yellowed market connections are the correct ones). If I have to check EINT0 and EINT1 along the PCB, can you show me where to check for continuity on the PCB? Quote Link to comment Share on other sites More sharing options...
SainT Posted May 24, 2021 Share Posted May 24, 2021 The ideal thing to do here is to test it on another Jaguar. There are two options -- The signal from the Jaguar is not making it to the cart. The cart is faulty. If another Jaguar isn't an option, if you can get an oscilloscope or signal probe on the EINT0 and EINT1 lines and check for activity at the cart pin when you try and load a ROM from the menu. It doesn't switch to this type of loading until this point. If you see signals but the cart is not loading, there must be an issue with the cart. And you are getting NO MOVEMENT AT ALL on the loading bar, right? I just want to be clear on this, as some movement indicates a different issue. Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted May 24, 2021 Author Share Posted May 24, 2021 Thanks for your help. Unfortunately I neither have a second Jaguar nor an oscilloscope. There is no movement at all on the loading bar when loading a Rom. I only have movement on the bar when the directory contents is loading. I can't imagine that the console is faulty I have quite e few consoles and have never seen on in such a good condition as the Jaguar (from an optical view). But sometime looks can deceive... Quote Link to comment Share on other sites More sharing options...
phoboz Posted May 24, 2021 Share Posted May 24, 2021 Did anyone get Gorf 2000 to work? I get the 2 options in the picture, but if I select either there is just a black screen. Quote Link to comment Share on other sites More sharing options...
SainT Posted May 24, 2021 Share Posted May 24, 2021 23 minutes ago, lomdar67 said: Thanks for your help. Unfortunately I neither have a second Jaguar nor an oscilloscope. There is no movement at all on the loading bar when loading a Rom. I only have movement on the bar when the directory contents is loading. I can't imagine that the console is faulty I have quite e few consoles and have never seen on in such a good condition as the Jaguar (from an optical view). But sometime looks can deceive... Unfortunately it could be the Jaguar, or it could be the cartridge. The easiest way to check would be to try another Jaguar or another cartridge. It’s worth trying a replacement cartridge if you can, if you get the same problem, then you know it’s the console. Quote Link to comment Share on other sites More sharing options...
Zerosquare Posted May 25, 2021 Share Posted May 25, 2021 Would it be possible to put a fallback in the firmware that would use polling instead of interrupts, at least for loading ROMs? Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted May 26, 2021 Author Share Posted May 26, 2021 On 5/24/2021 at 7:03 PM, SainT said: Unfortunately it could be the Jaguar, or it could be the cartridge. The easiest way to check would be to try another Jaguar or another cartridge. It’s worth trying a replacement cartridge if you can, if you get the same problem, then you know it’s the console. Yes you are right. Still can't imagine that there is something wrong with the console. It's in such a good shape... Could it be a hardware revision problem? Maybe some incompatibility with certain board revisions? Could we analyse the problem with something relatively cheap like this here: https://www.amazon.com/-/us/dp/B07C3F5BP3/ or this: https://www.amazon.com/-/de/dp/B06WW6QNFZ ? I case I would buy one of these (they are much cheaper here in Europe), would you be willing to help me analyse the problem? I've never worked with an oscilloscope before... Best regards, Michael Quote Link to comment Share on other sites More sharing options...
SainT Posted May 26, 2021 Share Posted May 26, 2021 2 hours ago, lomdar67 said: Yes you are right. Still can't imagine that there is something wrong with the console. It's in such a good shape... Could it be a hardware revision problem? Maybe some incompatibility with certain board revisions? Could we analyse the problem with something relatively cheap like this here: https://www.amazon.com/-/us/dp/B07C3F5BP3/ or this: https://www.amazon.com/-/de/dp/B06WW6QNFZ ? I case I would buy one of these (they are much cheaper here in Europe), would you be willing to help me analyse the problem? I've never worked with an oscilloscope before... Best regards, Michael I have one Jaguar, looks fine, it crashes randomly on some carts and not others! Cosmetic appearance does not necessarily have any bearing on functionality. I would see if you can get the cart swapped with DragonBox initially saying you've been chatting with me about the issue. If you get the same behavior, it's the console. I've had far more issues like this where the console is at fault than the cartridge, however I would never rule out a hardware issue with the cartridge either. While they work fine when they leave me, there certainly have been some cases (I think just one) where the cartridge has been the problem. Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted June 2, 2021 Author Share Posted June 2, 2021 Sorry for the delay. I was able to lend a oscilloscope. And I definitively get a signal on EINT0 when I try to load a ROM, but nothing on EINT1. Do I have to get a interrupt signal on both? Best regards, Michael 1 Quote Link to comment Share on other sites More sharing options...
SainT Posted June 2, 2021 Share Posted June 2, 2021 (edited) I've just had a look, EINT1 isn't used currently and EINT0 is used for the interrupt driven data transfer, so that is correct. So the cartridge *IS* correctly generating EINT0 to indicate data is available, but the Jaguar is not responding. So it does look like a faulty console, not cartridge at this point. Next is to check the connection between the cartridge port pin for EINT0 (B25) and pin 36 on Jerry. I think it's a direct connection without and buffer chips in-between. It's also pulled up to VCC at R9, so that's worth checking continuity there as well. You may find there is a bad solder joint on Jerry at this pin, or the trace may be bad at some point along the way. Try testing at the trace edge of the chip leg and also the top of the chip leg away from the trace. It can be useful to test at both locations, as with a dry joint you can appear to get a signal near the trace, but you will not at the top of the chip leg. If this is the case reflowing this joint with some flux should sort it. All very interesting! Edited June 2, 2021 by SainT Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted June 2, 2021 Author Share Posted June 2, 2021 Okay, thanks. I‘ll try this tomorrow. I just did a quick check. Jerry is the DSP, right? I can‘t find any indication on the board where pin one is… Quote Link to comment Share on other sites More sharing options...
Zerosquare Posted June 2, 2021 Share Posted June 2, 2021 1 Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted June 3, 2021 Author Share Posted June 3, 2021 Thanks for the picture! Quote Link to comment Share on other sites More sharing options...
SainT Posted June 3, 2021 Share Posted June 3, 2021 I have to say that's the weirdest pin numbering I've seen on a chip. It's customary to have either the middle or edge pin be pin 1, not a pin three in from the edge. That's nuts. I've doubled checked the schematics, though, and it is indeed correct. Pin 35 as marked above goes to R9 just below the chip, which is correct! 1 Quote Link to comment Share on other sites More sharing options...
Zerosquare Posted June 3, 2021 Share Posted June 3, 2021 8 hours ago, SainT said: I have to say that's the weirdest pin numbering I've seen on a chip. It's customary to have either the middle or edge pin be pin 1, not a pin three in from the edge. That's nuts. Every time you think you can take something for granted, the Jaguar proves you wrong. 3 Quote Link to comment Share on other sites More sharing options...
neo_rg Posted June 3, 2021 Share Posted June 3, 2021 Makes sense when you look at this one.. 2 Quote Link to comment Share on other sites More sharing options...
SainT Posted June 3, 2021 Share Posted June 3, 2021 Lol, wtf?! It does make sense, but then that means the pin numbering on the larger chips is all over the shop. Totally nuts. Quote Link to comment Share on other sites More sharing options...
Zerosquare Posted June 3, 2021 Share Posted June 3, 2021 (edited) It looks like the explanation is that the number of pins wasn't the same depending on the chip manufacturer: The extra pins on each side of the 160-pin version are unconnected, and they don't have a number. Edited June 3, 2021 by Zerosquare 1 Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted June 4, 2021 Author Share Posted June 4, 2021 On 6/2/2021 at 10:11 PM, SainT said: I've just had a look, EINT1 isn't used currently and EINT0 is used for the interrupt driven data transfer, so that is correct. So the cartridge *IS* correctly generating EINT0 to indicate data is available, but the Jaguar is not responding. So it does look like a faulty console, not cartridge at this point. Next is to check the connection between the cartridge port pin for EINT0 (B25) and pin 36 on Jerry. I think it's a direct connection without and buffer chips in-between. It's also pulled up to VCC at R9, so that's worth checking continuity there as well. You may find there is a bad solder joint on Jerry at this pin, or the trace may be bad at some point along the way. Try testing at the trace edge of the chip leg and also the top of the chip leg away from the trace. It can be useful to test at both locations, as with a dry joint you can appear to get a signal near the trace, but you will not at the top of the chip leg. If this is the case reflowing this joint with some flux should sort it. All very interesting! Hello, back again. I have continuity between EINT0 and pin 36 on Jerry and I can read the signal on pin 36 on Jerry when I try to load a Rom. Would be really great if you could help me through the process finding the problem. Best regards, Michael Quote Link to comment Share on other sites More sharing options...
SainT Posted June 4, 2021 Share Posted June 4, 2021 42 minutes ago, lomdar67 said: Hello, back again. I have continuity between EINT0 and pin 36 on Jerry and I can read the signal on pin 36 on Jerry when I try to load a Rom. Would be really great if you could help me through the process finding the problem. Best regards, Michael Ok, interesting. So if the interrupt is making it through to the actual pin on Jerry then either Jerry is faulty, or the external interrupt relay from Jerry to Tom is not working. It's the GPU (Tom) which actually deals with the interrupt, but it's triggered via Jerry. Looking at the schematics and technical reference you have DINT going from Jerry to Tom for this -- pin 24 on Jerry and pin 47 on Tom. This will be the next thing to check. Quote Link to comment Share on other sites More sharing options...
lomdar67 Posted June 4, 2021 Author Share Posted June 4, 2021 I have continuity between pin 24 on Jerry and pin 47 on Tom. I can also read the signal on pin 24/Jerry and 47/ Tom when I try to load a rom… 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.