+chue Posted February 18 Share Posted February 18 17 hours ago, JasonACT said: Can you try this re-compile of the "newest" Foundation firmware? It passes the SAMS burn-in test. 19 hours ago, JasonACT said: But only with one of the Foundation firmwares Now two of the three Foundation firmwares work. Oh, and the "failsafe" (3rd? version) firmware also works. 1 hour ago, JasonACT said: It would be good to get a report of what the original memory "bench-only" tester reports Both of my PICO PEBs blink the LED in the long-on, long-off pattern. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 19 Author Share Posted February 19 @RickyDean & @chue - This is an EXPERIMENTAL build of the new Foundation firmware, it runs at 266MHz (133MHz PSRAM) so may not work with all Pico PEB boards. It pushes out the timing as much as I can, without me seeing any issues when running 32KB programs in SAMS memory with SAMS paging turned on. This is mainly for testing on TI's that may seem to have issues with the device... If it makes things better (wouldn't that be great, but I'm not holding my breath) then I might make another to try which isn't so borderline. PPEB2.ino.uf2.zip 3 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 19 Share Posted February 19 (edited) 15 minutes ago, JasonACT said: @RickyDean & @chue - This is an EXPERIMENTAL build of the new Foundation firmware, it runs at 266MHz (133MHz PSRAM) so may not work with all Pico PEB boards. It pushes out the timing as much as I can, without me seeing any issues when running 32KB programs in SAMS memory with SAMS paging turned on. This is mainly for testing on TI's that may seem to have issues with the device... If it makes things better (wouldn't that be great, but I'm not holding my breath) then I might make another to try which isn't so borderline. PPEB2.ino.uf2.zip 379.54 kB · 0 downloads I'll try it tomorrow when I can, Thanks, Jason, and good night. Edited February 19 by RickyDean spelling 1 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 19 Share Posted February 19 Using the latest firmware, SAMS burn-in works on both my Pico PEBs. I let the tests run at least 3 passes. Beige TI only, as the other one is out of commission until I can look at the power supply. 2 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 19 Share Posted February 19 @JasonACT The first two pictures are the second pico I built, the one without risers. The last two are the 1st one with risers. This file did not work with either. 1 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 20 Share Posted February 20 I should have mentioned that my prior testing was with the FinalGROM. I am currently doing some testing with the Pico onboard carts. Will post results when I have them. 1 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 20 Share Posted February 20 Onboard carts are working for me. I ran the SAMS tester burn-ins and was able to do 3+ passes on both Picos. 3 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 20 Author Share Posted February 20 59 minutes ago, chue said: Onboard carts are working for me. I ran the SAMS tester burn-ins and was able to do 3+ passes on both Picos. That's good news @RickyDean I'll hang off posting any more firmwares, at least until @chue has fixed his power supply and reported back: a/if that was what was causing it trouble, or b/this experimental firmware is something that makes it better. I'm sorry it bombed out on your boards, I do recall the bench memory tester wasn't flashing with a high number for you. It does look like 266MHz/133MHz is pushing them too far. Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 20 Share Posted February 20 1 minute ago, JasonACT said: That's good news @RickyDean I'll hang off posting any more firmwares, at least until @chue has fixed his power supply and reported back: a/if that was what was causing it trouble, or b/this experimental firmware is something that makes it better. I'm sorry it bombed out on your boards, I do recall the bench memory tester wasn't flashing with a high number for you. It does look like 266MHz/133MHz is pushing them too far. Yea, I went back to the firmware you posted just before this one and they are both testing good. With Sams and Myarc/Foundation. 2 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 20 Share Posted February 20 I pulled the power supply from the silver TI and measured the rails: +5, +12, -12. The last one should be -5, so it indeed was the power supply. I have several parts TIs, all of which boot to a black screen, so I went looking for a replacement power supply. The first PS had the same problem: -12V on the -5V rail. The second power supply had the correct voltages, but the wrong connector - it had wires coming from the PS that went to what looked like a mini molex connector. The 3rd and last power supply had the correct voltages. This is the PS that is used in these tests. Silver TI test results 3rd Pico firmware "fail safe", with FinalGROM: 3rd Pico firmware "fail safe", onboard Pico cart: 8ma firmware, with FinalGROM: 8ma firmware, onboard Pico cart: Foundation #2 firmware, onboard Pico cart: Foundation #3 firmware, onboard Pico cart: So there we have it, only the latest "experimental" firmware works. Obviously I did not test all firmwares, with and without the FG. If there is some other scenario that you would like to see, please let me know @JasonACT. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 21 Author Share Posted February 21 @chue That's great news, and it shows not all TMS9900 chips are equal From what I'm seeing, 1983 chips have improved signals. @RickyDean Attached are 2 more experimental firmwares, you already have 260MHz(std) & 266MHz(exp) and here are 262 & 264 each with timing pushed out as much as I can, to test on your original TI (and working TI too, of course). PPEB2.ino.uf2.zip 2 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 21 Share Posted February 21 29 minutes ago, JasonACT said: @chue That's great news, and it shows not all TMS9900 chips are equal From what I'm seeing, 1983 chips have improved signals. @RickyDean Attached are 2 more experimental firmwares, you already have 260MHz(std) & 266MHz(exp) and here are 262 & 264 each with timing pushed out as much as I can, to test on your original TI (and working TI too, of course). PPEB2.ino.uf2.zip 759.01 kB · 1 download I'll try these tomorrow evening after I get home and rest from Physical Therapy, those sessions have been tough on my shoulder. 2 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 21 Author Share Posted February 21 @chue Attached is an archive with 4 x 266MHz firmwares - numbered 2..5 (this is the set NOP delays I'm using, the latest experimental firmware you are using is #6). The idea here being, you can test them in order until you find the first that works, then use the next one for production. I.E. if #2 passes the tests, use #3 to be on the safe side for your machine. I'll be interested in which one works first PPEB2.ino.uf2.zip 3 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 21 Share Posted February 21 1 hour ago, JasonACT said: Attached is an archive with 4 x 266MHz firmwares I will give these a try and report back. Thank you! 3 Quote Link to comment Share on other sites More sharing options...
+Ksarul Posted February 22 Share Posted February 22 On 2/20/2024 at 3:27 PM, chue said: The second power supply had the correct voltages, but the wrong connector - it had wires coming from the PS that went to what looked like a mini molex connector. These power supplies were used in 99/4s and early 99/4As. They are a bit HTF. Somewhere, I have a bag of the mating connector for those and the proper connector to make them work with the later motherboard connectors. The resulting cable needs to be kept a bit short so as not to get in the way of anything, but it does work to make them compatible with later consoles. 4 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 22 Share Posted February 22 I tested the 4 x 266 firmwares and the #2 firmware works on both my TIs. However, on the beige TI there is some stuttering of the spinning progress "wheel". Maybe it isn't really important, as firmware #3 (the recommended firmware) runs without any stuttering. There was no stuttering on the silver TI. Here's a video of the stutter: _beige_ti_ramtest_stutter.mp4 Just for comparison, what it looks like with another firmware (no stuttering): _beige_ti_ramtest_no_stutter.mp4 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 22 Share Posted February 22 (edited) 22 hours ago, JasonACT said: @chue That's great news, and it shows not all TMS9900 chips are equal From what I'm seeing, 1983 chips have improved signals. @RickyDean Attached are 2 more experimental firmwares, you already have 260MHz(std) & 266MHz(exp) and here are 262 & 264 each with timing pushed out as much as I can, to test on your original TI (and working TI too, of course). PPEB2.ino.uf2.zip 759.01 kB · 4 downloads Beige Console only PPEB2_262.ino.uf2: 1st pic- 1st with riser 2nd pic- 2nd w/o riser PPEB2_264.ino.uf2 3rd pic- 1st with riser 4th pic- 2nd w/o riser None of the 4 files destined for @chue worked on either- errors or lockups. On the beige console. Edited February 22 by RickyDean additional content 2 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 22 Share Posted February 22 3 hours ago, Ksarul said: These power supplies were used in 99/4s and early 99/4As. I had done some initial research on it and found the power supply page at MainByte - http://www.mainbyte.com/ti99/hardware/power_supply.html I got a little excited when I saw that the 99/4 PS has wires coming out of it. I was thinking that perhaps I had a 99/4 board inside of one of my dead 4As. I still have to take a look, just to rule it out. But alas, you are probably correct that I have an early 4A. Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 22 Author Share Posted February 22 3 hours ago, chue said: I tested the 4 x 266 firmwares and the #2 firmware works on both my TIs. However, on the beige TI there is some stuttering of the spinning progress "wheel". Maybe it isn't really important, as firmware #3 (the recommended firmware) runs without any stuttering. There was no stuttering on the silver TI. Weird! To be honest, I woke up this morning with the idea of generating these files for you before I left for work, so I didn't try them myself. I've just tried #2 on my (beige mid '83) console and... nope... no stutters. 2 hours ago, RickyDean said: Beige Console only PPEB2_262.ino.uf2: 1st pic- 1st with riser 2nd pic- 2nd w/o riser PPEB2_264.ino.uf2 3rd pic- 1st with riser 4th pic- 2nd w/o riser None of the 4 files destined for @chue worked on either- errors or lockups. On the beige console. Looks like 260MHz is your max speed then, I'll take a look at what code GCC is generating for the PSRAM code path, but I'm not sure I'll be able to get a couple of NOPs in there and working at that speed. 2 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 22 Author Share Posted February 22 1 hour ago, JasonACT said: Looks like 260MHz is your max speed then, I'll take a look at what code GCC is generating for the PSRAM code path, but I'm not sure I'll be able to get a couple of NOPs in there and working at that speed. Well, that got me nowhere. It's horrid generated code and not really readable. @RickyDean I have built a new 260MHz firmware with 3 NOPs where we think they are needed. I got back that time in the code path by moving things around, it makes my code "horrid, and not really readable"... But... That's life, see how this one goes for you. PPEB2.ino.uf2.zip 2 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 22 Author Share Posted February 22 This actually worked all the way down to 256MHz, then failed at 254MHz... So here is a 258MHz version to try too (always best to go one better)... PPEB2.ino.uf2.zip 2 Quote Link to comment Share on other sites More sharing options...
+arcadeshopper Posted February 22 Share Posted February 22 16 hours ago, chue said: I had done some initial research on it and found the power supply page at MainByte - http://www.mainbyte.com/ti99/hardware/power_supply.html I got a little excited when I saw that the 99/4 PS has wires coming out of it. I was thinking that perhaps I had a 99/4 board inside of one of my dead 4As. I still have to take a look, just to rule it out. But alas, you are probably correct that I have an early 4A. yeah they both had them in the beginning then the 4A went to the header on the ps board and just the cable from the motherboard I have both varieties here. I should update the FAQ with more detail 2 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 22 Share Posted February 22 @JasonACT, Tried both of these files to no avail, still had errors and lockups on both picos. I did get my new right angle headers in, will try to fix my third and do the fourth one, this weekend. Quote Link to comment Share on other sites More sharing options...
+chue Posted February 23 Share Posted February 23 @RickyDean I understand your first TI was bad. Do you have a third TI to try? Maybe there is an issue with your current TI. Quote Link to comment Share on other sites More sharing options...
+chue Posted February 23 Share Posted February 23 5 hours ago, arcadeshopper said: yeah they both had them in the beginning then the 4A went to the header on the ps board and just the cable from the motherboard I have both varieties here. You and @Ksarul are crushing my 99/4 dreams 😭. Seriously though, thanks for confirming. 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.