RickyDean Posted February 23 Share Posted February 23 4 minutes ago, chue said: @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. I'll try another, have several, including NOS one. Quote Link to comment Share on other sites More sharing options...
+Ksarul Posted February 23 Share Posted February 23 18 minutes ago, chue said: You and @Ksarul are crushing my 99/4 dreams 😭. Seriously though, thanks for confirming. The other thing to note is that the /4 motherboard and the keyboard are soldered together via a whole row of metal fingers. The chiclet keyboard and the mother board are basically a single unit, because you really don't want to try and separate them. 1 Quote Link to comment Share on other sites More sharing options...
+chue Posted February 23 Share Posted February 23 16 minutes ago, Ksarul said: the /4 motherboard and the keyboard are soldered together via a whole row of metal fingers. The chiclet keyboard and the mother board are basically a single unit Ah, well that puts the nail in the coffin on my theory of a /4 MB in a /4A case. There's no way my TI has a chiclet keyboard. Good to know though. Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 23 Share Posted February 23 3 hours ago, chue said: @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. I tried my NOS TI99, still having the errors and lockups. I haven't tried any of these without using the FG99, don't know if that would make a difference, but I haven't studied how to load onboard carts yet on the pico. Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 23 Author Share Posted February 23 18 minutes ago, RickyDean said: I tried my NOS TI99, still having the errors and lockups. I haven't tried any of these without using the FG99, don't know if that would make a difference, but I haven't studied how to load onboard carts yet on the pico. I've changed the attached version of the firmware to be more similar to the 260MHz you have working, see if it's any better? (258MHz, just 2 NOPs delay now and an extra [but small] timing change where I found the needed improvement for the last firmware.) PPEB2.ino.uf2.zip 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 23 Author Share Posted February 23 I did a bit of a clean-up... I'm now down to 252MHz (it works at 250MHz, so one level up...) with 3 NOPs again. If this doesn't work, then I can't fix it. Sad, but true. PPEB2.ino.uf2.zip 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 23 Share Posted February 23 8 hours ago, JasonACT said: I did a bit of a clean-up... I'm now down to 252MHz (it works at 250MHz, so one level up...) with 3 NOPs again. If this doesn't work, then I can't fix it. Sad, but true. PPEB2.ino.uf2.zip 379.42 kB · 0 downloads Okay I'll give these a shot, thank you. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 23 Author Share Posted February 23 @RickyDean Here is one more firmware, running at 260MHz / 130MHz PSRAM speed, which you have previously verified works on your devices... There are no other changes since the clean-up version, but it occurred to me that maybe your TIs are sampling the data lines earlier than my TI, which this version may fix? PPEB2.ino.uf2.zip Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 23 Share Posted February 23 (edited) 58 minutes ago, JasonACT said: @RickyDean Here is one more firmware, running at 260MHz / 130MHz PSRAM speed, which you have previously verified works on your devices... There are no other changes since the clean-up version, but it occurred to me that maybe your TIs are sampling the data lines earlier than my TI, which this version may fix? PPEB2.ino.uf2.zip 379.54 kB · 0 downloads Okay, I've about through testing the first of these 3 files. I'll skip the second and test this one next. Edit: I replaced the headers again on my third one and am using my last pico on it, as the third pico may have an issue. I am testing this one first before the other two, but I am using the last firmware here and will test on all 3, both in SAMS and Foundation mode. Then I will test with a couple of more TI's. Edited February 23 by RickyDean additional content 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 Just for interest, this is my logic analyser looking at the memory tester in SAMS mode (a write cycle to low memory, which does a read before write): All the trouble is happening at the first *PSRAM_EN signal, at the end when it rises - it needs to be well before A15 goes low... Which on my machine, it is working. (This is the latest 260MHz firmware.) 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 24 Share Posted February 24 3 hours ago, JasonACT said: @RickyDean Here is one more firmware, running at 260MHz / 130MHz PSRAM speed, which you have previously verified works on your devices... There are no other changes since the clean-up version, but it occurred to me that maybe your TIs are sampling the data lines earlier than my TI, which this version may fix? PPEB2.ino.uf2.zip 379.54 kB · 2 downloads The firsrt file and the last file both work good on all three of my pico's, both Foundation and Sams. I am trying the second file now. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 6 minutes ago, RickyDean said: The firsrt file and the last file both work good on all three of my pico's, both Foundation and Sams. I am trying the second file now. I'm very happy to hear that! Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 Here is the same (latest) firmware, and same speed (260MHz) except SAMS is configured as 2MB instead of 1MB... My logic analyser tells me there's a slight increase in the time to access PSRAM, but it doesn't appear to be enough for my TI to notice (about 12ns extra). @arcadeshopper has said nothing uses it, and I agree in that there's only a couple of people here on the forum that can or would use it. But, here it is anyway... One megabyte less to load large cartridges, which mostly also don't exist, you have to disable SAMS altogether to run the only 8MB cart there is. So, I'm conflicted in what to stick with, I'll probably go back to 1MB. Something to test though PPEB2.ino.uf2.zip 2 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 24 Share Posted February 24 (edited) 14 hours ago, JasonACT said: I'm very happy to hear that! Well the second file works on my 1st pico-riser and 3rd pico- riser, and passes Foundation on all 3, but the Sams test locks up on the 2nd pico- w/o riser. I'm testing it one more time now. This is only using one TI, may be different on others, but it takses up a lot of time to test 3 units extensively. I'll try this last one soon. Well testing is done on file 2 on this page, and it still locks up the 2nd pico w/o riser. I'm now starting tests on the 4th file. Edited February 24 by RickyDean spelling, added content Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 2 hours ago, RickyDean said: Well the second file works on my 1st pico-riser and 3rd pico- riser, and passes Foundation on all 3, but the Sams test locks up on the 2nd pico- w/o riser. I'm testing it one more time now. This is only using one TI, may be different on others, but it takses up a lot of time to test 3 units extensively. I'll try this last one soon. Well testing is done on file 2 on this page, and it still locks up the 2nd pico w/o riser. I'm now starting tests on the 4th file. OK, that possibly shows I've been trying to fix the wrong issue, PSRAM being pushed too fast, and instead I need to build the firmware 4 (or more) MHz faster than where mine is passing the test, rather than just 2MHz faster. 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 24 Share Posted February 24 58 minutes ago, JasonACT said: OK, that possibly shows I've been trying to fix the wrong issue, PSRAM being pushed too fast, and instead I need to build the firmware 4 (or more) MHz faster than where mine is passing the test, rather than just 2MHz faster. I can report that all 3 passed the 2MB Sams testing, now testing each on the foundation to be sure their still working. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 7 minutes ago, RickyDean said: I can report that all 3 passed the 2MB Sams testing, now testing each on the foundation to be sure their still working. Great news I've been trying to see where this version starts working on my board/TI - seems it's 256MHz (fails at 254). So the firmware posted above is running 4MHz faster. Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 24 Share Posted February 24 58 minutes ago, JasonACT said: Great news I've been trying to see where this version starts working on my board/TI - seems it's 256MHz (fails at 254). So the firmware posted above is running 4MHz faster. They all 3 passed the Myarc/Foundation tests too. Good work! 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 24 Author Share Posted February 24 1 hour ago, RickyDean said: They all 3 passed the Myarc/Foundation tests too. Good work! Thank you Ricky. Just for "completeness" attached is the same 2MB SAMS firmware, compiled for 262MHz, 264MHz and 266MHz. I fully tested the 266MHz version, but only partially tested the 2 others, the 2MB SAMS test takes ages to run! I don't think it's important to really try all of these, but I'm leaving them here for the Pico builds that can run the faster firmwares. PPEB2.ino.uf2.zip 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 25 Share Posted February 25 5 hours ago, JasonACT said: Thank you Ricky. Just for "completeness" attached is the same 2MB SAMS firmware, compiled for 262MHz, 264MHz and 266MHz. I fully tested the 266MHz version, but only partially tested the 2 others, the 2MB SAMS test takes ages to run! I don't think it's important to really try all of these, but I'm leaving them here for the Pico builds that can run the faster firmwares. PPEB2.ino.uf2.zip 1.11 MB · 2 downloads Well the 266 vesion works on two of these, but the 3rd got to page 11 and locked up, so tomorrow I'll try the 264 version on that one. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 25 Author Share Posted February 25 (edited) 25 minutes ago, RickyDean said: Well the 266 vesion works on two of these, but the 3rd got to page 11 and locked up, so tomorrow I'll try the 264 version on that one. Interesting! That has to be hitting the PSRAM limit on your 3rd unit. I've been playing around with mine this afternoon, using the logic analyser, and I think I've got 10ns more out of my PSRAM before A15 goes low - but that won't really help your 3rd unit. I'm trying to think why you are getting lock-ups though, if you're running from a FinalGROM then I don't think the PSRAM being tested here would cause that... You either get the right byte value - or you don't, that's all. Even if you were using my device as a cartridge, the first 8KB of ROM is always in Pico RAM, so is fast. That's all the memory tester is, really, 8KB of ROM. A bit of a mystery then - I look forward to you testing the 264MHz version Have a good night! Edited February 25 by JasonACT Quote Link to comment Share on other sites More sharing options...
+chue Posted February 25 Share Posted February 25 Here are more results from my testing as well. I only tested a single PI, but it seems they both run at 266 mhz. I tested these firmwares: 252 mhz (1MB SAMS), 260 mhz (2 MB SAMS), and 266 mhz (1 MB SAMS) These all passed the SAMS burn-in (3+ passes) on both the silver TI and the beige TI. I tested using the onboard cart functionality (without FG). I hadn't mentioned it before, but I have another beige TI. It came with a Mitsumi keyboard, which isn't working. I did a keyboard swap so I could do some testing using the PI Pico. The TI is a QI model (non v2.2 1981 ROM), so it has potentially some differences in the motherboard timings. I had to test with the FinalGROM on the QI TI because the Pico onboard cart would not work. I was able to choose the cartridge but after the reset, the cartridge would not show as an option. In any case, I tested using the same firmwares as the other two TIs. All passed the SAMS test. I will try to test the 264 mhz 2 MB SAMS firmware next, as the others have been accounted for. 1 Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 25 Author Share Posted February 25 23 minutes ago, chue said: I had to test with the FinalGROM on the QI TI because the Pico onboard cart would not work. I was able to choose the cartridge but after the reset, the cartridge would not show as an option. Yeah, I've recently been reading the "pgram+ operating & construction guide.pdf" which states it doesn't work with a QI motherboard because "These consoles, manufactured in late 1983 and early 1984 used different internal circuitry that does not permit access to GROM data through the Peripheral Expansion System." I'm assuming they also lock-out the ROM too from the side port - since the memory tester has no GROM and is ROM only. Quote Link to comment Share on other sites More sharing options...
JasonACT Posted February 25 Author Share Posted February 25 Very sad... Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 25 Share Posted February 25 8 hours ago, JasonACT said: Interesting! That has to be hitting the PSRAM limit on your 3rd unit. I've been playing around with mine this afternoon, using the logic analyser, and I think I've got 10ns more out of my PSRAM before A15 goes low - but that won't really help your 3rd unit. I'm trying to think why you are getting lock-ups though, if you're running from a FinalGROM then I don't think the PSRAM being tested here would cause that... You either get the right byte value - or you don't, that's all. Even if you were using my device as a cartridge, the first 8KB of ROM is always in Pico RAM, so is fast. That's all the memory tester is, really, 8KB of ROM. A bit of a mystery then - I look forward to you testing the 264MHz version Have a good night! I tested the 264 on that one last night and it passed. FYI, the QI motherboard is laid out differently from the standard TI motherboard too. 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.