+InsaneMultitasker Posted February 19 Author Share Posted February 19 41 minutes ago, Gary from OPA said: Good point I will dig out a normal original to made 32k card and try it with that instead of the corcomp and see. Corcomp Ramdisk is also one of those cards that also modify the TI99 startup screen. I also think there more than one version of corcomp dsr so should check to see what version I have. Another thought. When you load ROS, be sure to say "NO" when asked whether to preserve the previous info. Let's see if that changes the results. If you say yes, it is going to pull all the calls, subprograms, colors, etc. to save and restore them. For the CC DSR, there were a few versions. The binary that I grabbed was from the forum as posted by @RickyDean some time ago. Quote Link to comment Share on other sites More sharing options...
apersson850 Posted February 19 Share Posted February 19 My last PC that had a 5.25" drive is dead, so I don't have any simple way to moving files from modern machines to the TI. Maybe a text file serially. Have only tried it the other way. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 19 Share Posted February 19 1 hour ago, apersson850 said: My last PC that had a 5.25" drive is dead, so I don't have any simple way to moving files from modern machines to the TI. Maybe a text file serially. Have only tried it the other way. do you have a working 3.5" drive at least? -- my way I am using right now, since I don't have a TIPI yet, is running the CFHDXS1 for the nanopeb from flashrom99 or finalrom99 but using a standard rs232 card and serial cable to my pc (i have serial ports on a pcie card) and running HDX server on it, I can at least transfer v9t9 and tifiles that way, just can't mount disk images. - but lucky my 5.25 drives work, and i got a box of brand-new 5.25" floppies as well. -- sadly my 3.5" drives work, but i can't fully format brand-new unused 3.5" floppies without a few dozen bad sectors, but I can reformat used 3.5" floppies i had laying around, strange. i also have this device: Semi-Virtual Diskette (SVD) - TI 99/4a which can simulate a floppy drive, but I am not using it currently, you connect instead of physical drives on your peb, and then run a null modem cable to your pc and the software on the pc side can mount dsk images (mainly 90k ones work best), you could do that if you have no working floppy drives, but that requires buying new hardware, might be best to invest then in tipi instead in your case. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 19 Share Posted February 19 4 hours ago, InsaneMultitasker said: Another thought. When you load ROS, be sure to say "NO" when asked whether to preserve the previous info. Let's see if that changes the results. If you say yes, it is going to pull all the calls, subprograms, colors, etc. to save and restore them. For the CC DSR, there were a few versions. The binary that I grabbed was from the forum as posted by @RickyDean some time ago. this seem to be the trick that was needed for both systems, using cfg842c and ros842c and saying no and then formatting my drives again and copying all the files back over, and both my avpc and my corcomp 512k systems are working perfectly. more debugging i think the power-down or lithumin battery on my midwest card on my avpc system needs fixing as sometimes i get corruption after switching cartridges or turning off the console, so there something going on there, have to check into that one. and still have to look at why my replacement boot menu (ramos) doesn't like the avpc card, which is strange as it is design to support v9938 system. as for corcomp 512k system, the only glitch is cfg842c displays myarc hdfc where as cfg838 didn't but for that system, saying NO as well, worked and was able to load ros842c without issues, the only thing is when there is no drives formatted, it displays the title screen, once you format a drive, it autoloads into the cartridge you have inserted like e/a or x/b but locks up if no cartridge found until i copy the menu over to the newly formatted drive, strange. and that system both your supplied menu739c and my ramos works perfectly, so that is good. here some screenshots, but not really needed now, just need to fix the detection of the corcomp 512k like how it worked before in the older configs before 842c. 2 1 Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 19 Author Share Posted February 19 7 minutes ago, Gary from OPA said: as for corcomp 512k system, the only glitch is cfg842c displays myarc hdfc where as cfg838 didn't but for that system, saying NO as well, worked and was able to load ros842c without issues, the only thing is when there is no drives formatted, it displays the title screen, once you format a drive, it autoloads into the cartridge you have inserted like e/a or x/b but locks up if no cartridge found until i copy the menu over to the newly formatted drive, strange These are very good and interesting observations to see all together. I will add an issue to Github with a link to this topic. There is a lot of magic happening in ROS and maybe you'll spot the issue when you review the code. I seem to recall that the lockup can be bypassed by pressing the shift key? I only have the Geneve system right now, so my testing options are limited. Thank you for the pictures and confirming the CC ramdisk subprograms. I see some potential FAQ material @dhe Oh, I don't see MENU listed in the user options; this may be a clue as well. Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 19 Author Share Posted February 19 @Gary from OPA is this one of the 'bad' ramdisks or some other card that is UNDEFINABLE? Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 19 Share Posted February 19 29 minutes ago, InsaneMultitasker said: @Gary from OPA is this one of the 'bad' ramdisks or some other card that is UNDEFINABLE? that is on my avpc system, it is how it displays the avpc card which is fixed at >1400 cru since there is no subprograms or calls in the dsr just a power-up and interrupt program. normally the check is to look or the text DIJIT or AVPC within a certain dsr range, but not sure if that works for all versions of the DSR from 010 to 020 to 030 to 032 and Tony's one, I will have to compare all 5 known versions and find one solid way to detect the card. there is for sure some dying of the midwest one in this avpc system as well, as the dsr easy corrupts just switching cartridges without even powering off. time to start working on all my other horizon ramdisks i have and get the two best stable ones working, and at least one or two with rambo as well. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 19 Share Posted February 19 36 minutes ago, InsaneMultitasker said: These are very good and interesting observations to see all together. I will add an issue to Github with a link to this topic. There is a lot of magic happening in ROS and maybe you'll spot the issue when you review the code. I seem to recall that the lockup can be bypassed by pressing the shift key? I only have the Geneve system right now, so my testing options are limited. Thank you for the pictures and confirming the CC ramdisk subprograms. I see some potential FAQ material @dhe Oh, I don't see MENU listed in the user options; this may be a clue as well. on my corcomp system i have the menu as UL since i am running my menu replacement called ramos, and it is named UL so i can load it from funnelweb as userlist. 1 Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 20 Share Posted February 20 got my original 1536k horizon 4000 working, seems to be a battery issue, need to order a new 3.6 volt aa lithium for it, i think my midwest one has similar issue. one thing i noticed with horizon 4000 once i get new battery installed i will do some more testing, but i made 3 drives on it, a 3200 sector one, 1440 sector one and 368 sector one and then suddenly i had 2040k of rambo ram when the card is only 1536k in total, so something is wrong there in the calculation of free space. its 1am now, i will order the battery, install it tomorrow, and take some screenshots to show you the error. still no luck with my 3 other ramdisks, but they look in rough shape, one only sees 136k instead of 180k, and one not seen at all, and one just freezes the peb up on startup. i will take some pictures of them and maybe someone can point out what i should check first, but i think all 3 need to be stripped of their wires and resoldered, and well those germ diodes replaced as well at least. 1 Quote Link to comment Share on other sites More sharing options...
RickyDean Posted February 20 Share Posted February 20 16 minutes ago, Gary from OPA said: got my original 1536k horizon 4000 working, seems to be a battery issue, need to order a new 3.6 volt aa lithium for it, i think my midwest one has similar issue. one thing i noticed with horizon 4000 once i get new battery installed i will do some more testing, but i made 3 drives on it, a 3200 sector one, 1440 sector one and 368 sector one and then suddenly i had 2040k of rambo ram when the card is only 1536k in total, so something is wrong there in the calculation of free space. its 1am now, i will order the battery, install it tomorrow, and take some screenshots to show you the error. still no luck with my 3 other ramdisks, but they look in rough shape, one only sees 136k instead of 180k, and one not seen at all, and one just freezes the peb up on startup. i will take some pictures of them and maybe someone can point out what i should check first, but i think all 3 need to be stripped of their wires and resoldered, and well those germ diodes replaced as well at least. I would start with the logic chips first and test them, unless you see that the boards have some obvious damage on the wires and such (mice,rats, battery leakage).also reseat the IC's and look at the socket wiper to be sure their not corroded. Thes things you most likely know anyhow. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 20 Share Posted February 20 24 minutes ago, RickyDean said: I would start with the logic chips first and test them, unless you see that the boards have some obvious damage on the wires and such (mice,rats, battery leakage).also reseat the IC's and look at the socket wiper to be sure their not corroded. Thes things you most likely know anyhow. i made sure to remove the batteries before i stored them away 13 years ago, and they were left in the peb and then the peb was heavy wrapped in bubble wrap and then it was put in a sealed climate control box and then that was put in storage unit with steady climate control, so its not too bad shape, but the original soldering job is crappy, you can see flux leakage on the back, and the wiring was the old bell wires, bud mills had a habit of using that type of wires, and hand-soldering was not the best back in 1986 that is for sure, and some of sockets are very cheap. -- i might just take the ram and stack them on my working boards to increase the overall memory space, instead of running multiple smaller boards, for sure for now, i going to move the rambo off one of them onto one of my current working ones as i want have at least rambo up and running as well. i should start with the one that at least sees 130k and trace out why it not seeing the rest of the memory. 1 Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 20 Author Share Posted February 20 22 minutes ago, Gary from OPA said: one thing i noticed with horizon 4000 once i get new battery installed i will do some more testing, but i made 3 drives on it, a 3200 sector one, 1440 sector one and 368 sector one and then suddenly i had 2040k of rambo ram when the card is only 1536k in total, so something is wrong there in the calculation of free space. its 1am now, i will order the battery, install it tomorrow, and take some screenshots to show you the error. I recompiled CFG. This attached version contains a different, non-destructive SAMS card detection from 2023 (I don't know if it was tested), the HFDC fix (test for >27), and an update to the RAMBO code described below. I would test 8.42 and this version to see whether the results are different. cfg1cfg2 The page count is stored in @ldros0+2 via self-modifying code but as you can see below, I forgot the +2 for retrieval of the value. So the value of R0 is always >0202, the equivalent of "LI R2", which should force RAMBO to always show >2040K. I have never had the ability to test a RAMBO-enabled card with less than 8MB so I am glad you have some equipment for testing! I suspect the page count routine is incorrectly translating from 2K racks to 8k rambo pages, so we might not be completely out of the woods. LDROS0 LI R2,>0000 SAVED R2 FOR LOADING OF ROS MAP! (more code follows) *------ 7.8.2015 fix RAMBO maximum allocation ---- mov @ldros0,r0 ;2-15-2020, need to get the actual page count here! not remaining ram on card! CI R0,255*2 Check remaining (RAMBO) pages JL RAMOK2 if less than 255, we are ok otherwise LI R0,255*2 force maximum to 255 (can't exceed!) 1 Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 20 Share Posted February 20 @InsaneMultitasker thanks I will test it out tomorrow and let you know the results and what more I discovered. Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 21 Author Share Posted February 21 22 hours ago, InsaneMultitasker said: non-destructive SAMS card detection from 2023 (I don't know if it was tested), FYI, this was tested by another AA member in September 2023, so it should be good. 21 hours ago, Gary from OPA said: thanks I will test it out tomorrow and let you know the results and what more I discovered No rush. I'll probably check back this weekend. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 21 Share Posted February 21 53 minutes ago, InsaneMultitasker said: FYI, this was tested by another AA member in September 2023, so it should be good. No rush. I'll probably check back this weekend. Yeah I didn't get to it, as Amazon was late in delivering me some new 3.6 volt lithium batteries for the horizon 4000 and my Midwest version cards Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 22 Author Share Posted February 22 On 2/19/2024 at 4:04 PM, Gary from OPA said: on my avpc system, it is how it displays the avpc card which is fixed at >1400 cru since there is no subprograms or calls in the dsr just a power-up and interrupt program. normally the check is to look or the text DIJIT or AVPC within a certain dsr range CFG currently looks for the following: AVPC = "VIDEO" EVPC = "EVPC" So at least one version of AVPC is detectable, as far as I know. Quote Link to comment Share on other sites More sharing options...
apersson850 Posted February 22 Share Posted February 22 (edited) A bit funny is that it detects my real time clock card, in spite of it being my own design and thus there's only one copy in the world. Must be because one DSR entry is named "CLOCK". My IO card, also designed by me, goes under the radar, though. That's just too unique. Edited February 22 by apersson850 2 Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 22 Share Posted February 22 Got my new batteries soldered in and I put the Midwest one in my TI99 system with the corcomp 512k and the test cfg displays it right now. I put the 4000 into my avpc system but it still doesn't display the amount of free space for rambo correctly but cfg838 and cfg814 and tst displays rambo free space correctly but still not the new test cfg. I also had issues with making a drive at 3200 sectors after copying a bunch of files I ended up with the top two files with the same filename and coping more files after that screwed things up even more. I going to do more tests and see if I can recreate the corruption when the 4000 is formatted as 3200 sector drive. I had no issues with multiple 1440 drives on the 4000. 2 Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 22 Author Share Posted February 22 The RAMBO calc issue doesn't surprise me. It's been waiting for some good testing. The screenshot shows two drives <800K, is one of these your 4000? The 3200 sector drive partition has been stable for quite some time. I ran multiple partitions for a few years. I won't rule out a software issue but that wouldn't be my first place to look. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 22 Share Posted February 22 8 minutes ago, InsaneMultitasker said: The RAMBO calc issue doesn't surprise me. It's been waiting for some good testing. The screenshot shows two drives <800K, is one of these your 4000? The 3200 sector drive partition has been stable for quite some time. I ran multiple partitions for a few years. I won't rule out a software issue but that wouldn't be my first place to look. No that screenshot is my corcomp 512k system. My horizon 4000 has 1536k of ram and is in my avpc system. I will take some screenshots of it later on when I try the 3200 format again. And see what happens. As for the rambo free space only the newer configs get it wrong cfg838 and cfg814 get it right. Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 22 Author Share Posted February 22 2 minutes ago, Gary from OPA said: No that screenshot is my corcomp 512k system. My horizon 4000 has 1536k of ram and is in my avpc system. I will take some screenshots of it later on when I try the 3200 format again. And see what happens. As for the rambo free space only the newer configs get it wrong cfg838 and cfg814 get it right. Sounds good. The older CFGs didn't have the boundary checks for larger drives so something must be wrong in the logic that tests for >2MB. If I have time this weekend, I'll take another look. Quote Link to comment Share on other sites More sharing options...
+InsaneMultitasker Posted February 22 Author Share Posted February 22 Two more thoughts. If you are able to reproduce the problem with a set of files, we can test on another ramdisk or in MAME.The latter should rule out hardware in most cases. Also, exercise the DIPs and switches. Maybe there is something intermittent causing an issue. Same goes for sockets/chips. Quote Link to comment Share on other sites More sharing options...
Gary from OPA Posted February 23 Share Posted February 23 (edited) i given up trying to get my other horizon ramdisks working, i tried swapping around support chips and such, no luck, one sees the cru bus, and the led turns on, but nothing else, the other sees 138k out of 186k but with tons of errors, the other one with rambo it looks perfect, but nothing, doesn't get detected at all even with all new support chips, and the horizon 3000 one is yikes horrible, but it never had any memory on it, it needs to be stripped of all sockets and rebuilted. i thinking of doing that, and taking the memory from one other board and putting it on the 3000, or instead moving the rambo off the dead one and adding it to either one of my working 480k ramdisks, also maybe stacking up more memory on both of those 480k ones, as i don't need more than 3 ramdisks i got working already, just need the rambo and expand memory, i think that best for now, as i don't have flexcards or consoles for my other 3 peb's anyway, so having two peb's fully stocked with 8 cards is good enough, with 4000 in my main machine and two 480k ones in my second system. i still have prototype 4000 horizon board from jim as well, never built it up, so i can always do that as well, in the future. as my older ramdisks not working are mainly the 8k ram chip ones, so they not much use anyway, i think i just rob the 32k chips i have one the one add them to my two 480k ones, and move the rambo one to one of them. anyone have manual for midwest one so i can expand it to max. with memory? plus what a good source in canada for memory chips for my exisiting original 4000 that only at 1536k i can expand it as well memory wise? Edited February 23 by Gary from OPA Quote Link to comment Share on other sites More sharing options...
atrax27407 Posted February 23 Share Posted February 23 Here ya go! Construction of Midwest RAMDisk.pdf Midwest RAMDisk.pdf 3 1 Quote Link to comment Share on other sites More sharing options...
+Ksarul Posted February 24 Share Posted February 24 Here's one other set of manuals I've found for it. CHICAGO RAMDISK WHOPPERCART MISSING LINK.pdf Horizon - Midwest.pdf 2 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.