phaeron Posted February 18, 2019 Author Share Posted February 18, 2019 http://www.virtualdub.org/beta/Altirra-3.20-test15.zip http://www.virtualdub.org/beta/Altirra-3.20-test15-src.zip Automatically held keys are now re-enabled when a soft-reset occurs through the WarpOS 32-in-1. AltirraOS 3.14: Fix for boot loop with BlackBox 2.16 firmware, which intercepts type 3 poll requests. The boot code now exits the poll loop on Y=$80 (user break). Fixed crash with no sound card. 10 Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted February 18, 2019 Share Posted February 18, 2019 (edited) Thank you! Premature I'm afraid...Still doing it.. Easy way to recreate, 32 in 1, Happy 1050 and boot Jumpman.. No key held and Jumpman says to remove carts (tried other software and BASIC is enabled) Shift F5 and hold Option / F4, same thing... Shift F5 with slight delay and hold Select / F3 and Jumpman boots fine.. I showed the key numbers just to prove I'm holding the right keys.. Clear my settings and made sure no ini file was ini (ha) there...All same.. Showing as beta 15 but will reset the cache and try again.. Edited February 18, 2019 by Mclaneinc Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted February 18, 2019 Share Posted February 18, 2019 (edited) Sadly its the same, even tried a few different XL OS's and 32 in 1 dumps.(of the 3 I know of) Sorry to be a pain.. ED: And I'm getting that odd thing where cold resets sometimes after holding select and cold resetting Jumpman then boots fine with no keys held but if you cold reset it again its back to saying remove cart.... And then some times it works with holding F4....Its like its holding a memory of other keys being pressed after a cold reset and sort of acting on the last set of keys held and then completely not doing it the next reset...Again some times if F3 has been held it acts like it was F4.. I am totally confused.... This is all from a complete cleaned out Altirra and using only the roms that ALtirra scans in plus manual insert of the 32 in 1 OS....Warp OS And to confirm that enable basic is not ticked in firmware and no Basic shows up in the menu bar.. Edited February 18, 2019 by Mclaneinc Quote Link to comment Share on other sites More sharing options...
phaeron Posted February 18, 2019 Author Share Posted February 18, 2019 Are you seeing [Option] show up in the lower right corner on the reset when the emulator should be holding it? The part I really don't understand here is the Select key (F3). Holding that after Shift+F5 shouldn't do anything with the standard OS unless you're using the XEGS OS. Quote Link to comment Share on other sites More sharing options...
Alfred Posted February 19, 2019 Share Posted February 19, 2019 Just curious, I noticed Axlon emulation goes to 4MB. Is there a reason for this ? I thought the largest Axlon compatible board for the 800 was the 1MB card from Magna Systems. Quote Link to comment Share on other sites More sharing options...
Nezgar Posted February 19, 2019 Share Posted February 19, 2019 Just curious, I noticed Axlon emulation goes to 4MB. Is there a reason for this ? I thought the largest Axlon compatible board for the 800 was the 1MB card from Magna Systems.Since the banking register for Axlon is not shared for any other purpose (unlike PORTB) all 8 bits are theoretically usable for bank selection. 256 banks of 16KiB = 4MiB. Although technically 1 of those 16K banks is your base bank so it's really 4096-16 = 4080KiB extended RAM 2 Quote Link to comment Share on other sites More sharing options...
Alfred Posted February 19, 2019 Share Posted February 19, 2019 Since the banking register for Axlon is not shared for any other purpose (unlike PORTB) all 8 bits are theoretically usable for bank selection. 256 banks of 16KiB = 4MiB. Although technically 1 of those 16K banks is your base bank so it's really 4096-16 = 4080KiB extended RAM Yes, I get that. However he refused to add 65816 memory beyond 4MB because that was the largest upgrade, until someone pointed out that Rapidus had all 16MB. So I'm just wondering why Axlon is 4MB. Is there an upgrade somewhere that gives 4MB to an 800 ? Quote Link to comment Share on other sites More sharing options...
Nezgar Posted February 19, 2019 Share Posted February 19, 2019 Yes, I get that. However he refused to add 65816 memory beyond 4MB because that was the largest upgrade, until someone pointed out that Rapidus had all 16MB. So I'm just wondering why Axlon is 4MB. Is there an upgrade somewhere that gives 4MB to an 800 ? Not sure about the 800, but the Antonia upgrade for XL/XE has 4MB, which can be configured for various combinations of PORTB, linear 65816, and Axlon banked RAM. 1 Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted February 19, 2019 Share Posted February 19, 2019 (edited) Are you seeing [Option] show up in the lower right corner on the reset when the emulator should be holding it? The part I really don't understand here is the Select key (F3). Holding that after Shift+F5 shouldn't do anything with the standard OS unless you're using the XEGS OS. Hi Avery, Sorry its a pain, I've added the roms I'm using, I totally isolated every thing and made a new folder, used the roms that came Rainbow or whatever it was called, the origina AtariXL.rom and the AtariBAS.rom and Nezgars 32 in 1 dump. That was the only dir I scanned and it found them, I manually added the 32 in 1 and set it to A for the warp os (although I think it defaults to that anyway). All setting were previously removed so its all from this dir only. Yes, the Option button comes up in bottom right but it seems to be booting with BASIC as Jumpman complains about carts. I then manually held the Option down and it stilled complained re cart. I sat there randomly trying lengths of time holding the cold start combination and even held the F3 just in case and randomly it would work and then a cold reboot brought back the cart complaint... Its all very odd, its like there's something mixed happening in the space of time from boot to the time the keys are held either virtually or not??? I don't know if that makes sense.. Anyway, roms added and the CRC of the XL rom suggests its a version 2 XL rom, the Bas is version 3 and Nezgars dump seems to work and I get the same problem on a different 32 in one dump. System scanned for any resident naughties and none.. Apologies for this...The simple answer is not to use the warp OS and 32 in 1 and I'm happy to do that but I just thought it was worth mentioning.. Paul.. roms used.zip Edited February 19, 2019 by Mclaneinc 2 Quote Link to comment Share on other sites More sharing options...
phaeron Posted February 20, 2019 Author Share Posted February 20, 2019 Okay, now things make sense. The issue is that the Warp OS has the Option button behavior reversed from normal -- holding Option enables internal BASIC instead of disabling it. The problem is that the emulator has no way of knowing this, so it tries to hold down Option to turn off BASIC and the Warp OS turns it on instead. This is a common mod, so the emulator has a setting you can check on OS ROMs to indicate that the Option button works the other way. You can only set this on loose OS ROMs, though, and not the slots in the 32-in-1. This means that you should flip the BASIC option and tell it to enable BASIC so it doesn't hold Option. There is an additional complication, which is that if the OS ROM setting in the emulator points to an OS ROM that is set to invert BASIC behavior, it will use that even though the 32-in-1 is overriding the OS ROM. I have an override in place for Ultimate1MB to disable this, which I need to extend to 32-in-1. 4 Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted February 20, 2019 Share Posted February 20, 2019 (edited) Thanks Avery, I thought I was going mad, I'm betting that depending on if you choose to hold down F4 by mistake that itself muddles the way it boots. Thanks for looking at it, at least it was a genuine issue but not really with Altirra other than you kindly compensated for holding down the OPTION internally....And hopefully that will not get changed Look forward to the change for the 32 in 1...Hardly the most pressing change in terms of your time Edited February 20, 2019 by Mclaneinc Quote Link to comment Share on other sites More sharing options...
Kyle22 Posted February 22, 2019 Share Posted February 22, 2019 This may be a stupid question, but: I just downloaded Curt's Integrator ROM. I believe it goes @ $C000. Is there currently a way in Altirra to add a $C000 block ROM to the OS? Doesn't OmniMon, etc. also work this way? Quote Link to comment Share on other sites More sharing options...
phaeron Posted February 22, 2019 Author Share Posted February 22, 2019 This may be a stupid question, but: I just downloaded Curt's Integrator ROM. I believe it goes @ $C000. Is there currently a way in Altirra to add a $C000 block ROM to the OS? Doesn't OmniMon, etc. also work this way? Yes, 3.10 will let you use a 16K XL/XE OS ROM image layout in 800 mode. The first 4K of that image is mapped to $C000-CFFF, and the last 10K to $D800-FFFF. 4 Quote Link to comment Share on other sites More sharing options...
Alfred Posted February 24, 2019 Share Posted February 24, 2019 Bug in the debugger. It's duplicating bytes, must be an error in advancing the PC. Example: 42D0: 60 RTS 42D1: 20 29 32 JSR $3229 42D4: 20 C8 42 JSR $42C8 42D6: 42 KIL 42D7: 85 D4 STA FR0 42D9: A2 00 LDX #$00 42DB: 86 D6 42D6 appears twice, first as part of the JSR at 42D4 and then again as 42D6 by itself. So it incremented the PC correctly but the data byte fetcher only advanced 2 and not three bytes looks like. Quote Link to comment Share on other sites More sharing options...
phaeron Posted February 24, 2019 Author Share Posted February 24, 2019 Bug in the debugger. It's duplicating bytes, must be an error in advancing the PC. Example: 42D0: 60 RTS 42D1: 20 29 32 JSR $3229 42D4: 20 C8 42 JSR $42C8 42D6: 42 KIL 42D7: 85 D4 STA FR0 42D9: A2 00 LDX #$00 42DB: 86 D6 42D6 appears twice, first as part of the JSR at 42D4 and then again as 42D6 by itself. So it incremented the PC correctly but the data byte fetcher only advanced 2 and not three bytes looks like. You'll have to be more specific as to where you're seeing this. If it's in the Disassembly window, this is intentional and it happens when the current PC or frame address doesn't line up with the prior disassembly. The debugger has to do a reverse disassembly in order to determine instructions prior to the given address. Since those instructions may not have been executed and the CPU never executes instructions backwards, this is not well defined. The reverse disassembler tries to find a reasonable path to the anchor address, but if it doesn't, the disassembler will force the anchor address to appear in the disassembly. For instance, in the above, if the CPU jumped to $42D6 and crashed, you would see the above because that address is where the PC is actually at even though the PC would never get there in the normal instruction sequence. 3 Quote Link to comment Share on other sites More sharing options...
Alfred Posted February 24, 2019 Share Posted February 24, 2019 Yes, this is in the disassembly window. I set the address to $42C8 and that is the output I see. It's all code in that section, no data bytes to mess things up. I wasted quite a bit of time poking into CLSN pascal before I realized the disassembly had to be wrong, and then there it was. Quote Link to comment Share on other sites More sharing options...
phaeron Posted February 25, 2019 Author Share Posted February 25, 2019 I'd have to see the entire disassembly window contents to diagnose. The only way I can see this happening is if the target address was set to $42D6. Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted March 1, 2019 Share Posted March 1, 2019 Cosmetic observations: I notice on non-65C816 machines (i.e. with only 64K of linear RAM), the memory window still does not wrap back to $0000, but shows an inaccessible numbered segment: Also (not sure if it even matters or is intentional or not), the IDE IO indicator overwrites the file size indicator on the OSD when a video is recording: Quote Link to comment Share on other sites More sharing options...
Farb Posted March 1, 2019 Share Posted March 1, 2019 Is anyone else having a problem with full 1050 emulation? When I enable it, it just drops to self test on reboot like a drive isn't attached... Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted March 1, 2019 Share Posted March 1, 2019 Have you checked the firmware? It might be missing in the firmware manager, it will also show if its missing in the Device adding menu... Quote Link to comment Share on other sites More sharing options...
+DjayBee Posted March 1, 2019 Share Posted March 1, 2019 Is anyone else having a problem with full 1050 emulation? When I enable it, it just drops to self test on reboot like a drive isn't attached... With Altirra 3.10 you need to coldstart only the computer. Otherwise it seems that the disk drive is still reseting when the Atari tries to boot. 1 Quote Link to comment Share on other sites More sharing options...
Mclaneinc Posted March 1, 2019 Share Posted March 1, 2019 (edited) Ah, just saw it was 3.10 you are using..Seriously, go to 3.20 and the latest beta, there's been so many fixes since 3.10...Its perfectly stable.... Works here...I have a mouse gesture bind to do a standard cold start and the emulated 1050 (full) boots every time..Latest beta.(15) There's also a Power on delay setting you can adjust... Just to reaffirm, Phaerons beta's rarely ever have regressions or introduce extreme faults let alone any faults, the old wait for a stable release syndrome is not really an issue with Altirra... Edited March 1, 2019 by Mclaneinc 2 Quote Link to comment Share on other sites More sharing options...
flashjazzcat Posted March 4, 2019 Share Posted March 4, 2019 Crash every time I try create an ATR (360K DS/DD), drag these two files into it and then try to delete the ROM file: Scoured the HDD for the mdmp file but can't find it. Quote Link to comment Share on other sites More sharing options...
Alfred Posted March 9, 2019 Share Posted March 9, 2019 The SpartaDos Menu program fails with error 130, which implies the D: handler is missing. I have H: and P: intercepts on but not CIO Burst, but I don't see how those would cause that error. Quote Link to comment Share on other sites More sharing options...
phaeron Posted March 9, 2019 Author Share Posted March 9, 2019 http://www.virtualdub.org/beta/Altirra-3.20-test16.zip http://www.virtualdub.org/beta/Altirra-3.20-test16-src.zip Inverted Option flag for internal BASIC is forced off when 32-in-1 is active. AltirraOS 3.15: Adjusted ICBALZ/ICBAHZ to improve compatibility with Lightspeed DOS. Fixed the Customize button being enabled too often in the keyboard customization page (when custom layout isn't enabled). Fixed an issue with Shift key down not being reflected in SKSTAT while another key was already being held. (Thanks to MaPa) The Recommendations page now has a note if the keyboard is set to cooked mode. Fixed crash in Disk Explorer when write flush fails due to another program locking the disk image file, like a virus scanner. Disk Explorer now forces read-only mode when the disk image file being mounted is marked read-only. The SpartaDos Menu program fails with error 130, which implies the D: handler is missing. I have H: and P: intercepts on but not CIO Burst, but I don't see how those would cause that error. I can't reproduce this with the menu utility in SpartaDOS 3.2g. The one in SpartaDOS X doesn't use CIO at all to access the disk, so the options you mentioned wouldn't matter for that version. 2 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.