Altirra 3.90 released

[phaeron]

Version 3.90 of my emulator Altirra is out:

http://www.virtualdub.org/altirra.html

 

Thanks for everyone's continuing support, whether it be bug reports, feature requests, discussion, trying out the helper programs on real hardware, etc. 3.90 final is essentially the same as 3.90-test34, except for release changes.

 

Although previous versions tended to be about six months apart, it's almost been a year since 3.20, so I figured I'd better hurry up. Highlights of the 3.90 release:

 

• Accuracy: 800 System Reset timing fixed, undocumented RMW and WSYNC timing fixes, several fixes to 65C816 direct page wrapping and Veronica, many fixes to FDC/RIOT/6809 for full disk drive emulation, improved POKEY two-tone mode emulation, more accurate power-up hardware state.
• Debugger: Improved disassembly window with automatic block separation and inline call target preview, more disassembly options, better loop detection in the history window, Alt+Shift+click to jump to history for a pixel.
• Disk drives: 810 Turbo, Amdek, and Percom AT-88 full emulation; easier file import/export in the Disk Explorer.
• Display: Improved PAL artifacting and color defaults, gamma-corrected frame blending, color setting import/export, PERITEL and monochrome monitor emulation, fixes to color correction logic (esp. with VBXE).
• Firmware: Updated AltirraOS 3.26 for improved compatibility with hardware addons and software, improved autodetection for custom OS ROMs.
• Tape: Faster emulation especially in warp mode, audio filter compensation for better turbo decoding, and enhanced debugging for tape issues.
• Video recording: Aspect ratio correction, scaling, and direct H.264/WMV compression support through Media Foundation.
• UI: Dark theme, improved audio monitor/scope, improved timing for slightly reduced latency.

 

With 3.90 done, it's now time to start the 4.00 test series:

http://www.virtualdub.org/beta/Altirra-4.00-test1.zip

http://www.virtualdub.org/beta/Altirra-4.00-test1-src.zip

 

First, there are a couple of breaking changes in 4.00:

 

• End of support for Windows XP SP3 and Vista. The new minimum operating system requirement is Windows 7 SP1.
• End of support for DirectDraw, which has not been native since Vista, and OpenGL, of which only old versions were used anyway (and was not enabled by default).

 

New features and fixes:

 

• Palette solver: Matching color palettes in Altirra has been a long-standing issue, since it is unable to take .pal/.act files due to needing the generation parameters for the palette rather than the final colors for various features, and matching colors from real hardware is a difficult task. Pretty much everyone has their opinion on what "true colors" should be. Well, in 4.00-test1 there is now a solver to guess the parameters for you:
  
  
  
  You can either give it a PAL/ACT file, or you can give it a picture -- and although the picture needs to be a good one, it doesn't have to be perfectly straight as you can align it to compensate for projection by dragging the corner dots. It's compatible with any show-palette tool for the Atari that does 16 lumas across and 16 hues down. Click Match, and it'll start grinding on the parameters to try to get as close as possible, and tell you how good the match is. Or, you can have it overlay the current palette as dots over the picture to check the palette while manually tuning it.
   
• Debugger: On-screen watches now update continuously while stepping and the 'wx' command has options for hex formatting, bank-sensitive debugging is now supported for SpartaDOS X cartridges, improved handling of subdirectories when looking for source files, and the console output window is faster when flooded with output. Oh, and the Memory window has been improved with scrolling and more display options:
  
  
   
• Display: Added pure white monochrome mode, and fixed a bug in the high artifacting engine where chroma artifacts moved in the wrong direction for ascending hues.
• XEP-80: Fixed an emulation bug in the way that scrolling occurs, and added a couple of new toys to the Additions disk: an XEPVHOLD.COM tool to reprogram the XEP80's timing chain for a shorter display that's less likely to roll on modern displays, and a new ultra-speed ALTXEP8U.SYS driver that runs at symmetric 31.5Kbaud instead of 15.7K or 31.5K send / 15.7K receive.
• Tape: Added support for KSO Turbo 2000.
• Disks: Added emulation for the Percom AT88-SPD, and added 1791/1795 FDC selection for the AT-88. Fixed support for virtual SpartaDOS disks with directories and files whose names start with periods. Added a polling workaround option for virtual disks for environments where file change notifications don't work (some Wine environments).
• Audio: More accurate emulation of uneven volume bits, and rewrote cycle-level filtering emulation for improved high frequency aliasing rejection during downsampling.
• UI: Dark mode now has reskinned buttons. Added support for auto-hiding the menu bar. Fixed mouse wheel scrolling when the OS page-at-a-time option is on.
• H: device: Lifted 16MB limit for binary (untranslated) access, and fixed errors not being returned properly during burst I/O accesses.
• AltirraOS 3.27: Fixes to the printer handler for EOL handling, particularly for the Atari 1025, and improved compatibility of variable usage to work with Monkey Wrench II.
• Custom Devices: More scripting language constructs like break, while/do-while loops, and forward declaration of functions, better threading support, and more scripting methods for manipulating memory. Custom devices can now raise Parallel Bus Interface IRQs. Custom video outputs can now be created with either text or graphical output, and even support text select/copy like the emulator's built in ANTIC and XEP-80 video outputs. New sample custom devices: Bit-3 80 column, 1090 80-column, and a PBI-based metronome.

[TGB1718]

Many thanks for the update to this wonderful piece of software

[Mclaneinc]

Yes and its entered a new era which is nice but for those not wanting to upgrade their windows then Altirra is already at a stage where its amazingly good.

 

Going to have a play with the colour solver, looks fun.

 

Thanks Avery...

[Keatah]

Question/Comment: on the audio sub-section. I understand audio can be very subjective, and it's been years since I payed attention to real hardware in-depth. And I didn't hook up a scope.. So..

 

I noted that in order to hear what I think should be correct sound in Star Raiders hyperwarp I needed to turn on the Non-Linear Mixing option. This in 4.00 test-1. If I don't turn it on there seems to be noticeable discrete steps or bumps in the volume occurring 1.2 and 3.8 seconds into the hyperwarp.  Turning on Non-Linear Mixing fixes this nicely.

 

In 3.90 I didn't need to turn on Non-Linear Mixing. And it sounds good either way on or off. This supposed to work this way?

 

Anyhow thanks for the white phosphor option and we all love the ongoing work!

 

Feature Request:

A variable-speed turbo option. With today's hardware the Warp-Speed can become quite silly. Try playing Defender at 2700 FPS! So I was thinking of something like either a series of hot-keys for like 2x 3x 4x and so on with the last one being unlimited. Or maybe a dropdown menu item where you just specify how fast "F1" can go.

[Keatah]

Minor UI in:consistency:

 

Has to do with warp speed. Start by running in a standard window. Then turn on warp by clicking on the drop-down SYSTEM and WARP option. Now hit <alt>-<enter> to go full screen. Good. Now look at the status in the drop-menu. It shows warp speed is not checked. Then if you select it while in full screen, it shows the check mark but toggles to normal speed. IIRC this has been present in older releases.

 

 

[marcokitt2000]

Thank you for new release and new start of version 4.0.

 

 

Keep the good work.

 

 

Gr.

Marco

[Eyvind Bernhardsen]

3.90 works fine in Wine (CrossOver) on my Mac, but Version 4.00 doesn't: the window comes up, then it immediately generates a ~120kB .mdmp file and locks up. The menu is there but nonfunctional, the rest of the window is a dark gray, and I have to kill the process to get rid of it.

[Yaron Nir]

@phaeron, excllent release! many thanks

 

as for next versio 4.0

 

i am still facing the goto source issue with agreggated project folder, finding my self needing to select the source file myself, even though the folder to pick the file is correct.... will that be solved? 

also would be nice to be able to search for routines in the c file. currently the search doesn't do anything....

 

[phaeron]

12 hours ago, Keatah said:

In 3.90 I didn't need to turn on Non-Linear Mixing. And it sounds good either way on or off. This supposed to work this way?

No, this is a bug, I screwed up the linear curve code when I rewrote the mixing table (it now has to be split by volume bits). Will be fixed in -test2.

 

12 hours ago, Keatah said:

A variable-speed turbo option. With today's hardware the Warp-Speed can become quite silly. Try playing Defender at 2700 FPS! So I was thinking of something like either a series of hot-keys for like 2x 3x 4x and so on with the last one being unlimited. Or maybe a dropdown menu item where you just specify how fast "F1" can go.

Adding multiple commands for different speeds is probably a bit much, but I could see a warp speed limiter. Though I'm not sure why you would play Defender at warp at all, I suck more than enough at it at 60 fps.

 

11 hours ago, Keatah said:

Has to do with warp speed. Start by running in a standard window. Then turn on warp by clicking on the drop-down SYSTEM and WARP option. Now hit <alt>-<enter> to go full screen. Good. Now look at the status in the drop-menu. It shows warp speed is not checked. Then if you select it while in full screen, it shows the check mark but toggles to normal speed. IIRC this has been present in older releases.

Bug, will be fixed in -test2. The issue was that the menu was updated too late, after the menu had already been created. It refreshes if you select another menu temporarily.

 

8 hours ago, Eyvind Bernhardsen said:

3.90 works fine in Wine (CrossOver) on my Mac, but Version 4.00 doesn't: the window comes up, then it immediately generates a ~120kB .mdmp file and locks up. The menu is there but nonfunctional, the rest of the window is a dark gray, and I have to kill the process to get rid of it.

Usual caveat regarding running under Wine, but if you could attach or PM me the .mdmp file I might have a chance at figuring out what's going on. I wasn't able to reproduce with Wine 5 on Ubuntu, though it was running on llvmpipe since I couldn't get VMware 3D hardware acceleration working. If you get a .mdmp file, it's crashed, everything after that doesn't matter.

 

I have noticed that Wine does really bogus things if decent graphics hardware isn't available, such as trying to restrict Direct3D 9 APIs to using 1MB of VRAM. Doesn't really work when you can't even create the swap chain to cover the screen. You can try launching with the /gdi flag to disable all hardware acceleration, which will identify if the issue is graphics related (and thus may related to the DirectDraw removal). I wouldn't expect you could run any moderately current version of macOS with a graphics card that's too old to emulate D3D9 pixel shader 2.0, though.

 

8 hours ago, Yaron Nir said:

i am still facing the goto source issue with agreggated project folder, finding my self needing to select the source file myself, even though the folder to pick the file is correct.... will that be solved?

No current plans, if I do make changes you'll see it in the change lists.

 

[Eyvind Bernhardsen]

5 hours ago, phaeron said:

Usual caveat regarding running under Wine, but if you could attach or PM me the .mdmp file I might have a chance at figuring out what's going on. I wasn't able to reproduce with Wine 5 on Ubuntu, though it was running on llvmpipe since I couldn't get VMware 3D hardware acceleration working. If you get a .mdmp file, it's crashed, everything after that doesn't matter.

 

I have noticed that Wine does really bogus things if decent graphics hardware isn't available, such as trying to restrict Direct3D 9 APIs to using 1MB of VRAM. Doesn't really work when you can't even create the swap chain to cover the screen. You can try launching with the /gdi flag to disable all hardware acceleration, which will identify if the issue is graphics related (and thus may related to the DirectDraw removal). I wouldn't expect you could run any moderately current version of macOS with a graphics card that's too old to emulate D3D9 pixel shader 2.0, though.

 

This is on my 2015 iMac (AMD Radeon R9 M395X with 4 GB VRAM), so it should be okayish; I tried running with /gdi but got the same crash. Will PM you the .mdmp.

[avanze]

Amazing work. Once again, thank you.

[Krakerman]

Is there a fix for Atari 5200 Galaxian? Player ship keeps moving to right.

[+DrVenkman]

6 hours ago, Krakerman said:

Is there a fix for Atari 5200 Galaxian? Player ship keeps moving to right.

That's something to do with your controller setup. I just started a game of 5200 Galaxian using Keyboard --> Absolute (port 1) and it worked fine.

 

[Krakerman]

I did the same exact thing you did pictured above to test and does same thing still goes to the right.

[+DrVenkman]

8 minutes ago, Krakerman said:

I did the same exact thing you did pictured above to test and does same thing still goes to the right.

Hmm. I just tried again and got the same behavior you describe - immediate movement to the right. You can override that with the left arrow but as soon as you let go, your ship goes right.

 

HOWEVER - when I tried changing to Keyboard Relative control, and then back to Keyboard Absolute, it started working correctly. Is this a bug, @phaeron, or just a quirk of emulating the 5200's odd analog control scheme?

[Yautja]

You need to update the ver. from 3.26 to 3.90 in the initial screen; and also the copyright year to 2020.

 

 

 

[+DrVenkman]

28 minutes ago, Yautja said:

You need to update the ver. from 3.26 to 3.90 in the initial screen; and also the copyright year to 2020.

Those are the version number and copyright dates for the Altirra written-from-scratch OS, not the emulator itself. 

[Yautja]

13 minutes ago, DrVenkman said:

Those are the version number and copyright dates for the Altirra written-from-scratch OS, not the emulator itself. 

I've checked the previous releases. And the versions and dates change.

[dmsc]

Hi!

46 minutes ago, Yautja said:

I've checked the previous releases. And the versions and dates change.

Not anymore, starting at 3.20, See: 

 

Have Fun!

[phaeron]

3 hours ago, DrVenkman said:

Hmm. I just tried again and got the same behavior you describe - immediate movement to the right. You can override that with the left arrow but as soon as you let go, your ship goes right.

 

HOWEVER - when I tried changing to Keyboard Relative control, and then back to Keyboard Absolute, it started working correctly. Is this a bug, @phaeron, or just a quirk of emulating the 5200's odd analog control scheme?

It's a bug in 5200 Galaxian. The joystick calibration routine is broken and splits by even/odd POT values instead of port 1/2; it only happens to work on the real console due to pot noise. You'll only see this if you are using keyboard or D-pad in absolute mode; keyboard in relative mode or an analog stick will feed enough different values to eventually kick the calibration routine into behaving.

 

I have a workaround for this to add a small amount of jitter, but it appears to have been broken by, of all things, a timing change I did to reduce input latency. The input timing is slightly different and so the jitter doesn't happen at point for Galaxian to see it. I'll have to add some more noise to work around it more.

 

2 hours ago, Yautja said:

You need to update the ver. from 3.26 to 3.90 in the initial screen; and also the copyright year to 2020.

As others have said, the version being different is intentional, but I'll fix the copyright year the next time I rev it.

[Faicuai]

On 6/15/2020 at 1:49 AM, phaeron said:

(..)

• XEP-80: Fixed an emulation bug in the way that scrolling occurs, and added a couple of new toys to the Additions disk: an XEPVHOLD.COM tool to reprogram the XEP80's timing chain for a shorter display that's less likely to roll on modern displays, and a new ultra-speed ALTXEP8U.SYS driver that runs at symmetric 31.5Kbaud instead of 15.7K or 31.5K send / 15.7K receive.
• (..)
• AltirraOS 3.27: Fixes to the printer handler for EOL handling, particularly for the Atari 1025, and improved compatibility of variable usage to work with Monkey Wrench II.
• Custom Devices: More scripting language constructs like break, while/do-while loops, and forward declaration of functions, better threading support, and more scripting methods for manipulating memory. Custom devices can now raise Parallel Bus Interface IRQs. Custom video outputs can now be created with either text or graphical output, and even support text select/copy like the emulator's built in ANTIC and XEP-80 video outputs. New sample custom devices: Bit-3 80 column, 1090 80-column, and a PBI-based metronome.

 

First of all, congrats for reaching this (monumental) v4.0 milestone... Hard to believe how far and high this has gone... I know that (one of these days) all this greatness may well have to stop (as nothing runs forever)... But in the mean time, I don't even want to think about that day...

 

Secondly, congrats as well for this sweet v0.6 Ultra-mode XEP80 driver. I can confirm that its performance essentially matches that of 5x ACCELERATED E: on Gr.0 mode (HYPE and SDX Quick ED's), char by char, especially on "console" mode output, when sending characters to the screen, linearly sequenced. That is NO small feat, by any means. It is FAR, FAR ahead of SDX's own XEP80 driver, throughput wise.

 

(UPDATE: here are the timings, a full dump of FastBasic Manual, via SDX's regular TYPE command:

 

1. XEP80-Ultra: 1m:30sec

2. SDX QED: 1m:33sec.

3. HYPE-E: 1m:34sec.

 

KUDOS!!!)

 

Now, what I think it would be special is to have this latest v0.6 also set for PAL output, since it will render almost a perfect 80x24 grid on Capital letters, and almost perfect on lower-caps, except the last ~two scan-lines. Your utility is really nice for extracting the 25th line, but it does cut one or two-lines on each character row, which makes anything that uses special characters looks a bit weird on screen (I know XEP80 NTSC font is disproportionately tall vertically and seemingly narrow horizontally). Maybe if you could run your config. utility on PAL mode, and it may solve this problem, automatically.

 

Thanks, again, for everything!

 

 

 

Edited June 17, 2020 by Faicuai

[Krakerman]

@phaeron Thanks for the update and look forward to the Galaxian workaround!

 

[Gearhead Lu]

Thank you phaeron for the emulator and update. It is great that you do this work to keep Atari 8 bit alive and for those that don't have the hardware.

 

When I run any atr, atx file of Dimension X by Synapse on Altirra, thetext and "Desert Map" (square indicators at top right) are grey with vertical lines and should have color. Same problem v3.2 or v3.9.

When I run the same files with SDrive-Max or from the original disk on my 800, the squares have color.

 

Dimension X on physical Atari 800.

 

Altirra 3.2 or 3.9

Do I have something set wrong?

Thanks

[phaeron]

28 minutes ago, Gearhead Lu said:

When I run any atr, atx file of Dimension X by Synapse on Altirra, thetext and "Desert Map" (square indicators at top right) are grey with vertical lines and should have color. Same problem v3.2 or v3.9.

When I run the same files with SDrive-Max or from the original disk on my 800, the squares have color.

Turn on artifacting.

[flashjazzcat]

If I set a breakpoint on read in zero page (e.g. ba r $80) using 4.00-test1, the breakpoint immediately triggers and messages run away in the console window, apparently regardless of whether the location was actually read.

 

EDIT: And just like that, I close down the emulator and re-start it again, and ba works as expected. Weird.

 

Edited June 17, 2020 by flashjazzcat