Altirra 2.90 released

[Alfred]

What PC joysticks, if any, will work as an Atari joystick under Altirra ? Mostly seems to be only expensive flight sim sticks for sale for sale for pc's these days, hence the question.

[SS]

What PC joysticks, if any, will work as an Atari joystick under Altirra ? Mostly seems to be only expensive flight sim sticks for sale for sale for pc's these days, hence the question.

2600-daptor

http://www.2600-daptor.com

[phaeron]

What PC joysticks, if any, will work as an Atari joystick under Altirra ? Mostly seems to be only expensive flight sim sticks for sale for sale for pc's these days, hence the question.

 

Any controller that shows up in Control Panel as a gamepad will work with Altirra, and the default joystick setup should work for any controller with a D-pad. That having been said, it's best to avoid getting any controller designed for PCs and just get a Xbox or Playstation controller. My experience with PC controllers is that they are universally garbage, even the name brand ones.

 

Xbox controllers are automatically detected and will work in XInput mode, including the LB/RB buttons acting independently. The original 360 controller had a pretty bad D-pad that would give you diagonals unless you modded it, though. The Xbox One controllers are claimed to be better but I haven't tried one of them. A USB connection is easiest as getting them working over wireless requires adapters. I use a modified wired Xbox 360 controller for most testing.

 

PS4 controllers will also work will as long as you connect them by USB. Recent versions of Altirra will detect these and show PS4 button assignments. Don't try using them over Bluetooth as I have never gotten this to work -- they'll pair but not do anything. I have code lying around to access the accelerometer / gyro / touchpad on these but haven't gotten around to plugging it in.

 

In 5200 mode, simply booting the emulator without a cartridge will show a controller test you can use to check bindings. There's no built-in controller test yet for 800/XL mode.

[serj]

new turbo 2000 games.

record from the real cassette:https://yadi.sk/d/MQu7NXlh3Pchnd

result:https://yadi.sk/d/GLjTb_B83Pch5a

Edited November 12, 2017 by serj

[Mclaneinc]

Thanks Serj,

 

what does the read me say.....Its I presume in Russian but my notepad does not support the characters.to translate it..

 

Notepad++ to the rescue...

 

All translated..

 

For others the readme says

 

could not create a wav file for game # 26, although the * .xex file was obtained from the record.

side b:

could not translate 4 games, bad recording.

1.wav - game F15 strike eagle.

2.wav - the game is popeye.

10.wav - the game Air Strike.

21.wav - the game Los Angeles (SWAT).

 

 

 

Edited November 12, 2017 by Mclaneinc

[serj]

a question for Phaeron or baktra.

folder side b ---> error --->10.wav file

This file is loaded and running in the atari800 2.0.3 turbed emulator.

but, not works in altirra.

error or...

Edited November 12, 2017 by serj

[baktra]

10.wav. There is a dropout in the pilot tone of the data block. A turbo decoder or a turbo loader can consider one of the pulses in the droupout to be a synchronization pulse.

There can be subtle differences between atari800 2.0.3turbed and Altirra when it comes to interpreting the signal (e.g. neutral signal: wave file sample==0 - is it interpreted as logical 0 or 1 on the SIO data pin?). This might result in different interpretation. The sync pulse is detected in the middle of the droupout - see sample # 347564 in the WAVE file.

 

One note to atar800 2.0.3turbed - It performs absolutely no filtering of the signal. Just takes a sample and decides if it results in logical 0 or 1.

The sampled signal is somehow weak.

[WavePulseDecoder] C:\10.wav 2/0:44100
[T2000] Pilot tone found. Pulse boundaries: 11,23,47. {10196}
[T2000] Sync pulse found {26832}
[T2000] Pilot tone found. Pulse boundaries: 11,23,47. {40114}
[T2000] Sync pulse found {280144}
[T2000] HEADER: AIR STRIKE, LO:7048, LN:18432, RU:7048, TP:0 <OK: Success {282625}>
[T2000] Pilot tone found. Pulse boundaries: 11,23,47. {291151}
[T2000] Sync pulse found {347564} !!!!
[T2000] ERROR: Bad checksum {2710126}

Edited November 13, 2017 by baktra

[phaeron]

http://www.virtualdub.org/beta/Altirra-2.99-test19.zip

http://www.virtualdub.org/beta/Altirra-2.99-test19-src.zip

 

Fixes for Rapidus:

• HLE system is aware of Rapidus OS switching so that acceleration hooks no longer cause a crash when switching OSes.
• Fixed issues with the $FFD000-FFD7FF hardware mirror.
• Additional 16MB of banked SDRAM is now supported.
• Fast_mem_0 ($0000-3FFF write-through disable) is now supported.
• Hardware protect unit is now supported. (Who wants to write a VM?)
• Added .rapidus command to debugger to dump control register and EEPROM state.
• Implemented enough functionality in the 16-bit bootstrap PBI for the stock menu options to work -- this means you can flash the 6502 PBI and firmware images and the menu will be functional. The bootstrap flash will also apply options, but that's of limited use since it still has no menu.
• Rapidus, and several other "internal" device types, will now auto-reset the computer when added or removed.

Special thanks to laoo and drac030 for assistance.

 

Regarding the turbo tape, I don't think the glitch at 347564 is it. It's too early and the decode shows Altirra's AGC compensating for it with no issue in the decoded data (top is waveform, middle is threshold level amplified, bottom is turbo decoder output):

 

 

Ignore the notching in the square wave, that's just the analyzer's way of showing weak vs. strong detection. I need to remove that as it just makes the data hard to read.

 

This dropout is also far too early to be the problem. It's definitely a decoding error somewhere in the data and it's not significant enough to break the game -- if I bypass the checksum validation in the loader the game (mostly) seems OK. That implies that the error is late, possibly in the checksum itself. What complicates tracking this down is that the loader only uses a single flimsy XOR checksum byte for an entire block >4K, which is optimistic to say the least. This can fail to detect errors very often -- I have seen dozens of false negatives on a single Apple II disk because of such a weak checksum.

 

[baktra]

So it is time I stolen AGC from Altirra and implemented it in Turgen System's decoder... There is always something...

Edited November 14, 2017 by baktra

[Alfred]

What's the hardware protect unit ? Sounds interesting.

[drac030]

HPU is a circuitry in Rapidus which allows to intercept direct reads or writes (or read/writes) to the hardware registers at $d000-$d7ff.

 

When HPU is active and an access to that area is attempted, the CPU gets the abort interrupt.

Edited November 14, 2017 by drac030

[phaeron]

http://www.virtualdub.org/beta/Altirra-2.99-test20.zip

http://www.virtualdub.org/beta/Altirra-2.99-test20-src.zip

 

Polishes up the built-in Rapidus bootstrap firmware so you can actually configure the accelerator even if you don't have the real images to flash, so it's usable out of the box. Figured, as long as we're working with the decadence of a 20MHz accelerator and I have to implement the menu from scratch, might as well try to fancy it up a bit:

 

 

Note that the bootstrap flash is not meant to work on a real Rapidus and should not be flashed to one. The menu is "real" in that it uses stock hardware features, but the accelerator configuration takes shortcuts like not uploading the FPGA core and not following all setup and timing reqs for talking to the EEPROM. So don't do it. There is also currently no Rapidus OS in the image, though you could flash one into it.

 

Other changes:

• Removed the strong/weak indicator from the turbo waveform in tape analysis output. Begone, weird notches.
• Fixed MADS label parser not resolving banks properly for banks above $A0. (Stupid bug.)
• Fixed symbols not resolving in expressions above 64K.
• Relicensed the Additions disk and the .dumpsnap loader from GPLv2+ to permissive. This is just a formality as there is not a flood of people looking to enforce or violate the license on replacement firmware for 300 baud modems, but I noticed it while setting up license text for some other files and it's suboptimal for stuff that can be used independently of the emulator.

 

 

[flashjazzcat]

Any idea what could be causing this kind of corruption in the CPU history window?

 

 

Only crops up from time to time (this was just as a breakpoint was triggered). Possibly only in folded sections? Anyway: using 2.99 Test 19 on Windows 10 64-bit, DirectDraw and Direct3D 9 checked in display preferences. Only seemed to happen after the last big OS update, which I understand broke some aspects of display handling.

 

[phaeron]

Windows 10 version 1709 does indeed have some major and somewhat broken display changes, but as far as I know those are mainly in the Direct3D paths. What you've got there is an old fashioned GDI refresh bug and probably related to the rework I did on the history control to allow it to work on traces.

[phaeron]

Minor update:

http://www.virtualdub.org/beta/Altirra-2.99-test22.zip

http://www.virtualdub.org/beta/Altirra-2.99-test22-src.zip

 

• Rapidus added to help file.
• Rapidus NVRAM is now saved.
• Adjusted scroll bounds on history view.

 

[Mclaneinc]

Who stole beta 21

 

I take it that it was too minor to publish?

[phaeron]

Bug found after publish, skipped. Never reuse version numbers for different builds.

[Roydea6]

I have scanned through the DEVICE tree and have not found a stereo pokey item so that my Ultimate one Bios can be activated . And the AUDIO tree selecting Stereo does not activate it either.

My real hardware 800XL with u-switch actives okay, but my 130XE with a stacked pokey chip plays stereo even though the Bios is greyed out not seeing this older style up grade. [ Is there a wire I could add to the mix to have the Ultimate sense this old upgrade. ]

[phaeron]

Altirra doesn't currently support soft-switchable stereo, if that's what you're asking for.

 

Flashjazzcat would be able to answer authoritatively, but my guess is that the menu is sensing whether the switch output toggles the visibility of the second POKEY. In order for this to work with your stacked POKEY setup you would need a gate on the A4 to chip select connection so the second POKEY could be dropped out of the memory map. Without that your second POKEY is essentially always active and can be detected as such by software.

[R0ger]

Somebody halp ! I'm going mad. I'm quite sure it is (was ?) possible to set scanline brightness in Altirra. I mean brightness of the space between scanlines if you enable 'scanlines' in system/video. I have videos recorded from Altirra where the space between scanlines is pure black. But now it's like 50% of the scanline. Or did it simply change ?

[José Pereira]

Somebody halp ! I'm going mad. I'm quite sure it is (was ?) possible to set scanline brightness in Altirra. I mean brightness of the space between scanlines if you enable 'scanlines' in system/video. I have videos recorded from Altirra where the space between scanlines is pure black. But now it's like 50% of the scanline. Or did it simply change ?

Not sure if is what your refering to but on VIDEO if you enable PAL Artifacting it blends but this is real pixels scanlines.

But also on these Artifacting stuff if choose Artifacting High?

Maybe try different combinations on the VIDEO window for those 3options at the top: Frame Blending, Interlace and Scanlines.

Why not post a screen of what you have now?

[phaeron]

No, there was never an option to configure scanline brightness. I changed the brightness in between several versions back.

[SS]

Windows 10 version 1709 does indeed have some major and somewhat broken display changes, but as far as I know those are mainly in the Direct3D paths. What you've got there is an old fashioned GDI refresh bug and probably related to the rework I did on the history control to allow it to work on traces.

 

My WIN10 upgraded itself last night to v1709 and it did indeed break my ability to go fullscreen on some emulators such as Altirra 2.90 (and earlier) and the C64 emulator WinVICE. I downloaded Altirra 2.99-test22 and luckily it fixed my issue. No luck yet with WinVICE though; I'm sure that I'll find another program or two that will also have been affected as well. Stupid Microsoft, I hate that I am forced into these mandatory upgrades in WIN10.

 

Unfortunately, the "Atari 800 Best Game Pack" uses Altirra 2.40 so it is affected by this display problem. Are there any workarounds for fullscreen mode that I might be able to use to make version 2.40 work correctly again?

[phaeron]

 

My WIN10 upgraded itself last night to v1709 and it did indeed break my ability to go fullscreen on some emulators such as Altirra 2.90 (and earlier) and the C64 emulator WinVICE. I downloaded Altirra 2.99-test22 and luckily it fixed my issue. No luck yet with WinVICE though; I'm sure that I'll find another program or two that will also have been affected as well. Stupid Microsoft, I hate that I am forced into these mandatory upgrades in WIN10.

 

Unfortunately, the "Atari 800 Best Game Pack" uses Altirra 2.40 so it is affected by this display problem. Are there any workarounds for fullscreen mode that I might be able to use to make version 2.40 work correctly again?

 

Right-click on the .exe for the affected program, choose Properties, and under Compatibility check the "disable fullscreen optimizations" tab. This should bypass the broken code in Windows and let exclusive fullscreen mode work again. I didn't know that WinVICE was also affected, but I shouldn't be surprised given the number of AAA game titles that were also broken.

 

In Altirra, you can also use Direct3D 11 mode in Tools / Options / Display, but 2.40 is so old I don't know if D3D11 mode will work. Better to just replace it with 2.99-test.

 

Also, for each program that is affected, use Windows Feedback Hub (Win+F) to file an issue about it under Apps > (specific program). The feedback popularity contest is stupid, but it's the only way from the outside to get some priority on fixing it.

 

(Update: Checked WinVICE on 1709, and yeah, it's the same problem. Different symptom, in that only part of the screen isn't covered, but same cause: WinVICE is also trying to Present() from a background window and in 1709 this causes the foreground window to be cut out of the drawing area. Disable Fullscreen Optimizations works around it.)

Edited December 2, 2017 by phaeron

[SS]

 

Right-click on the .exe for the affected program, choose Properties, and under Compatibility check the "disable fullscreen optimizations" tab. This should bypass the broken code in Windows and let exclusive fullscreen mode work again. I didn't know that WinVICE was also affected, but I shouldn't be surprised given the number of AAA game titles that were also broken.

 

...

 

(Update: Checked WinVICE on 1709, and yeah, it's the same problem. Different symptom, in that only part of the screen isn't covered, but same cause: WinVICE is also trying to Present() from a background window and in 1709 this causes the foreground window to be cut out of the drawing area. Disable Fullscreen Optimizations works around it.)

 

Thanks! That seems to have fixed the problem for now.