Search the Community
Showing results for tags 'testing'.
-
Let's make the testing a bit more structured and useful. So please put all your Atari 2600 games, controller etc. related test results in here. If possible, verify your results with original consoles. First, there are numerous games (especially homebrews) which are known to not work (yet). In brackets you find the bankswitching type name of Stella. Pitfall II (DPC) Activision (FE) (most likely): Decathlon, Space Shuttle, Robot Tank - Note: there exist cart versions (e.g from HES) which use Atari's F8 and thus work M Network (E7): Bump'n Jump (not the Telegames version!), Burgertime, Masters of the Universe CBS (FA): Mountain King, Omega Race, Tunnel Runner CommaVid (CV): MagiCard, Video Life Megaboy (F0) CompuMate (CM) Arcadia Supercharger (AR) games Melody enhanced homebrews: DPC+ (e.g. Space Rocks, Stratovox...) CDF(J) (e.g. Draconian, Galagon, Lode Runner, Mappy, Stay Frosty 2...) FA2 (Star Castle Arcade) CTY (Chetiry) Homebrews using EF, DF or EF bankswitching: e.g. Zippy the Porcupine, Penult Toyshop Trouble (0840) (there exist carts which use standard F8 bank switching and thus work) Boulder Dash (3E) Circus Convoy (SB) Stella's Stocking (X07) Realsports Boxing (F8) (there is a bug in the dumper which affects this game) Thrust+ Platinum (constant thrusting because properties cannot be found) Also some rare carts/prototypes like Pink Panther (WD), Pleiades (UA) or certain Brazilian bank switching types will not work. All other games should work. Exceptions found so far: certain bankswitched white Coleco carts seem to have problems Also not working: Harmony cart works only for 4K games in developer mode PlusCart As of now only standard Joysticks and Paddle seem to be supported. Sega Genesis and BoosterGrip might work (please test). Unsupported controllers: Keyboard, Video Touch Pad Driving controllers Lightgun Mindlink KidVid Trackballs (CX-22, CX-80, Atari or Amiga Mouse ) (in real trackball mode, works in joystick mode) AtariVox, SaveKey QuadTari(?) Joy2B+(?)
-
Let´s collect test results for the 2600+ regarding its ability to play 7800 PAL carts! Please add more! (and let´t keep it PAL 7800) ---- unplayable: - Ballblazer is very slow, glitches, unplayable - Commando (resets when you press fire to start) - Choplifter (black screen) - Food Fight (black screen) - Karateka (black screen) - Ms. Pac Man (crashed screen) - Joust (black screen) playable, but messed up: - Galaga (playable, but bottom quarter of screen not visible) - Kung-Fu Master (playable, but background graphics messed up) - Title Match Pro Wresting (playable, but sound issues) - Tower Toppler (playable, but tower graphics differ from original)
-
Anyone who prefers to use AtariAge as their bug reporting thread, can do so here. Please try to provide a concise description, doesn't have to be verbose, but enough information to reproduce the problem, e.g. if the problem is with a disk image, please provide the disk image (or a TNFS location). We'll try to fold this into github issues, so we can keep track. -Thom
-
Hi @Ben from Plaion and @Albert Here is my suggestion for what to include in the dumper fix. The dumper creates a 128 byte header for the game and sends it to the ProSystem emulator. Currently the dumper picks a few bits from the ProSystem.dat to guess what the header should look like. Having a copy of this 128 byte header in the ROM would take out the guessing work. Simply let developers make a copy of the a78 header by adding the header at $FF00 in the game ROM. This is the logical place for it as the signature for NTSC carts is at $FF80. The cart would work correctly on a real Atari 7800 (NTSC) and thanks to the info in the ROM the dumper could send the correct settings to the emulator as well. If the data in the dumped rom has the data: (version_byte+'ATARI7800') at $FF00 the dumper just copies the 128 bytes from there and sends it to the emulator followed by the dumped game. If the (version_byte+'ATARI7800') is not there then the dumper goes back to guessing. Oh, and thanks for making a really modular and elegant 2600+. The more I have played with it - the more I like it. Edit: I just added a small example file with the header in the ROM. The end of the file with the header, the signature and the vectors: 007f80 004 A T A R I 7 8 0 0 \0 \0 \0 \0 \0 \0 007f90 \0 E x a m p l e \0 \0 \0 \0 \0 \0 \0 \0 007fa0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 007fb0 \0 \0 \0 200 \0 \0 \0 001 001 \0 \0 \0 \0 \0 \0 \0 007fc0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 007fd0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 007fe0 \0 \0 \0 \0 A C T U A L C A R T 007ff0 D A T A S T A R T S H E R E 008000 \t 245 300 324 321 \b 031 \r 253 340 307 ! 326 I 375 [ 008010 265 3 210 ' 367 211 ^ ' L 370 B & > n 207 227 008020 257 W 314 346 5 266 220 303 344 | d 355 V 016 300 017 008030 352 367 255 022 ~ 301 036 277 ? 367 360 257 N 247 313 0 008040 1 W b 203 332 \a P 371 257 | \ 204 i 365 230 276 008050 311 354 \v 270 266 A [ 373 ! + 250 016 006 276 O 374 008060 263 332 201 s y O + \f 344 211 177 275 4 017 + C 008070 204 210 024 356 235 m 310 306 377 207 ` 200 \0 200 a 200 Example.a78
-
Jupiter SS Pants changer in CBS RAM I recompiled my 10 line BASIC contest game entry Jupiter SS with Atari Flashback BASIC. The Manual and SuperCharger version are here. No changes were made to the code, but the Pants changed to black on the Harmony Cart. The Pants may be Rainbow on the Retron77 and White or Black on other Flashcarts. Unstable binary This binary may also require repeat attempts to load, or loading of a functioning program before it will start on some emulation platform setups. From these differences I am unsure what color the Pants and Black Star would be under actual CBS RAM. I am interested if someone can modify an existing CBS RAM cartridge, into an EPROM board for accurate testing. Diagnostic Tests so far -------------------------- Javatari testing is showing Black Pants like on the Harmony Cart: https://javatari.org/?ROM=http://relationalframework.com/CBSRAM_Jupiter_SS_Firestorm_GAMMA.bin&SCREEN_FULLSCREEN_MODE=1 Other differences observed: ------------------------------- In Javatari "The beat" is missing from the soundtracks on the descent but I think that is a sound reproduction bug specific to Javatari, the 5 chiptunes are algorithmically generated and use the Tracker sequencer to change the envelope to shape the notes. Also noticed a "phatter sound" in the SuperCharger version, likely from minute differences in the timing for shaping the sound envelope. Expected "easter egg" changes: ----------------------------------- There are cyan clouds that animate and can turn black when hit by the BlackStar in the CBS RAM version, these are an artifact of the top two layers of the virtual world holding the player colors and virtual world color arrays, CBS RAM has only 256 bytes while SuperCharger RAM is unlimited): Atari Portable unit Fix: This Atari Flashback program will not yet work on the Atari Flashback portable because the program is too small, I can add "ballast code" to make it run there (that is the fix for small programs as per the manual) as well and the Pants should be white with the Portable device. I'll add that ballast code so we can confirm white Pants, and so that the game is playable on the Atari Portable, presently. Currently there are no changes to the BASIC program, and no ballast code to make it easier to test compare the binaries. CBSRAM_Jupiter_SS_Firestorm_GAMMA.bin
- 1 reply
-
- cbs ram
- atari flashback basic
-
(and 4 more)
Tagged with:
-
Hello, I have found some time to continue my work on "porting" DIS6502 described here: http://atariage.com/forums/topic/231300-the-best-dissassembler-for-atari/ I have now written the conversion, so the new version can read the workspace (*.WRK) files created by the old versions of DIS6502. Now I need such files for testing. There I intentionally post this here outside of the programming forum, hoping to collect some example .WRK files you may have created since 1998 :-) - Peter.
-
Hey all, I hope this hasn't been asked previously but I was wondering what some of you guys do when you're buying hardware from unknown individuals but you want to test it before handing over cash. I don't have an diagnosis cart so I'm thinking of doing these simple things. Insert a cart and play the game a little and game port Test out all of the keys using notepad Inspect for signs of significant damage Run a BASIC command to check for total amount of RAM or test areas of RAM for flakiness. (Would be interested in this BASIC command if someone has one handy) Would love to test the other ports but I think that would require much more setup time so I'm shooting for the minimum time for the maximum output. I'm just considered with the computer itself. Testing peripherals would be nice but not necessary. OK, retro computerphiles. Hit me with your best shot (fire away)!
-
Hello everyone! I need to make sure that I'm testing my XEGS power supply correctly. Most pinout diagrams on Internet don't specify if the pinout is viewed from the connector's point of view or from the console's plug. Using the pinout configuration displayed on the guides, I get -5VDC instead of 5VDC in the respective pins; therefore, I assume that the guides show the pins reversed (i.e., shown from the console's plug), but I want to make sure if I'm doing the right thing. Any words of wisdom?
- 1 reply
-
- xegs
- ac adapter
-
(and 2 more)
Tagged with:
-
hack CarnivHack: 6-digits-score Carnival hack (and testers wanted!)
rasty posted a topic in Atari 2600
Hello everyone! The premise: I've always liked Carnival on the 2600, it's a nice and fast brainless shooter but I find it way too easy to roll the score, considering that it's just 5 digits. To keep a not-so-long story even shorter, after a pretty long stretch of inactivity concerning all things Atari (years go by so fast!), I've decided to put the extra time available during the holiday break to some good use and add the much needed 6th digit to Carnival's score. Obviously the task at hand proved to be quite more time consuming than expected, but in the end I think I've done it. So I proudly submit for your judgment and possibly enjoyment: CarnivHack! Playing notes: Carnival is a game that requires a non-standard setting on emulators (i.e. Stella), since the display starts at row 26 and should be 214 rows high, instead of the default 34x210. Stella features a game-specific preset configuration for Carnival, so once you load this hack game in Stella please press TAB and open up Game Properties / Display, where you should set YStart to 26 and Height to 214, as mentioned above. You will only need to do it once. See image below: Some details and tech stuff: I had to completely replace the 6-digits display routine. Also, finding some free RAM was obviously a nightmare and forced me to implement a few workarounds. With some additional time on my hands I could have probably done things a bit more elegantly here and there. I wish I had the Stella Debugger a few years ago though I also had to remove the 2-players option and the logo.. not a big loss IMO! Interestingly enough, I've found out that the game actually used just 2 BCD bytes to display the 5-digits score, since the least significant digit is always zero TESTERS WANTED!!!: Before calling it a "final" version and spending time on tidying up the source for release, I welcome any kind volunteer to further test, focusing on the following: Testing on real hardware: my 2600 is PAL, so testing on real NTSC hardware would be awesome. I'm mostly concerned about the display being stable and the score displaying and vertically aligning properly, possibly comparing it to the original, non-hacked game Testing the scoring itself General gameplay testing, possibly comparing it to the original game And... testing!!! I won't have much time available in the coming months so hopefully nothing huge will show up.. hopefully! The fine print, i.e. known minor issues I (probably) intend to fix: - During the attract mode, when the duck starts eating your bullets the score will temporarily show garbled graphics - This game features an easter egg that shows the programmer's name if you keep the fire button pressed while powering up the Atari. Since I had to remove some graphics I'll have to do part with this hack as well since now it shows garbled graphics too. Once again, not a big loss: if you want to try the easter egg, use the original game CarnivHack - beta1.zip -
Operating from a Windows PC development environment, I would like to be able to test my 6502 Assembly code. So, for example: I might have this code: org $0600 CLC LDA #5 ADC #3 STA 203 <end of code> I would like to be able to execute a command from the PC which will check that the above code puts the value of '8' into memory location 203. It may run Altirra or a 6502 emulator in the background, but I need to be able to extract that memory location and then report it back to the calling program. At the end I want "Test passed" or "Test failed" or something similar displaying. This way I can build up an automated test suite for all of my procedures / sections of code / macros. I have no problem with it bringing up windows whilst testing is going on, but I want all of the windows closed at the end of the tests. Any ideas of the best way of implementing this? I guess I'd need to automatically dump the memory contents and then extract the memory values that I require...