Jump to content

TailChao

Members
  • Posts

    958
  • Joined

  • Last visited

  • Days Won

    2

TailChao last won the day on February 18 2019

TailChao had the most liked content!

4 Followers

Profile Information

  • Custom Status
    Bup?
  • Gender
    Male
  • Interests
    Being a busy person.

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

TailChao's Achievements

Dragonstomper

Dragonstomper (6/9)

1.5k

Reputation

  1. Woah, now that was totally unexpected! BupSystem definitely wasn't licensed from me, so I'm guessing these are all updates to the ProSystem core. Fully supporting BupBoop / BupChip music will be difficult if there's no support for audio mixing from the cartridge slot, but the data is small enough that it could be included with the 2600+'s firmware. It took five years to get here. So maybe in 2028 someone else will use it in a shipped game, lol.
  2. The YM2413 is also a write only device, so placing it alongside your paging registers above $8000 is a good option. This would avoid any conflicts in the $04xx or $08xx areas, or at least that's what I was planning on doing when it was still a candidate.
  3. 768KB if we include the BupChip's software and data. Very little, unfortunately. Creating new versions of each track is a big ask, and there's only 2KB or so of fragmented space available in the game's binary. It might be possible to add support for streaming ADPCM data to the MiSTer though, similar to how the 7800 GameDrive handles BupChip support. But I don't have free time for any of this right now. That said, RushJet1 did a few TIA covers of the game's music if you're curious about how it'd actually sound .
  4. This is where most of the support for the game seemed to originate from, rather than AtariAge's typical demographic. There's also a Windows version available on Steam or through our Humble Widget. It includes the game's ROM and music data which can be played on the MiSTer, LibRetro / RetroArch, and 7800 GameDrive. You can find out more in the FAQ. Right now the game is on sale for $1.99 on Steam, a little cheaper than buying a $129.99 device which can't run it anyway .
  5. R34 and R35 are both series resistors on the trigger inputs (J3-6 and J4-6 on the NTSC schematic, which go to the TIA's I4 and I5). So no, this wouldn't help with the paddle inputs drifting. If you wanted to build a solution into the console, it'd resemble a mirror of the Q5, Q6, and Q8 transistor network which strongly pulls up the trigger inputs when ProLine mode is enabled through PB2 or PB4 on the RIOT. But instead of pulling up the trigger inputs when PB2 or PB4 and TIAEN are low, it'd pull down the paddles. Something like one NPN and a pulldown on each paddle line, fed by !(PB2 | TIAEN) or !(PB4 | TIAEN) through a few logic gates. I'd honestly leave the wedge alone though. It's easier to take a female DE9 then short pins 5 and 9 (paddles) to pin 8 (ground) - now you've got a terminator. You can even 3D print a little shell for the solder side to make it look like a fancy port cover. Isn't ProSystem licensed under the GPL? You should just be able to ask Atari / PLAION to release the source for compliance if everyone is so eager to fix their product. Or even better...
  6. R34 and R35 are series resistors for the CX-40's trigger on Pin 6. Removing them will, of course, disable fire button functionality as you're also removing the connection from the CX-40's trigger to the TIA. However, they're also used by the ProLine to aggressively drive the paddle inputs high using the transistor network of Q5, Q6, and Q8. Not exactly. Pin 6 on either joystick port, which is used for the CX-40's trigger, is pulled high by R40 or R41. In single button mode (or for any 2600 game), this prevents the pin's state from changing and spuriously detecting a press. So in this case, it's difficult to get any surprises unless something has gone catastrophically wrong. The paddle inputs on Pins 5 and 9, which are used by the ProLine controller for both of its triggers, have nothing to keep the pin's state from wandering. The two resistors inside the ProLine pull these low to indicate both buttons aren't being pressed. Pressing either button will short Pin 6 to its respective paddle input and force it high, using the aforementioned transistor network. The ghost button presses in Rikki & Vikki, Xenophobe, and Rampage are from the paddle inputs wandering high without a ProLine attached to hold them low. They're effectively an antenna. If the ProLine is being used in single button mode (Q5 or Q6 are off), its pull downs are strong enough to overpower R40 or R41 and register a press on Pin 6. This is how its CX-40 compatibility works. Refdes above are all for the NTSC schematic, of course.
  7. There's a lot of accuracy improvements I wanted to add in BupSystem, the input block was definitely one of them. But for now its development is still suspended, I've gotten more support for reverse engineering the Watara SuperVision and have to give that priority. Right, and in that respect I think omitting support for Rikki & Vikki is totally fair. Most of the interest in the game seemed to come from younger players anyway, who wouldn't be likely to purchase the 2600+. However, I do think the 7800 emulation in the 2600+, as it is, needs significant improvement to live up to the marketing copy. For now, let's say "probably not". But you can still purchase the game digitally and run it on the 7800 GameDrive.
  8. That was my understanding as well. These are just current limiting resistors, so while you're seeing the behavior go away it's likely more of a happy accident. Same as when swapping the TIA out for another one seems to fix it. Not that I'm doubting it's worked for you - but electrically you're not doing much other than changing the behavior of an antenna without attaching a ProLine or grounding both paddle inputs. Mostly I want to dissuade from viewing this as a "bug in the TIA" since it's fair game for the paddle inputs to operate this way, based upon the TIA's internal schematics and how it was used in the 2600. If anything the "bug" would be in the 7800's original documentation for omitting it, or perhaps not realizing it'd be a problem. Ah, but speaking of terminology misuses... ...should be ADC. May bad.
  9. Hey, Rikki & Vikki's lead here... The physical cartridges have been out of production since late 2020, but still I have a few of our prototype boards and could send one over. For digital goods or other inquiries, send me a private message. Not really, as I mentioned earlier in this thread. The game is already supported in Libretro / RetroArch, which is based upon ProSystem. It's also supported by BupSystem, which I wrote with low-end platforms in mind and is available for commercial licensing.
  10. I'd argue the BupChip is even easier to support than ARM / Melody, as are most of the 7800 audio expansions. There's a dedicated pin for analog audio input right on the cartridge slot. Put a DAC on there, forward the emulated software's writes to the actual chip, capture the output and mix. Aside from fiddling with some timing here and there, it's a much less stressful job than accurately reimplementing a gajillion different audio chips. While your workaround of adding pulldown resistors is correct, I wouldn't call it a bug in the TIA - rather, it's a quirk of how the paddle inputs are designed. You can also mitigate this in software by strategically dumping the inputs to ground. I didn't think any precautions were necessary at the time as none of the other "press button to play" games like Xenophobe or Rampage took them. Only after launch did the quirk show up in any sort of scale 😕 . So don't blame the TIA, but rather my expectation that the ProLine's curse would be limited to some minor wrist damage.
  11. I dunno, if 1.3e is still before any of Maria's fetch penalties were emulated then the performance will be amazing.
  12. Just a heads up about checking for opposite directions pressed on the controller (i.e. UP + DOWN or LEFT + RIGHT) - some of those other console manufacturers recommended one should avoid this practice in their development literature. It's not just the CX78 that can yield spurious presses this way, but many controllers using a rocking directional pad. Especially after three decades of use. Of course, if supporting the Flashback controller is a value add then you'll have to leave it in (or make it optional). But generally this sort of behavior will become more common on NES, SNES, SMS, and certain types of Genesis controllers as they all age.
  13. If you're going with ProSystem 1.3, I recommend checking out of the forks by Tachi and what's currently used in Libretro. These have improved compatibility, and in particular support Rikki & Vikki's custom hardware. BupSystem, the emulator used for Rikki & Vikki's development and distribution, is also available for commercial licensing. This wouldn't be free, but the cost would go towards new features and improving compatibility. As of writing its accuracy is higher than ProSystem, but it'd take a little effort to catch up with a7800. It also runs fairly quick and is a good fit for low end targets.
  14. I've updated the first post's FAQ to cover a few cool new things, such as support for Rikki & Vikki's custom hardware on the 7800 GameDrive (thanks to SainT) and ProSystem-libretro (thanks to Tachi). The latter even supports Retro Achievements, neat! That would be nice, but it's also largely up to Atari to ensure the 2600+ supports the game. If this sort of thing is important to you (or anyone else reading), please let Atari know you'd like to play Rikki & Vikki on the 2600+ and/or purchase the game from them.
  15. Safety first! I've submitted my build of BupBoopConv to Microsoft for analysis, looks like it's clean. @Trebor, if you keep getting the Wacatac false positive send me a copy of your machine's logs as per the directions above (in the analyst comments section). Well, that was exciting. Time to go hide in my secret moon base for another year or two.
×
×
  • Create New...