tangobattle / tango Goto Github PK

@luckytyphlosion

• ensure opposite version gigas are not allowed
• ensure exe6 chips (hakushaku, gdsex, falzar, gregar, dblbeast, django, otenko) are not allowed on bn6 games
• navicust+kaizou legality checking
• check linknavis
• apparently you can set normal shot/chargeshot to something weird? need to make sure this is not set

asymmetric input delay should, in theory, allow the player with the higher input delay to experience less rollback. currently it does ????

a dropdown bar may also be a good place to show filter options, like being able to view only the replays from a specific game. Maybe you could also track whether a replay has already been rendered at least once, so users can view only unrendered videos which then get hidden from the list once rendering finishes.

Originally posted by @ssbmars in #9 (comment)

works on 3.2.33, almost definitely due to shadow input changes

todo:

• rewrite rx packets out of replayview to see what they are and compare

20220705144710-ROCKEXE4.5ROBR4Jbn45_us_real_bn_gameplay-v0.6.0-vs-I_Drink_Jet_Fuel-unhinged-aquadrgn-occurroico-round3-p1.tangoreplay.gz

Tango's launcher doesn't detect Switch's JoyCons, NES, and SNES controllers, ignoring any button requests when attempting to map these controllers.

Strangely, Switch Pro Controllers work perfectly fine, not sure how the above controllers operate differently from SPCs.

When opening Tango, automatically select and display the save file, associated game, and patch that was being used the last time Tango was closed. This may be best as an option that can be toggled on and off in settings.

For some reason, whenever a version change is made(3.2.44 to 3.2.45, for example) the program won't auto-update for me. I have stable update channel on and I don't know what I'd have to enable to allow it to auto update.

get rid of the refresh button by watching for fs changes

There's always an external program that will help out with that, but not everyone is a tech literate person willing to take a minute to set up a keyboard emulation program. With Tango 3.0 focusing on user-friendliness (or so it seems to me), I feel like adding controller support fits in with that theme.

wgpu is a little too cutting edge i think (vulkan crashes on intel igpus), glium should support everyone

return better messages to the launcher instead of just bailing out with anyhow::bail

i forgor to add this 💀

maybe via a dropdown at the top of the replays tab?

i dont even know where to begin with this one i hate windows

currently tango will only do lightweight: there is some support for selecting this (matchSubtype) but this is not surfaced through the UI

we can do this during a match somehow but not while in the lobby?

this works fine most of the time but is strictly incorrect on misprediction

this replay is complete but gets wedged at the end

20220725163331-MEGAMAN_EXE2AE2Ebn2_simpcn-v1.0.0-vs--12345-round1-p2.tangoreplay.gz

the electron stuff has to go first, though

i want to get rid of electron because it kind of sucks

unclear why this happens: is rng2 generated twice by accident? however, rng1 is still in sync? weird!

[2022-07-07 00:01:20.936] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow rng1 state: aeee908b
[2022-07-07 00:01:20.937] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow rng2 state: 3e755a8a
[2022-07-07 00:01:20.939] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow state committed on 0
[2022-07-07 00:01:20.946] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] primary rng1 state: 333e77a4
[2022-07-07 00:01:20.947] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] primary rng2 state: 3e755a8a
[2022-07-07 00:01:20.949] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] battle state committed on 0

[2022-07-06 23:01:20.644] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow rng1 state: 333e77a4
[2022-07-06 23:01:20.644] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow rng2 state: 5ca4f9b7
[2022-07-06 23:01:20.644] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] shadow state committed on 0
[2022-07-06 23:01:20.649] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] primary rng1 state: aeee908b
[2022-07-06 23:01:20.650] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] primary rng2 state: 5ca4f9b7
[2022-07-06 23:01:20.650] [info]  core: [2022-07-07T04:01:20Z INFO  tango_core::hooks::exe45] battle state committed on 0

https://drive.google.com/file/d/1rXSUaLT-UHAmbBoSjQIr7G-i9VosMhM4/view

things to do:

• navicust support
  • get navicust cata
  • find where navicust data is stored in EWRAM
  • expose navicust info via getNavicustEditor
• modcard support
  • get modcard cata
  • find where modcard data is stored in EWRAM
  • make bn4 modcards UI

i'm not sure why:

• this only happens during the turn (also the audio gets scrunkly)
• in the custom screen the audio is not scrunkly
• doesn't seem to cause any timing issues
• maybe skipping samples?

Known issues:

• Supports Blue Moon US version only.
• You can only play on Final Destination.
• You can only play a single round (will crash when going into round 2).
• Do not use saves that, when being loaded, open the New Game+ menu.
• Reveal setup doesn't work.

sdl manages to detect this but is unable to handle events from it

things to do:

• folder support
  • find where the folder info is stored in EWRAM
  • add chip data and icons
  • expose chip info via getFolderEditor
• navicust support
  • get navicust cata
  • find where navicust data is stored in EWRAM
  • expose navicust info via getNavicustEditor
• style support
  • find where style data is stored in EWRAM
  • make a UI for this

this seems relatively complicated to implement, as during the random battle screens there is SIO communication going on. i don't really know what the SIO in question is, though, and how much of it is actually required (and how much of it can just be fudged).

things to do:

• folder support
  • find where the folder info is stored in EWRAM
  • add chip data and icons
  • expose chip info via getFolderEditor
• navicust support
  • get navicust cata
  • find where navicust data is stored in EWRAM
  • expose navicust info via getNavicustEditor
• modcard support
  • get modcard cata
  • find where modcard data is stored in EWRAM
  • expose modcards info via getModcardsEditor

when winit was used it apparently just didn't care about this!

marwe — Today at 16:34
nice, I think that's pretty much the only way to make the dynamic dream a reality
otherwise the other option would have to be like the patch devs supplying additional hardcoded metadata for the save viewer