Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
For me, console verification is the most impressive part of TASing. So I thought I would make this thread to see what games (with or without existing TASes) people would most want to see a console verification of.
Personally, I'm mostly interested in seeing new work on N64. This will probably need a new emulator, but I'm hopeful it might happen one day, and maybe there are still some games out there like SM64 and Mario Kart where verification is possible with Mupen. Anyway here are the games I most want to see console verified.
N64:
Goldeneye
Perfect Dark
Rogue Squadron
Banjo Kazooie
Ocarina of Time
Diddy Kong Racing
Body Harvest
1080 Snowboarding
SNES:
Mega Man X
Super Mario RPG
Donkey Kong Country 1/2
NES:
Punch Out Done!
Ninja Turtles
GB/C:
Oracle of Ages/Seasons Done!
Shantae
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
There's been a lot of interesting stuff happening in N64 TASing recently, maybe some of those creators can give some commentary? Iknow WhiteTed has some cool Goldeneye stuff, and the recent SM64 120 star WIP post would be cool to have some commentary with it too. Also new improvements in Mario Kart 64.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Yeah the file I uploaded accidentally cut off the last couple of inputs. The console desync though happens around frame 12300.
I think we need a sprites + window mode 3 test. The boss part of the screen is all window and I don't think we have a lot of details about timing in that situation. The game is checking for end of mode 3 each scanline. Not entirely sure this is it but at least need to rule it out.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
RetroEdit wrote:
Alyosha wrote:
Here is a file for gambatte (2.4.2) that desyncs on console on the second boss in a deterministic way.
Does that file sync in 2.5? Gambatte was updated in 2.5.
Not sure, I haven't gotten the chance to look at the new GBA BIOS stuff.
EDIT: I hacked back in the fake GBA bios stuff and can confirm the run syncs on the current build of gambatte.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Gensan 2 continues to provide challenges for console verification.
http://tasvideos.org/userfiles/info/66301728356726713
Here is a file for gambatte (2.4.2) that desyncs on console on the second boss in a deterministic way. I have confirmed a version in GBHawk matches this run cycle for cycle in the boss area. GBHawk has an IRQ priority issue that is fixed in my local build that made it desync in a different way, but once I fixed that it syncs to the end of the game just like Gambatte.
So there is some un-implemented behaviour in both GBHawk and Gambatte that is causing desync on actual hardware.
I didn't see any uninitialized RAM so I don't think that is the cause. Also in this run mode 3 timing does not seem to be the issue (at least to the extent testable.)
Currently I don't have any idea why this desyncs.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Nice run, yes vote. It's good to see expert knowledge put into a TAS like this. A lot times I go into games like this with only superficial knowledge and inevitably miss things that I would only ever really know from time and experience. This TAS really shows the difference those things can make.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
NES games can also have issues with uninitialized RAM, it's not exceptionally rare.
It is good to identify that problem though and ways to work around it for games that encounter it. Just having things written down in an organized way really helps.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
EZGames69 wrote:
What do you mean by this?
I mean LibTAS and the examples of games with built in TAS tools. Maybe not console verification in the usual sense but still a lot of progress compared to say genesis.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Making and re-syncing TASes can take a lot of time and effort, are there any thoughts on which specific games or TASes should be looked at that can most help console verification?
GB and GBC have progressed a lot recently, and PC of course too, but not so much for other consoles, where should effort be focused?
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I thought I would make a thread specifically for runs that failed to sync on console. There is a page with a lot of old tests here but a lot of those are too old to be relevant to modern development.
Failed cases tend to find hardware edge cases not expected or covered by test ROMs, so keeping track of them seems like a helpful way to drive improvements to emulation and verification. More generally keeping track of what things can go wrong and why helps get a clearer picture of what path development needs to take.
I expect a lot of older TASes to fail to sync just by virtue of being made on old inaccurate emulators, so I intend this thread mainly for new TASes made on accuracy focused emulators.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
ThunderAxe31 wrote:
I just found another graphical bug with GBHawk and SubGBHawk. The first 11 frames of the GBC BIOS sequence consist in a black screen. I can confirm this behavior is NOT present on any real console: GBC, GBA, GBI.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I find it fascinating how there was a hint of this all the way back in 2007 but it took over a decade of technological development to be able to arrive at the point where this TAS could be made.
Jumping to the middle of the credits seems a bit strange, but if there isn't any way to tell the game is beaten besides seeing the end screen then good enough I guess, yes vote.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
GBHawk still applies inputs on VBlank boundaries, so you cannot save time simply by converting between it and gambatte, unless the game samples inputs while the screen is off, which this game doesn't do.
My guess is you just found some places where I didn't press input on exactly the right frame, like in the 2nd K .Rool fight where I am indeed 3 frames late.
However trying to save after truncating my movie file when fixing this also corrupted my TASproj, seems like something bad is going on with TAStudio.
EDIT: yeah i missed a few presses, here is a corrected file.
http://tasvideos.org/userfiles/info/65880395759689215
I'll probably take another look and see if I missed any other small things.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I (mostly) fixed 4 in a Row. This was actually a very difficult case. Just figuring out what the ROM format was took a lot more googling I expected (it's something called 'XROM' apparently, and only in 2 games.) After accounting for that there were several small bugs that needed fixing that slid under the radar until now.
There is still a problem that a counter interrupt is happening one scanline too late, but that kind of thing could take forever to properly sort out and I currently don't have the time. So for now it will have to be good enough, at least it's playable.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
Thanks for the correction, not sure what happened there.
Also I realized that I only saved frames in one spot, not two. Where I originally thought I saved frames in mashing was just a re-sync error, the only savings are in the cutscene skip.
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
ThunderAxe31 wrote:
Why do the tilt controls need to be explicitly set? It's confusing, user-unfriendly. It's not like you can choose to not use them when you play on real hardware, as it's part of the cartridge itself, or decide to use it for other games.
Eventually I want to make different motion models for the control scheme. Right now it is Tilt (realistic.) Eventually I will add Tilt (simplified) and maybe some others that allow linear accelerations (speedrunners seem to use quick linear jerk motions to make Kirby jump, the accelerometers work just as well sensing any kind of acceleration.)
jlun2 wrote:
Thanks. Changing to tilt mode however disabled my dpad, which I need to select stuff in menus. Is there a way to enable both?
Also do I need to change anything else? I can't seem to move still in game.
The tilt controls don't have default settings. Go to config -> controllers to set them.
EDIT: Maybe I can make the tilt scheme a different sync setting, I'll think about how to do it.
EDIT2: Ok controller setting now only effects MBC7 games
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
jlun2 wrote:
Not sure if this is the right place to ask, but how does one use motion control for games like Kirby's Tilt n Tumble? I can't seem to find an option for it.
GB -> settings -> sync settings -> controller -> tilt
EDIT: Also I went through Gambatte's Halt tests and fixed a couple more edge cases. Passes all but the double speed mode tests now (which rely on mode 3 timing anyway.)
Editor, Experienced Forum User, Published Author, Expert player
(3602)
Joined: 11/30/2014
Posts: 2749
Location: US
I have made various improvements to GBHawk's timer emulation. It now passes all of Gambatte's timer tests in both GB and GBC mode, while still syncing console verified runs and passing the other available timer tests.
This doesn't fix the problem in Gensan 2, but it does narrow it down to being a purely GBA issue. Still not sure exactly how I will handle that one yet.
I also made a test movie for Pinball Fantasies, and that run worked on console, so that was good.
EDIT: I implemented a basic version of the GBA glitch. Basically if you are enabling the timer while changing frequency and the old frequency is different from the new frequency, and the old div position was a high bit and the new one is a low bit, then an extra increment of timer happens. This doesn't capture all of the details of the glitch, but gets all the basic cases, including Gensan2. The glitch is rare enough that it doesn't impact existing runs, but the details are strange enough that it's something to keep an eye on.
