(Link to video)
Submission Text Full Submission Page
This TAS aims to reach the crash that happens in NES Tetris. To best experience it, you will have to patch an existing rom dump to increase the digits on the score display from 6 to 7, which can be found here: https://drive.google.com/drive/folders/1VYmem92eAhcOhl_sLAOkIuFlpeqANx2I. Make sure to select "7digitscore.ips" as that was the patch used. I also used game genie codes (KUYPZS and EXYPLI) to uncap the score past level 29, up to 49, after which the level display breaks once more. You can find a table to convert the level numbers to something human readable here: https://cdn.discordapp.com/attachments/987042025624981534/1001200368086171709/Level_Conversion_Table.jpg
A sequel TAS is currently in the works.

feos: Claiming for judging.
feos: As others have mentioned in the thread, this run is far from optimal. It plays through levels 19-29 in 07:17.37 which is only 10 seconds faster than this AI that doesn't use Pause to manipulate a better piece. This run that appears to be a real TAS completes those levels in 2 minutes 46 seconds. This TAS demonstration from 2014 completes those levels in 02:53.98. The main problems seem to be that this submission doesn't do tetrises all the time (clearing 4 lines at once), and it doesn't play at the top to clear lines faster.
The goal is also strange. Tetris is a game that will show you an ending after you die, and this movie does get an ending after reaching the kill screen at the end, where the game pauses for a moment and then pieces start falling again and you can't do anything. But this movie uses a patch that does some tweaks to the score counter, and that decision is questionable.
We do accept ROM hacks if they are of decent quality and not overly obscure. The ROM patch used here is indeed obscure, because I could only find it in one place on google, and there was little to no explanation on what it actually changes. Here's a full list of bytes it's changing.
Why is this so important, one may ask? It's important, because that patch avoids the crash that happens on the original ROM with this exact submission, half way through (frame 126784). In a "fastest crash" category.
We have accepted "fastest crash" and "fastest softlock" categories to Playground, but it's not an acceptable goal for Standard (because it's not a standard goal in its nature) or Moons (because it doesn't entertain the audience all that much) movie classes. So Playground would be an appropriate place for this goal if the submission was better optimized. But not when the game is tweaked to contradict the very goal of the movie. If you aim for the fastest crash, you shouldn't patch the game to make it happen later.
This post mentions that FCEUX is just not good enough to emulate this kind of crashes accurately, but this run desyncs on BizHawk early on. I haven't tried running it on Mesen, which is also very accurate, but for a potential future attempt BizHawk is a must, ideally its NESHawk core.
adelikat asked which goal could be more suitable for this game, but it's hard to discuss the potential if the original game has to be hacked to continue working after a certain point. And if we don't patch it, the only goal I can see as technically interesting is playing until the difficulty resets. But that would still not be Moons content, and if we're limited with Playground, we can have quite some variety of arbitrary goals, as long as they are worthwhile in their nature and as long as they are actually met.
Rejecting this attempt though. Better luck next time!

TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 14777
Location: 127.0.0.1
This topic is for the purpose of discussing #7624: m_riss's NES Tetris "Fastest Crash" in 1:10:26.37
adelikat
He/Him
Emulator Coder, Expert player, Site Developer, Site Owner (3581)
Joined: 11/3/2004
Posts: 4736
Location: Tennessee
The crash isn't guaranteed to happen at a particular level or line count, correct? Would it theoretically be possible to avoid it crashing until a level wrap? If so, that would be a far more interesting goal.
It's hard to look this good. My TAS projects
Judge, Skilled player (1275)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Does this movie also sync on the unpatched ROM?
my personal page - my YouTube channel - my GitHub - my Discord: thunderaxe31 <Masterjun> if you look at the "NES" in a weird angle, it actually clearly says "GBA"
Bigbass
He/Him
Moderator
Joined: 2/2/2021
Posts: 153
Location: Midwest
Is this submission different from the movie you had me test on hardware a few weeks ago? Is there any explanation for what's happening? What causes the crash? How did you make the movie?
TAS Verifications | Mastodon | Github | Discord: @bigbass
Player (12)
Joined: 6/17/2006
Posts: 498
Last time I checked, a crash doesn't allow gameplay to resume to a top out like we can see in your video... Gameplay also appears to be sub-optimal. I find it odd that the legal screen appears for several seconds longer than other submissions, for instance. You could also be playing closer to the top of the playfield like the other published NES Tetris movies to clear lines faster. Also, this other video showcases what appears to be a genuine crash in less lines and without pausing, which suggests better overall optimization: https://www.youtube.com/watch?v=l_KY_EwZEVA I'm all for an NES Tetris TAS that pushes the game to its limits. but I don't think crashing the game should be the goal in itself. Best theoretical score/lines/level before an inevitable crash would be much more interesting in my opinion.
jmosx36
He/Him
Player (122)
Joined: 12/7/2013
Posts: 98
Location: Colombia
I also think it is a bit suboptimal, based on this https://www.youtube.com/watch?v=fy5qt_SOapM
Currently making a route for Super Bomberman 5 200% TASes i'm planning: aero fighters, tg3000 (again), gradius 3, bust a move, bust a move plus, gradius rebirth, smg, smg2, mp9, gh3 (wii)
Post subject: Unpatched ROM Sync
m_riss
She/They
Joined: 7/31/2022
Posts: 2
Location: America/Delaware
ThunderAxe31 wrote:
Does this movie also sync on the unpatched ROM?
This will sync on an unpatched ROM. However, there is a chance that it will crash earlier than intended on original hardware.
adelikat
He/Him
Emulator Coder, Expert player, Site Developer, Site Owner (3581)
Joined: 11/3/2004
Posts: 4736
Location: Tennessee
m_riss wrote:
ThunderAxe31 wrote:
Does this movie also sync on the unpatched ROM?
This will sync on an unpatched ROM. However, there is a chance that it will crash earlier than intended on original hardware.
A chance? Shouldn't these inputs be deterministic? Also, regarding optimization, I think the more important thing to establish is goal choice. At level 256, the levels actually wrap back to 0 and it goes back to level 0 speed. Avoiding this potential crash to get to that point I think is inherently interesting, especially since no human or bot has done this (yet), so it shows off something unique to TAS. I'd rather see that goal choice than this one, which is bot achievable and theoretically human achievable (humans are a long way away from these levels, but it is a matter of endurance and luck than it is skill at this point). I'd like to establish what the reasonable goal choice here, rather than simply rejecting for optimization.
It's hard to look this good. My TAS projects
GJTASer2018
He/Him
Joined: 1/24/2018
Posts: 242
Location: Stafford, NY
adelikat wrote:
Avoiding this potential crash to get to that point I think is inherently interesting, especially since no human or bot has done this (yet), so it shows off something unique to TAS.
Actually, that's not true: Link to video
c-square wrote:
Yes, standard runs are needed and very appreciated here too
Dylon Stejakoski wrote:
Me and the boys starting over our games of choice for the infinityieth time in a row because of just-found optimizations
^ Why I don't have any submissions despite being on the forums for years now...
adelikat
He/Him
Emulator Coder, Expert player, Site Developer, Site Owner (3581)
Joined: 11/3/2004
Posts: 4736
Location: Tennessee
GJTASer2018 wrote:
adelikat wrote:
Avoiding this potential crash to get to that point I think is inherently interesting, especially since no human or bot has done this (yet), so it shows off something unique to TAS.
Actually, that's not true: Link to video
What exactly isn't true? This is a TAS and proves my point that a level roll-over is a better goal for a TAS, than the goal chosen in this submission.
It's hard to look this good. My TAS projects
Player (16)
Joined: 7/3/2012
Posts: 33
GJTASer2018 wrote:
adelikat wrote:
Avoiding this potential crash to get to that point I think is inherently interesting, especially since no human or bot has done this (yet), so it shows off something unique to TAS.
Actually, that's not true: Link to video
That was done using a Lua script to avoid the crash, so it doesn't count. Since the crash is related to obtaining high scores in a single line clear, the easiest way to avoid the crash is to stop scoring Tetrises after a certain point. I think it'd be more interesting to investigate if it's possible to route around it and still score Tetrises, though. (In fact, this submission continues to score Tetrises until 6 levels after the video with the AI crashed, proving that it is indeed possible to some extent.)
negative_seven
She/Her
Active player (354)
Joined: 11/25/2012
Posts: 101
Location: Europe
Arcorann wrote:
GJTASer2018 wrote:
adelikat wrote:
Avoiding this potential crash to get to that point I think is inherently interesting, especially since no human or bot has done this (yet), so it shows off something unique to TAS.
Actually, that's not true: Link to video
That was done using a Lua script to avoid the crash, so it doesn't count. Since the crash is related to obtaining high scores in a single line clear, the easiest way to avoid the crash is to stop scoring Tetrises after a certain point. I think it'd be more interesting to investigate if it's possible to route around it and still score Tetrises, though. (In fact, this submission continues to score Tetrises until 6 levels after the video with the AI crashed, proving that it is indeed possible to some extent.)
Crashing actually has to do with an interrupt happening right in the middle of some code that handles jump tables; it's just that this doesn't happen until quite late into the game, when the score addition logic starts taking a significant amount of time. This is a delicate enough occurrance that just making small changes like delaying the line clear or toggling the next piece visibility can be enough to avoid the crash. There is a legitimate movie that gets the level counter to wrap around and even scores a tetris almost every time. Well, somewhat legitimate - I'm pretty sure FCEUX is not accurate enough to emulate this rather timing-sensitive behavior correctly, and that's backed up by the fact that the movie does run into a crash on Mesen.
TASVideosGrue
They/Them
Joined: 10/1/2008
Posts: 2728
Location: The dark corners of the TASVideos server
om, nom, nom... juicy!