Also FWIW I would like to reiterate that I believe I save some frames outside of the difficulty change itself as well, so all that aside there is that as well. I could technically slap the difficulty back to how it was and still be ahead, so at best there's a little bit of fine tuning.
I was sorry to hear about your retirement and hope you make a return sometime.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
No ill will and no offense taken. I'm a fan of the better of the 2 runs being published, whichever that is deemed to be. I'm less concerned about my own TAS and more concerned about showing the best TAS. No different than RTA running, any new WR is great for the game.
Amen. Thanks for the cool attitude. There's always the option for two categories, but that's probably not worth it.
It's just my opinion that somebody watching on YouTube would put more weight on the final number than the TAS number. It's also how the RTA runs are timed. And obviously 20 minutes is about as big as it gets. I'd hate to see somebody mistaken that you *have* to wait the full 40 minutes as they aren't going to be aware of the difficulty settings.
It really all comes down to how heavily we can abuse the TAS timing, which in both scenarios is "a fuckton" and "a double fuckton". Obviously in any other scenario RTA timing would be implied and the difference would be apparent and there would be no question. Here, it's a bit blurry.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
FWIW I'm fine with any decision that gets made.
You're correct in that it was a happy little accident that I managed the "best ending" and I wasn't even trying to manipulate such. I used the hardest difficulty to be cheeky (a trait I've displayed quite a bit in the past two months or so) knowing full well about the longer time run out. Hardest difficulty isn't a requirement anyway.
I'm on board with whatever you and the judges want to do. I wasn't expecting things to play out like this.
Hey there. Hopefully you don't take offense to me submitting, I wouldn't have done so if I hadn't spent hours on the game months prior (and glitch hunting dating back to 2011) and only didn't submit because I figured it was a bad game choice. Obviously you could imagine my emotions when I realized this was not the case, hah! I had a finished run sitting in a drawer for months.
No ill will or anything. It's just a TAS. Seems there is way too much energy in this thread than there needs to be. I mean I don't even know you.
I made a comment on Discord that the grue bot was spamming the forum and had a screenshot of some of your cancelled/rejected submissions (most of which were cancelled, not sure how that would be offensive as there's no shame in cancelling), I was trying to make a joke about the bot itself, and Spikestuff went off on me, which probably doesn't look good on this turn of events either. Probably explains their reply. It's really not that complicated and I have no ill will for anybody.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
I'll look into it. Tried some basic RNG manip (start delay, movement on sled, movement in house) and always got the same result.
Also combed through all the speedruns on speedrun.com (including ShesHardcore's 3rd place run which followed the same standards of my TAS including difficulty and AI house progression), and on easy mode like all of the rest,
and no runner was able to perform with these circumstances. All had the scenario play out as in my TAS. There was some that triggered The Old House instead of The Mansion but it doesn't seem feasible to get a different result than those without a difficulty swap. I'd say if they were ending up with one of the different 5 houses there would be some RNG going on, but the layout was like 90% Mansion, 10% The Old House, 0% the others. Unless there is some serious RNG delay going on (to the point of it being of questionable worth), I'm pointing to the difficulty of altering the order the AI gets to the houses.
I'll have to look into if the difficulty settings changes the order of the houses. Since it wasn't mentioned in the submittal text, I figure it was likely a coincidence as there's no obvious RNG manipulation present in their file (I checked this morning), rather than an intended outcome. The movement and inputs in the entire run are clearly done with the intention of the fastest result. If there needed to be some RNG manipulation, some up and down spam on the sled is a good way without losing any frames.
I'll keep poking at the RNG but if it's difficulty based I would certainly prioritize the run ending twenty minutes earlier.
I think losing a few frames AND having the run last 40 minutes instead of 20 minutes is a dealbreaker for best ending if it comes down to not being able to get the best result with RNG manipulation (sans difficulty), especially because it's the standard for the speedrunning community. I definitely think a movie last 20 minutes to 40 minutes would be more ideal for the YouTube channel, and all in all the ending is just text on a screen.
I've seen runs with better "best endings" settle for the "quickest ending" and this run has the same endings besides just text on a screen.
Either way, I'll keep looking into it but will await a judge. (Hopefully one that isn't friends with the other author for fairness)
Hoping I'll crack open some RNG door but I'm pretty sure it's just difficulty based.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Just wanted to say I tried this and it works great
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
As we approach the 10 year anniversary of this thread, #ReleaseTheInputs
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Oh and I would like to welcome everybody to share in the cringe of my roots
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Dug this out of an old hard drive. It should be noted it was in .tasproj form, so the date on the bk2 is current as I had to convert. Not that it matters.
I'd also like to wish ShesChardcore a happy retirement after an incredible tenure of TASing.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Submitted mine. It ended up being 9 frames faster due to several reasons.
Incredibly I found the hard drive, that was the hard part.
https://tasvideos.org/8058S
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
I have a run of this on my old hard drive I spent a really really long time optimizing the sledding, I can hook up my old hard drive and check it and see who's faster. I remember frame perfect drifting past the snowman's hitbox so I think mine is but maybe I'll bite my tongue.
Never submitted it because I didn't think the game would be accepted because you have to wait 20 (or 40) minutes to finish.
You're adding 20 minutes to the finish by playing on hard. Easy is 20 minutes, Hard is 40. Only difference (in the whole game let alone this run) is the timer so it will pass judgement with flying colors and save frames.
Also, you can land frame perfect on the clip spot. I know it doesn't really matter, but if you were going for a cleaner look it's possible.
EZGames69 wrote:
Since movement can screw with enemy AI, doing even the slightest change can make the whole run impossible.
The robot is hardcoded and as long as you pass the pipe you're safe. There is no fine tuning this run. Only way to lose is basically screwing up the sledding.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Run looked pretty nice! Is there any reason to remap the controls at the start?
No. Personal preference. Then I realized the TAS timing rules. Unfortunately the bug mentioned above prevents changes.
I plan to redo it at some point but until then I think it would be a real shame to have the world record sitting so much higher.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Unlike dwangoAC, I'm an outsider to the TASVideos community and I don't have a vested interest in whether the project is accepted here or not.
This is the way.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Is data input through control inputs seriously a copyright concern? It's an unmodified ROM, right, so the payload is input through controls.
A) Nintendo would likely never figure this out let alone act on it
However, okay, we'll run with that:
B) It doesn't run on emulator anyways so 99% of people wouldn't download the movie file
C) Let a site admin run the movie on a different TASBot and call it a day? Verified, but not publicly downloadable. This is a playaround anyhow with massive time losses doing other activities, so it's not like it's a race anyhow.
Man, just found this - and what a discussion.
I'm moreso on team "if there ain't a file, it's a denial".
I also think it is worth noting not everything needs to be accepted to TASVideos. It's already been seen by a bazillion people... is there really a benefit to having it on the site, let alone performing massive site changes to accomodate one movie out of thousands? Additionally, will those changes potentionally compromise future standards of REAL runs?
Honestly, deny it and move on. It's a cool piece and it's on YouTube. Good enough.
Also, with the other performances, why not just dump the input properly to a file? Is it because it's running on console? Then run on emulator or don't submit. Doesn't seem that hard.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
very impressive! do you plan on making a tas of the “raptor” branch as well?
Thank you! I am close to developing a cam hack as most of the run is off screen, might give it a go then. Most of the enemies are bypassed as they are off-screen so figure I will use lua to camhack and despawn all enemies (or most) and see how it starts to sync with the camhack and everything disabled. Grant is my favorite category though. :)
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
I'm playing Jurassic Park on Genesis and occasionally I get a lag frame that spawns on frame #2993.
I had this several times throughout the creation despite zero changes at nor before the frame which should be impossible.
The final time it fixed itself lasted through several days, several reboots of BizHawk and TASEditor, ROM loads, LUA, etc., with no problems, up until publication which lasted without a hitch and synced on Samsara's PC.
I published a world record TAS succesfully through a bk2, but now going back to edit the file (skipping the splash screen in the startup process and re-aligned inputs to match, which does not alter lag nor RNG in memory otherwise the run would fail almost immediately and this is the most finicky sync game ever) triggered the glitch again.
Normally I would write this up to the change, but it is the exact same change that plagued me through zero-input-change scenarios several times, so I have a feeling it is related, and if skipping the splash screen changed a single time, there is no way the run would make it precisely exactly the same to frame #2993 without desyncing in a major manner.
Every single input, frame change, etc., in this game will desync the run so I have no doubt any side effect would not make it past the first few seconds let alone 2,993 frames later.
Same exact frame I was having problems on prior.
Love the emulator otherwise and the TAS Editor is superb. Thanks guys.
Will attach files below.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
I have encountered a bug in BizHawk(?) that randomly causes a lag frame in the middle of the Jungle level to jump from one position to a prior frame, desyncing the entire run. Additionally, this frame is essential for both speedy play as well as RNG purposes, and mitigating this causes a slight time loss in the moment, a large time loss in the third level and possibly also two parts of the second.
Unfortunately this seems to be at random even in the same movie project or movie file. This happened several times but was remedied several times by just getting annoyed, coming back tomorrow, and it was fixed. It's so odd.
This file did sync for myself as well as Samsara and will likely sync for everybody.
However, this did not allow for the removal of the sega splash screen before submitting. I was not familiar with the TAS timing "from power on" when starting this, as such the splash screen was overlooked until after initial submission when I was contacted by EZGames69 mentioned I could. optionally, skip the splash screen.
Removing/changing elements, inputs, or even frames from the run destroys the entire run by desyncing both lag and RNG.
Thankfully, this does not seem to include the splash screen (as its removal would destroy the run within the first few frames let alone over 1,000 of identical play and RNG), however, upon fixing it, this seems to trigger the glitch in BizHawk.
I am looking into it further (apparently as well as Samsara).
I'm happy and proud of the 4:51.8 "RTA timing" and I'll think of that in my head, especially because that is how the speedrunning community views it.
Anyhow, now that I'm done glitch hunting I do plan to restart this from scratch at one point (no real way to go back and edit things without rocket science, I tried once and it took 2 hours to adjust from the ramifications of one frame change and that was only for the next level because that's as far as I had gotten) and will skip the splash screen, menu optimization (using default controls instead of my own preference), and pray the glitch either occurs on a frame that can be compensated by next (i.e. running, doesn't matter if the lag flip flops from frame 5 to 6 or vice versa, you're holding run and that is it) and as such is basically a non-issue (shouldn't affect RNG either in theory), or doesn't occur at all.
There's also a few things I would like to do, which saved time but produced undesirable RNG, which may be mitigated or altered in a fresh restart with the splash screen skipped and menu optimized.
For now, not counting prior to the game starting officially, I'm very proud of this run and hope you all enjoy.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Voted no for the threatening-style comment about rejection. Very immature.
350 re-records over 30 minutes says all it needs to. I am sitting at 20,000 for 3 minutes.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Ahh. Only other thing I could think would be editing how many pixels Mario travels forward per frame. (Almost like frame skip)
I stand corrected then.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
Can't you just continually do simple ROM editing until you get the speed so fast that it is unplayable & unTASable?
I don't think this hits that wall and as such is kind of pointless. You should hack into Super Mario Bros. and increase the speed to that level. Right now it's just arbitrary on what some guy made it. Picked a random number and went with it to create the ROM.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023My rule is quality TASes over quantity TASes... unless I'm bored.
