Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Very cool run!
I have a question, though: What happens if you use a Repel, say, one step before the "Repel doesn't work anymore" text appears?
I saw some speedrunners use that method in other Pokemon games to save time, since it skips the message, but I don't know if that works here.
Also, it should probably be said that V1.0 of the game was used (GoodGBA gives me Pokemon - Fire Red Version (U) (V1.0).gba). The run doesn't sync on V1.1.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
mkdasher wrote:
- Getting an encounter on first grass step: I'm unsure about this. I know RNG mod 100 < 5 works for like the 5th step, but it's something lower for first step.
Masterjun wrote:
The game uses the upper 16 bits of the normal RNG, modulo 100 and gives an encounter when under 60.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Due to the lower frame rate (50 Hz) and the fact that games are usually not modified in order to compensate for this, PAL versions of ROMs are generally not preferred, unless there are significant technical and/or entertainment merits to using this version.
I don't know if the run syncs on the (U) version, but if not I don't really see a good reason to use the (E) version over the (U) version.
Still waiting for an encode though.
PAL games were only 50 Hz on console until Gen. 5 (with a few exceptions in Gen. 6).
Oh, ok then =)
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Due to the lower frame rate (50 Hz) and the fact that games are usually not modified in order to compensate for this, PAL versions of ROMs are generally not preferred, unless there are significant technical and/or entertainment merits to using this version.
I don't know if the run syncs on the (U) version, but if not I don't really see a good reason to use the (E) version over the (U) version.
Still waiting for an encode though.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
You would have to fill all the level bytes and the level score, then run the save routine. That would take a long time to execute and this is actually pretty unstable (since I use DMA to let the joypads refresh).
Also, IIRC there is no indication in the credits scene that you have 100%, so it wouldn't even be visible for the normal viewer...
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Just looked into the code and wanted to say what I found is that if you walk from not-grass to grass, it is always (I think) possible to avoid an encounter.
The game uses the upper 16 bits of the normal RNG, modulo 100 and gives an encounter when under 60. And since normal RNG advances once every frame (and 2 times when it checks for an encounter, and like 87 times when it calculates a wild pokemon) it is possible to manipulate.
The second step however (and probably every other) is different. I might look into that soon...
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Thanks for all the good feedback =D
Carl Sagan wrote:
Do you know if it's possible to do something like this with the infinite tongue glitch?
I have yet to test things out since that is not that easy to tracelog and find out about (basically it's hard to figure out what exactly is going wrong and how to use it right).
Carl Sagan wrote:
Also sometimes the game crashes when you bring certain sprites, and other times bringing the exact same sprites does not cause a crash. Any idea why this may be?
Yes, the game often jumps to complete random addresses (sometimes even RAM) that can be different every time you jump into the goal ring. Also, the game is good at jumping over code with a lot of 00 in it so sometimes it can somehow get back to normal code.
Carl Sagan wrote:
These rats are also in 6-2 and 6-4, but I'm not sure if it would be faster to get to them or if those levels have all the necessary components to set up the glitch.
6-2 is long as hell + the time it needs to get enough coins for warping and 6-4 doesn't have a goal ring.
Carl Sagan wrote:
I'm also not sure about the route you took through 2-2
I used this to make my route, I wonder which route you think is faster...
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Carl Sagan wrote:
I think Masterjun is working on an even more SMW-style glitched% run using other tricks Arnethegreat, the #yoshi IRC community, and I have discovered over the past year or so, and it would be substantially faster than using this null sprite glitch, suffice it to say.
Just saying that I am indeed using this null sprite glitch. What I can say is that I won't use it to take sprites to other levels.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
MESHUGGAH wrote:
but could you tell me which part of this run was hard? If you say the last trick, then I'm apologize in advance, since I really didn't wasted any time to understand why would it be really hard to make that positioning trick.
Well since the last "positioning trick" is not only positioning yourself, it is positioning the camera too (X and Y position), which is quite a hard thing to do. (And trust me, that was really a hard thing to do, since I did that run myself)
MESHUGGAH wrote:
Yes, I've read that you made many efforts and couldn't make an improvement
Well there was one idea that unfortunately didn't make it.
MESHUGGAH wrote:
I would like to know the difficulty in implementing the same strategy
Doing a run with only the goal of doing the same stuff as the previous movie in mind is not as hard.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
MESHUGGAH wrote:
Let me explain you how converting movies are done with the example of [1945] SNES Super Mario World "game end glitch" by Masterjun in 02:36.40 (the obsoloted one).
1. As far as I saw, FCEUX and snes9x has "bigger" loading times compared to BizHawk (nes and snes (two different) core), so just add enough empty frames to easily use the previous input (to start the level at 1st frame, move in the level at first frame, so you will need to use this everytime the game loads something)
If we don't look at the frame rules and assume that this would sync, the it would be copied input and so:
Movie Rules wrote:
it's generally accepted that if you simply copy large chunks of gameplay from an earlier movie as they are, that effectively makes a new submission coauthored.
In case he redid the level it would be fine.
MESHUGGAH wrote:
2. Because the emulation is different, the RNG cycle is also different so all you need to do is to read and reimplement the strategy (rolling the RNG until a specific value got by fishes bouncing).
In this case the RNG cycle wouldn't even change, but copying input will have the same result as above. If the RNG cycle would change, then it wouldn't be as easy to just "reimplement the strategy".
MESHUGGAH wrote:
3. When you are done, i think there's another tricky place that depends on very specific subposition which only should be a problem if you get a random lag-frame or it's misplaced, this time you need to reoptimize your sub pos.
Well it depends on 3 values which are only correct on a single frame so that is probably the hardest part of the run.
MESHUGGAH wrote:
If someone would do this and submit this movie and say "I made enough attempts but it's can't be improved" and of course will yield in a slower time because of the longer loading times, should this movie would be accepted if there wouldn't be the new published TAS ([2380] SNES Super Mario World "game end glitch" by Masterjun in 01:39.74)?
If the person really did all the hard stuff by himself, I would have no problem with seeing my run obsoleted. I would contact him and ask if he would be ready to try out a few changes to this run to make it even more optimized.
I see that in this run the problem is that optimizing is an easy thing. You just have to press a button the first frame it does something. The thing is, that in many other games, doing things on the first frame possible is not always the best solution.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
turska wrote:
Masterjun wrote:
Tangent wrote:
SmashManiac wrote:
- p4wn3r's run is on a deprecated emulator
This part's incorrect, or at least not fully accurate. VBA's deprecated, but it's still accepted
And where is that part now not fully accurate?
I mean of course it can still be accepted, else it would be listed as Not accepted at all.
However, deprecation can still be a reason for rejection.
That's not the case. VBA movies are currently accepted, and there are no plans for rejecting VBA submissions; if there were, you'd see news about it on the front page like with the Snes9x 1.43 deprecation. If VBA has major issues emulating a game, then emulator choice could be a reason for rejecting movies of that particular game, but outside that there will be no VBA rejections for the foreseeable future.
How the hell was that relevant to anything I said?
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Tangent wrote:
SmashManiac wrote:
- p4wn3r's run is on a deprecated emulator
This part's incorrect, or at least not fully accurate. VBA's deprecated, but it's still accepted
And where is that part now not fully accurate?
I mean of course it can still be accepted, else it would be listed as Not accepted at all.
However, deprecation can still be a reason for rejection.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Cyber_Kun wrote:
My point being here is that if all the port is some bot work that some idiot at TAS'es like myself could do with the bare minimum knowledge
And my point wasn't that I'm bad at this, my point was that it is indeed not easy to recreate these TASes to be optimized.
That's also why I think that if someone managed to recreate the current any% run on SMW on an accurate emulator with every single optimization the current run does too, the credit to him is justified. Except when a single flaw in the run is discovered, then it has no right to obsolete the previous run.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
Experienced Forum User, Published Author, Site Developer, Skilled player
(1971)
Joined: 10/12/2010
Posts: 1179
Location: Germany
Tangent wrote:
but dupicating someone else's work is also a lot easier than doing new or original
If you mean by "duplicating" that they are doing the same RNG and lag reduction strategies and such, then I have to disagree.
I agree that it is harder to do a new TAS, but it is not harder than finding optimizations and then applying them.
Warning: Might glitch to creditsI will finish this ACE soon as possible
(or will I?)
