Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Niamek wrote:
Why would it not be acceptable? I don't see any problem in making a single file that comlete the game twice. One time for Donkey Kong and one time with Diddy Kong.
The TAS goals must not be too much obscure or exoteric.
Maybe it could still be accepted, you can't never tell for sure without first going through the actual judging process and analyzing all aspects. Still, it's just common sense that completing the game is something that should be done only once for each branch, except where there is some extra content after the main game quest that doesn't require to repeat most of the already seen game content. Some examples would be Pokémon Gold and Toki Tori.
Niamek wrote:
Eitherway, I don't think I've said it on the submission thread, but I enjoyed the movie and would watch a full completion of the game.
I'm looking forward to it too, that's why I'm trying to help about following the guidelines.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
PiePusher11 wrote:
ThunderAxe: Thanks! I'm trying! At the moment, I don't plan on TASing any other games, but I'm definitely considering TASing other categories of this game. My end goal is to have a 200% TAS of this game, which involves 100% DK, 100% Diddy, Time Attacks with both characters, and the Jungle Jams (minigames of sorts).
After reading the guide on GameFAQs, it seems to me that such 200% run would be not be acceptable; instead, Donkey Kong 100%, Diddy Kong any%, and Diddy Kong 100%, should all be submitted as separate runs.
Since Diddy Kong requires to beat the game once, it will be necessary to record the movie starting from SRAM. In order to create such movie, you need to go on File > Movie > Record Movie... > Record From: > SaveRam > OK. Then you will also be able to continue your movie in TAStudio. We usually don't accept save-anchored movies, except for cases as these. For more details, read the first section of the Movie Rules.
For the Time Attack: from what I've read, it looks like it doesn't introduce any real new content. If I'm right, then a run done in such mode couldn't get accepted in any case.
For the Jungle Jams: there shouldn't be any problems. Start from SRAM and beat all minigames in the same movie.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
PCachu wrote:
Is it just me or is there a lot of Sonic design philosophy in this game? "Run through the level on high or low paths as fast as your feet can take you to reach the end-of-level sign."
No, it's actually the TAS making it look this way. Every character has different running speed, and running all the way through the level without dying is actually pretty insane.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
boct1584 wrote:
I don't have any issues with personal projects, but if your run is suboptimal (which is what I'm getting from this post), and you KNOW and ADMIT the run is suboptimal (which I'm also getting), there is basically no chance it will be published. If I'm wrong, then my bad, but let me offer a suggestion. Upload it to the Userfiles.
If the run has chances or not, it will be decided by the judge who will handle this submission, after having finished the process of judging.
I'm still waiting for an encode, but as a watcher I can tell you that what I find more entertaining in RPG TASes is a good route planning, rather than overworld movement optimization.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Hi, I'm fine, what about you?
In order to enable the memory card emulation, you first need to open a PSX ROM with BizHawk, then you go on PSX > Controller / Memcard Setting.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Great TAS. Yes vote.
By the way, the second screenshot suggested has been taken after the movie end, right? I'm not sure if that would be applicable.
Edit: oh and also, why switching from SGB to GB?
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Turns out there is a major problem with this run. While there is already a similar run already published, this case is different. DS Game & Watch Collection doesn't feature any credits sequence, so there is no definite game end point and thus was possible to come up with a custom solution. But for the game of this submission run, GBA Game & Watch Gallery 4, there is an actual credits sequence that can be triggered after getting 170 stars.
Link to video
More details here: Thread #16824: Game & Watch Gallery 4
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
So, I've researched for a value that indicates the game difficulty. I'm not sure if there is a single value that controls it, but at least I've found the values that affect 4 distinct difficulty aspects:
SystemID GBA
03A3A0 w u 0 EWRAM score
03C121 b u 0 EWRAM max enemies
03C122 b u 0 EWRAM everyone speed
03C123 b u 0 EWRAM spawning rate
03C124 b u 0 EWRAM currently unknown
These values increase when the score increases. If you poke the score address to a bigger value, the following 4 values will change according to it. If you lower the score again, they won't change, suggesting that there could still be a master value somewhere.
Also note that these 4 values sometimes get temporarly lower...
Edit: getting the score to 990 or any higher number gets the same values. I suppose this could be considered the score required for triggering the maximum difficulty.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
#5730: The8bitbeast's PSX Crash Bandicoot: Warped "105%" in 1:54:18.52. Right now, it is currently being published in Moons.
Crash Bandicoot: Warped is one of the most popular games for PSX of all times, and the branch of this run features very extensive optimization and huge entertaining value.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I did enjoy watching this run enough for a Meh vote. That's a nice little game, but the portions of the game that needs optimization are minimal. It should still be eligible for Vault though; I'll tell in the next days.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
The movie file syncs fine on the Japan version, but the header refers to the USA version. Please fix it. Something like 7-Zip and Notepad will do. In particular, you need to fix "GameName" and "SHA1" fields.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Well, from what I've seen the game looks enough hex-friendly: you will probably need to resync only the two levels in which the improvements need to be implemented. So it should be fine keeping this submission active.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
The run itself isn't too bad, but I found two possible improvements that I need you to look at. Have here some unrefined movies from which you can see how to make the improvements:
1) The first eagle in 3-1 can be dodged without having to use the block. 21 frames saved, desync at 15763: User movie #43587441108271535
2) The "pressers" in Stage 4-2, at frame 28200, can be passed earlier. 68 frames saved, desync at 28427: User movie #43587447183350358
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Most games can behave different after seemingly trivial edits in a movie. I think most times it can't be explained without going serious with debugging and stuff.
Something similar happened with Zook Man ZX4 and Crash Bandicoot: The Huge Adventure, both GBA games: after a very minor edit in the WIP, some frames where wasted some levels later. (GBA Crash is continuing to resemble to Zook Man, and this is scaring me.)
By the way, succeeding or not, it sounds you did a good job looking if you could recover these 6 frames lost.
Also, it's not like a movie could be rejected for such a minor downtime, nor could get in Vault instead than Moons just because of that. You're doing a good job, working on it seriously and keeping improving yourself. I hope you will TAS other games as well, in future.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
feos wrote:
Bobo the King wrote:
Well, maybe you could waggle the joystick back and forth quickly enough that it interprets it as left+right, but I'll take a "I'll believe it when I see it" standpoint.
In that very same post you linked, it's also said this, with microscopic font:
AnS wrote:
(well, of course in the asynchronous circuits there's always a race condition of propagation delays, so the signal about releasing Left may somehow arrive later than the signal about pressing Right, and the sampling may just happen at this very moment, but it's too improbable and unreliable, definitely not something with which you should justify L+R)
We already accepted a TAS that makes 8000+ inputs for second, so why not allowing this?
Edit: I know the TAS in this submission is actually doing L+R inputs on the same frame, while the one I've pointed out makes extensive use of subframes inputs. My question is, if it would be possible to achieve the same results of L+R with very fast subframe inputs.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
I wonder, would be *theorically* possible to do L+R on a cabinet if you move the joystick toward the L and then the R during the same input poll, assuming a superhuman moving faster than the electric signals inside the hardware?
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Hello again.
Despite the fact that I'm feeling much better this year, I'm still unable to be productive. Basically, my only activity is TASing, and oddly, I need to rest a lot all days and save energy in order to do it. I'm not even playing games casually anymore, from about 2 or 3 years now.
In my previous post I've given a precise diagnosis of my problem, but despite that I'm still unable to understand why I am like this.
I've always thought to be doing fine all alone, but the truth is that I'm not. Yet, I'm the one that is keeping the distance, even if I feel lonely.
I don't know what I want.
Thanks for reading, and sorry to anyone who could have been waiting to read about how I did overcome insomnia and paranoia. Maybe I'll write about it at another time.
Well done! It looks well executed.
By the way, I've just noticed that the actual value type for the coordinate values is Fixed Point 16.16: go on the RAM Watch window, double click the addresses and then click on Display Type. This should make things simplier.
Experienced Forum User, Judge, Published Author, Skilled player
(1288)
Joined: 9/12/2016
Posts: 1645
Location: Italy
Solved the mystery. The right values can be found in the EWRAM. Go on the RAM Search window > Settings > Memory Domains > EWRAM. Start a new search.
You can either copy the right addresses from here:
SystemID GBA
022010 d s 0 EWRAM X position
022014 d s 0 EWRAM Y position
022018 d s 0 EWRAM X speed
02201C d s 0 EWRAM Y speed
This is what you get when you open a wch file with notepad. These values are all 4-bytes, signed, Little Endian.
