Joined: 6/4/2006
Posts: 267
Location: CO
Damn 37 speed is fast. ;) It's not quite the same as a walkathon though because pressing y+forward in the air is still a form of running, just not on the ground. Finished YI3 and the beginning of YI4: http://dehacked.2y.net/microstorage.php/info/3337/thegreginator%20SMW%20Walkathon%20YI3.smv
Joined: 8/2/2004
Posts: 98
thegreginator wrote:
Damn 37 speed is fast. ;) It's not quite the same as a walkathon though because pressing y+forward in the air is still a form of running, just not on the ground. Finished YI3 and the beginning of YI4: http://dehacked.2y.net/microstorage.php/info/3337/thegreginator%20SMW%20Walkathon%20YI3.smv
<3<3<3<3<3<3
-Oz. If practice makes perfect, but nobody's perfect, why practice?
Joined: 8/12/2004
Posts: 651
Location: Alberta, Canada
bkDJ wrote:
the mario64 crawlathon used to be "must hold z" but then they realized backflipping and picking stuff up allowed for faster movement, so it became "must crawl as much as possible".
I actually knew that from the start, but it totally defeats the purpose of the crawling run.
Tompa
Any
Editor, Expert player (2139)
Joined: 8/15/2005
Posts: 1932
Location: Mullsjö, Sweden
Hehe, nice Thegreginator :D. Is it faster to go in the water instead of pressing the P and jump on the koopas...? Probably:P.
Joined: 7/26/2006
Posts: 1215
Tompa wrote:
Hehe, nice Thegreginator :D. Is it faster to go in the water instead of pressing the P and jump on the koopas...? Probably:P.
not if he chooses to lose yoshi and jump off of him to where the koopas are.
Tompa
Any
Editor, Expert player (2139)
Joined: 8/15/2005
Posts: 1932
Location: Mullsjö, Sweden
That could also be a way... =P.
Joined: 6/4/2006
Posts: 267
Location: CO
I need Yoshi for DP1, so I dont want to jump off him here, since getting him again at DP1 would take too much time.
mrz
Joined: 8/24/2006
Posts: 119
Location: New York City
So your going to do a shell jump in YI4? :D
Post subject: Diffrences between (E) (1.1) and the used (U) (1.0) version
Joined: 12/17/2005
Posts: 77
Location: Bjurholm, Sweden
I don't know if this has been brought up before, if it has, I'm sorry about that. I may have done some mistakes here but I don't think I have. I was playing the Eu version of SMW and noticed that the speed value was 63 and realized later that it was because it run on 50 frames and I started testing if the E version could've been faster and it seems like it is. 63 * 50 means the speed is 3150 per second on the E version. 51 * 60 means the speed is 3060 per second on the U version. So in theory it should be faster so I tested it out on YI2. From restart without the hop glitch the frame counts looks something like this: YI 2 (no hop glitch): (E) (1.1): 1400 (28) - 2749 (54,98) = 1349 (26,98) (U) (1.0): 1596 (26,6) - 3265 (54,41666..) = 1669 (27,8166..) The numbers inside the ( ) is how many seconds. you can see that the E version is a lot slower at starting the game but the time inside the level is slightly faster. With the hop glitch: The E version YI2 was: 1320 (26,4) The U version YI2 was: 1631 (27,18333...) With the flying: (E) (1.1): 113 - 1396 = 1283 (25,66) (I accidently flied at 63 instead of 64) (U) (1.0): 125 - 1698 = 1573 (26,21666...) So the E version is indeed moving faster but I don't know these things between the levels and such things like that. But I would like to see this tested out a bit more. I also noticed that on the E version you can Jump a bit higher/longer, for example the first castle before the door you can jump over the whole gap without grabbing the net. You do also "fly" higher. I'm sorry if this was confusing but this is how I think. I hope this helps in some way though I'm not sure at all that the E version is faster. // Arne the great Edit: Btw, THIS should be faster at least. I bowser could be faster on PAL too. I don't think this could save any time on the any% but it might save time on the full.
Joined: 6/4/2006
Posts: 267
Location: CO
Cool discovery. But I vote right now that we continue using the (U) version for all SMW TASes. Obsoleting a run just because of a different ROM seems like a bad idea to me, just look at the problems it created with SMB.
Senior Moderator
Joined: 8/4/2005
Posts: 5769
Location: Away
SMB had nothing to do with this discovery, since the new run was using unconventional methods of going through and finishing the game. This particular discovery means that a ROM change expands the possibilities for the better, like with Rygar (or some other PAL run on the site).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Experienced player (600)
Joined: 10/23/2004
Posts: 706
thegreginator wrote:
Cool discovery. But I vote right now that we continue using the (U) version for all SMW TASes. Obsoleting a run just because of a different ROM seems like a bad idea to me, just look at the problems it created with SMB.
I vote that runners move to the E version if it's faster. In the world of TAS, faster is pretty much always better. And, in this case, the game may offer greater opportunity. Perhaps we can pass all of the spikes in Bowser's Castle without stopping, things like that would make the whole run more enjoyable.
Current Project: - Mario Kart 64
Joined: 5/29/2006
Posts: 200
It does raise some issues though. I think most people will agree an old run shouldn't be obseleted just because the new run uses a different ROM without any strategy change. However, it would be difficult to tell whether a strategy change on the new ROM is faster due to the change in strategy, or solely due to the PAL version. If someone can figure out how to measure that, then I see no issue, but otherwise...
Joined: 10/24/2005
Posts: 1080
Location: San Jose
We often use other versions of roms (1.0 vs 1.1 vs 1.2) because they potentially could have more glitches (like OOT). If the (E) version is fater, I begrudgingly say we should use it. I don't like it, and I wouldn't use it myself, but sometimes you gotta make sacrifices. Also, it's purely up to the runners. If they still want to use the (U) ROM, their run shouldn't be rejected because they used that version.
<agill> banana banana banana terracotta pie! <Shinryuu> ho-la terracotta barba-ra anal-o~
Joined: 1/17/2006
Posts: 775
Location: Deign
The rules page wrote:
If there is already a published video, do not use a different ROM than what it uses, unless the new movie is obviously better for reasons that are not only caused by the version change, or the ROM is superior (or a language change), and you can show how it should be compared to the existing movie.
So far it seems to me that all improvements are caused only by the version change. Cheetah, if the improvement is caused by a strategy change rather than a rom change, chances are the author would know.
Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign aqfaq Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign Deign
Post subject: Re: Diffrences between (E) (1.1) and the used (U) (1.0) vers
Editor, Experienced player (729)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Arne_the_great wrote:
I was playing the Eu version of SMW and noticed that the speed value was 63 and realized later that it was because it run on 50 frames and I started testing if the E version could've been faster and it seems like it is. 63 * 50 means the speed is 3150 per second on the E version. 51 * 60 means the speed is 3060 per second on the U version. So the E version is indeed moving faster I also noticed that on the E version you can Jump a bit higher/longer, for example the first castle before the door you can jump over the whole gap without grabbing the net. You do also "fly" higher. Edit: Btw, THIS should be faster at least. I bowser could be faster on PAL too. I don't think this could save any time on the any% but it might save time on the full.
I see here a bunch of version changes that leads to strategy changes. quite simply, I see a version change leading to 5 or most likely more seconds saved in the 11 exit. the 11 exit was improved by 4 and then by 3 seconds using [U]. I think this discovery should merit a TAS, maybe not obsoleting the current run, but definetly because a much faster time is achievable. based on what Arne has said; we have a few very slight changes in gameplay, namely faster speed, differences in jumps, and higher flying. All of these strategically change how the 11 or 96 exit will turn out. All levels can have time saved from increased maximum speed; not sure about the water level though. The Bowser spikes can be surpassed, which saves seconds alone Higher jumps can lead to some shortcuts, or to not lose time; YI castle comes to mind, with the one lava pit that is just too barely long that you have to grab the screen wall.
Homepage ☣ Retired
Joined: 5/3/2004
Posts: 1203
Great bit of testing, Arne. I hope someone makes use of it!
Joined: 6/15/2005
Posts: 1711
I hate ROM version changes. Always have, so I'm unhappy to hear about this. It was already known that the european version meant you could clear the lava pit in Castle #1, but it was also believed the 50 frames per second meant the game was much slower anyway. I'll do some basic testing of stuff if I can find the rom. Guys, please brainstorm and come up with other close call type things like the lava pit in castle 1 or the spike room in Front Door. The more ideas the better. I can't think of much else at the moment.
Zoey Ridin' High <Fabian_> I prett much never drunk
Joined: 4/16/2004
Posts: 1286
Location: Finland
Fabian wrote:
I hate ROM version changes. Always have, so I'm unhappy to hear about this.
Same here. IMO things that make the run faster only because of a version change should not be taken into account. I think it should be a rule to always use the (U) version if one exists. Too bad it seems that not many people (including the one that counts, Bisqwit) share my view.
Joined: 8/12/2004
Posts: 651
Location: Alberta, Canada
From the past I was under the impression that he did share your view. I think I have seen him say that you should use the same region when trying to obsolete a movie. Keep in mind that Super Metroid would be have been faster too if run with the E version, but it was decided to keep with the U version.
Joined: 6/15/2005
Posts: 1711
Ok some initial tests of the physics: First without cape. It takes 26 frames for the P-meter to start its acceleration. Compared to 27 for US version. The P-meter increases like this: +3 +3 +3 +3 -1 and repeats every 5 frames. So 11 / 5 = 2.2 per frame. I haven't tested if you can cheat here like short charging works in Super Metroid. Once it reaches 101 you can start flying, or run at full speed. It maxes out at 104. Compared to regular version which increases at 2 per frame, once it reaches 108 you can start flying, max is 112. If you don't feel like doing math, this means slower acceleration in PAL. For example, according to Arne, you don't reach flying speed in the ghost house for the 11 exit run, so flying through the stairs wouldn't work (with slowdown it would work, of course). "Middle speed" follows the pattern 46-45-43-45-44 and repeats. Compared to 37-36-35-36-35 and repeat. Not a lot different. Full running speed though. Bleh. I wrote down the first 100 or so values. There are some patterns (especially 61-60-60-60-59 repeating a lot) but it's definitely complicated. Thankfully you wouldn't need to worry much about it in a TAS, except that this game's equivalent to 6/5 might be horribly complicated, I don't know. In case someone wants to try making some sense of it, here are the numbers: 60-60-61-60-58-58-59-61-60-60-60-59-61-60-60-60-59-61-61-58-60-59-59-60-58-60-59-59-60-58-60-60-61-60-58-58-59-61-60-60-60-59-61-60-60-60-59-61-61-58-60-59-59-60-58-60-59- 59-60-58-60-60-61-60-58-58-59-61-60-58-59-61-60-60-60-59-61-61-58-60-59-59-60-58-60-59-59-60-58-60-60-61-60-58-58-59-61-60-58-58-59-61-60-60-60-59-61-61-58 Flying. Goofy as hell. Normal flying is 63-62-61-60-58 and repeat. BUT!!! Sometimes (and I don't know when) you all of a sudden change pattern to 64-62-61-60-59 and repeat. Also when you hop at speed 64 you can't jump higher than about one Mario's length up in the air. This is regular jumping, spinjumping doesn't work at all. Hopping at speed 63 works like normal. I've reached the 64 speed cycle when first flying into a chuck, and also once when taking off right after running up a pipe. Arne has a theory that it's after the first jump in each "room" you reach 64, but even if that's accurate, there's obviously more to it (the chuck thing). Another thing. The swoops/dives (gaining lots of height while flying) are also really goofy. First of all the height you gain is HUGE. Wow. Also, after you do a dive, regular flying starts behaving differently, unless I'm going completely crazy. Each time you catch air, you gain more height than catching air normally would result in. All in all, I hated this game's physics. Not being able to hop at speed 64 is of course the huge thing I imagine will make everyone else agree. These are just some quick thoughts, not very organized. Sorry about that. If after some more testing it turns out an 11 exit run is faster in minutes:seconds, I'll ask JXQ if he feels like redoing it with me, even though I'm against switching rom versions, like I said earlier. I'll definitely not switch version and restart the 96 exit run, no matter what. Edit: Another thing I tested but forgot to mention. You can't survive stompers (the things in the sidescroller room in castle 1) by jumping every frame. You jump through the floor instead, and die obviously.
Zoey Ridin' High <Fabian_> I prett much never drunk
Joined: 12/17/2005
Posts: 77
Location: Bjurholm, Sweden
I don't think we need to switch to PAL just because Mario runs a bit faster, I would like to leave it up to the TASer but however to bash the PAL version more, the world map i about 7% slower in PAL and that would probably make it even more slower. Although we should have in mind that it actually could be faster and we need to test more things, I suspect PAL have faster bosses (bowser?).
Active player (277)
Joined: 5/29/2004
Posts: 5712
BoltR wrote:
Keep in mind that Super Metroid would be have been faster too if run with the E version, but it was decided to keep with the U version.
Are you sure? I thought somebody tested it and decided that the one extra sequence break didn't make up for the more imprecise movement. Maybe I'm wrong.
put yourself in my rocketpack if that poochie is one outrageous dude
Post subject: Small mario SMW run
Joined: 2/18/2007
Posts: 63
Location: New York City
A while ago Jimsfriend (I think that was his account name) was going to work on a SMW tas where mario would never change his initial size in any way. What happened to it? Is he still working on it? Does anyone else plan on a SMW run?
Joined: 6/15/2005
Posts: 1711
hi scwizard, jimsfriend did a non-assisted (console) run of SMW which from what I've heard is extremely awesome, I haven't seen it yet though. Currently he's doing a 120 exit Super Demo World (hack of SMW) TAS. There is a small mario TAS going though, which aims to complete almost all of the exits (I believe it will do 85 exits or something). It's being made by thegreginator and skamastaG, and I can promise you and everyone else that it will be very awesome too. Hope this helps, if you have any other questions feel free to ask. Edit: In case you're extra interested, there is some discussion about the small run, as well as a small teaser video of Castle #2, buried somewhere in the last 20 (or maybe even more :) ) or so pages of the main Super Mario World thread in this forum.
Zoey Ridin' High <Fabian_> I prett much never drunk