Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Oh nice. Thank you for testing that parrot14green. I haven't gotten very far in kuwanger yet(namely because of an unsuccessful camera-less elevator climb, lol) so I'll definitely go back and implement that. I am sure sub pixels are the reason the lower coordinate sometimes doesn't work, but luckily the higher one is more optimal.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Fractal, have you ever messed with these dash speed gap jumps in the climb before Vile in Sigma 1?
https://www.mediafire.com/?l0g04e45e1b82dr
They're pretty easy to set up in Sigma 1, but I am wondering if it is possible to do on the first ladder in Kuwanger's stage. It seems like the ladder gaps in both stages have the same width and distance from the wall, but I can't seem to get it.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Wow I was able to save 129 frames w/ that capsule trick. Ty for showing me that!
Ohh so you can zip into the cave!
I'm not sure I understand exactly what you mean. You can zip down into the cave once the camera lock is lifted, but then you can't move in the cave because of camera lock? Does that mean there is a second camera lock? Or that only the vertical part gets lifted when you zip down and you have to go back to allow horizontal movement?
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Omg parrot14gree. These are amazing, lol
This is now making me wonder. Although the camera lock prevents you from moving left/right until it is lifted, does it prevent you from moving up/down? I am thinking perhaps you could zip from inside the armor capsule on chameleon's stage down into the cave so that you already down there and can start moving right as soon as the screen lock is lifted.
If possible, this could potentially save time since the advantages of chameleon 3rd only amount to about 3 seconds (and even less if an underwater phantom grab in chameleon is possible).
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
What lua should I be using to see X/Y position values in snes9x (or BizHawk for that matter if there is one)? The one I'm trying to use now is giving me an error that says "\megamanx.lua:29: attempt to index local 'fle' (a nil value)".
Does that essentially mean that (1600,252) is the upper left corner of the area which the 2nd set of platforms spawns, and then you just fly horizontally from there until you drop down to get past the corner of the building (and the e-tank spawns as some point during that horizontal movement)? If so that would make this pretty simple and easy to optimize.
Do you have a way of also finding out the coordinates that trigger the graphical glitches? I think it may possibly be worth avoiding those glitches to reduce lag later on.
That grab after the capsule looks really interesting. I'm gonna mess with that today to see if I can keep it from despawning somehow.
Between this glitch, the zip in Mandrill, and some unfortunate RNG that I ended up taking in my run, I think I'm going to be making a v2 of the X1 100% TAS pretty soon lol (and Any% too). Right now though I'm working on X2 Any%. Is there anything in that game you've found that isn't in any TASes/RTAs?
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Haha I wasn't sure what to do w/ that blank corner of the overlay, so I put that gif there since it has Storm Eagle in it (and cause I like Idolm@ster). :P
Oh nice, I didn't even think to use the second platforms. That looks like it cuts out a good 6-7ish seconds of backtracking.
Now I'm wondering if it could even be potentially faster in 100% to ditch the first platform in the place that it likes to despawn near the "Airport 1001" logo, then go under the E-tank building to grab a second platform and ride it up to get the E-tank then over to the capsule instead of taking the detour with the first platform.
On a side note, the more you glitch out the background/sprites it seems to generate a lot of extra lag, and the further left you are once you rise above the clouds in the initial ascent seems to get less the laggy glitches.
I also think this could save time at the beginning of Sigma 2 and possibly Sigma 1 (but that might be tricky since you also have factor in taking damage for Vile).
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
I have been messing with grabbing the platform in Eagle, and this is what I have so far for Any% and 100%.
The main problem I'm having with the 100% version of the trick is that if you take too direct of a path to get to the E-tank you despawn it, so you have to take kind of a detour.
Do you know if it would be possible, Fractal, to find out what exact location causes the E-tank to depsawn in order take the most optimal route around it using lua scripts or something?
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
@Atma
I actually have no idea how it works or why it works aside from the obvious (spawning an item the same frame you collect another w/ a boomerang), but there was a short discussion about it in the Mega Man X thread.
Also, interestingly enough you can even grab things that aren't even items with it, such as boss doors. I wonder if you can even grab capsules with it, lol.
@RaijinXBlade
That's pretty crazy. I've been messing with it for a little while now but still can't seem to get it.
@mklip2001
The first place I saw the Kuwanger elevator skip was in FractalFusion and Rolanmen1's 100% WIP.
I wanted to do the swagdouken on Velguarder Sigma, but unfortunately I didn't have time to do it and dboost through the head and let the claw take me to the other side, so I had to choose 1 or the other. This 2pronged attack thing is some silly thing I do in my RTA runs, so I really wanted to include it in the TAS as well, lol.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Thank you again everyone for all the votes/comments.
RaijinXBlade wrote:
I have a question about Mammoth's stage. I noticed your climb to the arm upgrade had you double back on the first wall kick to wall kick a second time before ascending.
In an old freerun TAS of the stage I made a good while back, I managed to avoid having to double back.
Were you just not able to avoid that, or does it stop some slowdown by doing that?
Hm I just tried doing what I think you were describing and I can't fit between the wall on the right and the block on the left without bonking my head on the right ceiling.
I've been messing around with the sled zip after the E-tank in Mandrill, but I haven't been able to get it just yet. The flickering lights make it pretty hard to tell exactly where I'm placing the sled.
Though I did find out that you can start the charge for the sled at least 11 frames earlier if you use a phantom grab to get it. I did it with a drop from the 3rd mech in the stage but I'm sure it would be more optimal with a drop from the 4th. Strangely enough, when you approach an e-tank from level ground (as opposed to from higher up like in Armadillo's stage), it doesn't spawn until you get much closer to it (around the right side of the nearest ladder in this case). I was able to test this in Armadillo's stage since that e-tank spawns in the air, but is already on the ground if you spawned it earlier.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Yeah the intro stage bees (and the submarines in Octopus' stage) are transparent until they reach ground level.
@FractalFusion
Ah yeah I think I figured out that the optimal pattern is for him to dash and then walk after you interrupt his jump with a shot, but I had spent so much time trying to manipulate him already that I just wanted to continue with the run at that point, lol.
Is the thing in Spark Mandrill you're talking about something like using the sled to glitch though the first wall and forgo the first 2 ladders with the same trick from Sigma 3 in your glitched walkathon?
Thank you everyone for watching it and for the comments!
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Ah I see, thank you.
I think I will still want to find out how much more optimized my run is for my own curiosity if nothing else though. Is there a still accurate but less tedious way to do this than the method I had in mind?
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
I just finished making a Mega Man X 100% TAS that beats the currently fastest published run by 400 frames and am working on my submission text, but I'm curious to know how much more optimal my TAS is since Mega Man X has a substantial amount of lag which isn't accurately emulated on Snes9x 1.43.
So I have a couple of questions regarding this: Is something like counting lag frames and subtracting them from total frame counts something that would be done during the adjudication process, done only if my TAS was slower due to emulator differences and this was necessary to prove I had more optimized gameplay, something I would have to do myself before submitting, or something I can do if I want but isn't necessary for submission?
Also, what would be the most efficient way to go about doing this? Is there any alternative to frame advancing all the way through both TASes and identifying and counting each lag frame individually? The lag frame counter doesn't seem to be very reliable in either emulator.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Akisto wrote:
it could be so easy to cheat on emulator by having a false rom which again is dumb because its just as easy to splice a run let alone add a rom to your flashcart and trick around with it without anyone noticing it.
This continues to boggle my mind. It seems as if 99% of the speed run community is so ignorant that any one could splice console footage together using virtualdub and no one would even think twice to question its legitimacy, lol.
I think that the topic of "World Record" is just a sensitive one for some people and that they will come up with whatever excuse they can to pretend their favorite streamer has WR over someone with a better run.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
The awkward thing about the mmx community is that the majority of top tier players play on original hardware, and some of them tell their viewers on stream that keyboard is the "superior input device" to make it seem like that is the reason for the disparity between the fastest run and their own, but refuse to switch to it either because they secretly acknowledge the inherent disadvantages of keyboard/emulator which I mentioned in an earlier post or because of this widespread ignorant idea that their runs are more "legitimate" because they think it's impossible to cheat on original hardware.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Ah I figured as much since you said certain games and not mmx specifically. Mmx is really the only game where I get flack for my input device because of the game's popularity, so I figured those things would still be worth stating for the record, lol.
I agree that neon jumps are easier on certain keyboards but harder on others; it all depends on the size/actuation of the keys and how deep the valleys are in between them. Regardless, emulator is already separate in x2 because of inaccuracy and the x3 community just recently decided to consolidate the categories after the frame testing Luiz and I did on lag, so I doubt anyone would want to change them back.
@Alakorn Darn, looks like I'm not a real player then. Apology accepted.
Oh and you don't have to worry about the frame rate discrepancy between emulator and console. RTA runs are measured in real time (hence "real time" attack). My run is the fastest by 11 seconds (it would be by 17s if I played on a 60.098814fps platform), so there's no need to worry about it being slower by any metric.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
PJ wrote:
The reason I thought they were different categories for MMX is because hetfield plays on keyboard and Caleb uses an SNES controller. Keyboard has some distinct advantages for certain tricks in certain games.
This is not the case in Mega Man X. On a snes controller you can mash fire twice as fast per wrist movement as you can on keyboard simply by wiggling the controller back and fourth. You can attribute an objective time loss to this (I can usually kill bees 1-1.5 seconds faster on controller than I can on keyboard).
You also have a dpad which can be rotated to reach almost 60hz by itself to get out of octopus grips after only losing 1 hp, which is virtually impossible on keyboard. This saves nearly a second per grip as well as an additional second for every extra hit that it eliminates by the end of the fight (the thresholds for this are octopus siphoning 2hp, 5hp, 8hp etc..., so an additional 1 second lost after 1 grip, 2 seconds after 3 grips, etc...).
So where exactly does keyboard gain its advantage over controller in an mmx run? A lot of people say keyboard has an advantage over controller simply because the fastest runs are on keyboard whereas much more popular streamers play on controller, but they themselves have never played on keyboard so they have no way to form any sort of educated opinion on the matter. The most common phrase thrown around is "a finger for every input". Not only is this argument completely invalidated by the fact that you can have a finger for every input on SNES controller (besides dpad which would be unecessary, and even my directional inputs are shared by the same fingers that I use for L and R buttons) by playing fight stick style (Theherpderp88, Darkomegamagic, and Caracarnvi are a few MMX runners I know of who do this), but where exactly does this input setup save any time whatsoever or allow someone to do tricks that would be otherwise impossible? The answer is nowhere. Controller is unquestionably the superior input device to keyboard for MMX when you look at the facts, and keyboard offers absolutely nothing in return. Why do I speedrun on keyboard then? Because I was too lazy to buy a USB converter and a snes controller when I started learning the game.
As for the OP, I told you this maybe 3 or 4 times in Caleb's chat and it still never seemed to quite sink in. You can call the 2nd fastest time the "World Record" all you want, and it won't change which run is actually faster (not to mention the emulator I play on is unequivocally slower than console). It doesn't really matter what you think, neither yours nor anyone else's opinions change facts. Even if all of the leaderboards in the world decided to deprecate the usage of emulators in RTA, I would still have the fastest and most optimized run.
As far as legitimacy is concerned, I can't really put it any better than Patashu. My credibility as as a player has been established by streaming my practice and runs of MMX1-4 almost daily over the past year. I can guarantee that there is not a single person who actually watches my stream on a regular basis who doubts I have the skill to pull off the times that I do. I will never stream attempts with mic or webcam because they distract me from concentrating on runs, and I do not stream to entertain. All I care about is the quality of my play, and if this causes clueless individuals such as yourself who think that "TAS" is synonymous with "emulated RTA" and that Snes9x 1.5x emulators have no lag to question the legitimacy of my play, I guarantee you I won't lose a wink of sleep over it.
And since you do seem to be so confused about the difference between RTA and TAS, I figure I should at least let you know that tasvideos.org is not the best place to submit your complaints about RTA leaderboard policies. Perhaps you should try the RTA leaderboard forums instead: http://jokaah.com/mmrta/forums/
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
hegyak wrote:
Hetfield90 wrote:
Also regarding Bizhawk, whenever I try to load a savestate while making a movie in a either of the two cx4 chip games(X2/X3), all things related to the chip either become invisible or disappear entirely which even soft locks the game at one point. Is there any way to fix this?
I think that particular bug was fixed. It was related to how the savestate method behaved. If it still being bad in BizHawk 1.8.1 let me know.
I just downloaded BizHawk 1.8.1 but can't seem to get X2/X3 to work in them at all. When I open either rom, the frame count starts running but the screen remains black.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
creaothceann wrote:
Super Metroid for example has "realtime" runs and "ingame time" runs - there's a timer that shows its time in the ending, but it pauses during door transitions, in menus and (I guess) cutscenes.
As far as I know every TAS of the SNES Mega Man X series games has gone by real time, so I want to make my TAS while aiming for fastest real time completion as well.
Would the most affective method of comparing them then be to discount lag frames from total frame counts, as well as show how many additional gameplay frames I am causing by doing certain lag reduction strats that the Snes9x 1.43 TAS doesn't have to do? Also would these comparisons be something I would have to include as "proof" that my run is more optimized as part of my submission? Or would these be things done by the people who are judging my TAS to determine if it gets accepted or not?
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
creaothceann wrote:
Personally I'd just subtract the lag frames of the SNES9x TAS from that TAS's frame count and the same with the BiHawk TAS, then compare both. But BizHawk is more accurate anyway, so imo it should be able to replace any SNES9x movie.
I thought about this method, but then doesn't this undermine the fact that lag reduction is a significant part of optimization in tool-assisted as well as real time speed running? For instance, there are a lot of places in the game where dust sprites from dashes need to be avoided or strike chains omitted to reduce significant amounts of lag (resulting in more gameplay frames/less lag frames and faster real time completion), which FractalFusion and Graveworm's TAS takes into account. If I were to use this method and aim for lowest number of actual gameplay frames, it would be advantageous for me to omit all lag reduction strats, resulting in a run that lasts even longer in real time.
Again, thank you everyone for taking the time to read and respond to my questions.
Experienced Forum User, Published Author, Skilled player
(1221)
Joined: 8/29/2014
Posts: 301
Thank you for the responses.
These timings you did, DeHackEd, I believe were to measure the gameplay differences between a run on (J) version and a run on (U) version (so that text speed differences were not taken into account), correct?
Would there actually be an accurate way to compare game play in this manner of runs done on 2 separate emulators where the differences are differences in lag that affect the game play within the stages as well?
For example, a TAS on 1.43 is able to visit Bubble Crab's stage 4th in order to target the Counter Hunters on the stage select map sooner, use speed burner to pass through the giant fish's laser beam without taking damage, then ignore the fish for the rest of the stage since it does not cause a significant amount of lag. On 1.53, I have to visit Bubble Crab 7th or 8th (after Overdrive Ostrich), so that I can kill the giant fish with sonic slicer to reduce the lag. This causes more counter hunters to warp on an off the map during stage select, prevents me from being able to burn through the fish's beam, and I have to wait for its beam to disappear while it is getting killed by slicers before I can pass.
This probably the biggest example I can think of, but there are lots of small differences between 1.43 and 1.51 in which you must take extra measures in order to reduce lag on more accurate versions in the SNES Mega Man X games.
If you say differences in lag, not actual gameplay, from version to version are discounted when judging the TAS, isn't this a problem when different lag reduction strats bring about differences in gameplay?
