JXQ's run was very tightly optimized, I wouldn't expect to find too many "loose" frames floating around.
Though I found 20 or so frames just in the escape which I took as one of his more optimized parts. Still Any big improvements without different room strategies is not easy.
don't forget that you'll get around 25 seconds just by skipping the torizo, so you'll need only ~30 seconds of gameplay improvements.
Torizo skip saves only 18 seconds, unfortunately.
hero of the day wrote:
JXQ's run was very tightly optimized, I wouldn't expect to find too many "loose" frames floating around. I do think the 36 is possible, but it will no doubt be very close.
Actually, JXQ's run is pretty full of exactly that: loose frames. Remember how much was saved in pre-Torizo segment alone, and that's considering how easy to optimize it generally is, compared to the sloped/rugged areas and such.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Though I got it to something like 21 seconds IIRC. I measured how long it took for JXQ to get through the pre-torizo room the torizo and back and how long it took for me, I think the difference was something like 21.7 seconds. Taking in to account the loss in the diagonal room it is still over 21 seconds. Or else I'm missing something.
moozooh wrote:
Actually, JXQ's run is pretty full of exactly that: loose frames.
Yes I noted you changed your technical rating on his movie from 8 to 7 some time ago ;)
sorry, you're right. Yesterday I did a rough estimation using the avi and my watch and got around 25 seconds.
time spent in the torizo-room itself is this:
JXQ 100% v2: 24374 - 26212 = 1838 frames
cpadolf v4: 24301 - 24844 = 543 frames
~21.5 seconds saved (small differences in the pre-torizo-room not accounted for)
also, despite being ingame-oriented, 73 realtime-frames were saved in zebes until the torizo-room by minor optimisations and CWJs alone.
Well, don't forget that JXQ's performance in pre-Torizo room and the fight itself weren't optimal. It still saves no more than 20-20.5 seconds overall, which I agree is still more than 18 seconds, though (might have remembered that figure wrong).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
it doesn't matter if JXQs torizo fight was good or not, since we're calculating improvements over JXQs run.
of course you might label those 21 seconds as "1 second: better torizo fight. 20 seconds: torizo skip", if you really want to. ;)
Joined: 11/14/2005
Posts: 1057
Location: United States
To be fair, much of the pre torizo improvement was the result of "moozooh jump", ombr room, "saturn shaft". Other than those things, we are talking about 1 to 2 frames per room of brute forced optimizations. I consider the later parts of his run to be even more optimized, mainly because of the powerful items Samus had acquired make room strategies more simple. I could be wrong though, I never meticulously combed through his entire run. I just know that if someone were to spend so much time trying to find ever single last frame in every room, a 100% run would take a few years to finish. Let's see how the new run turns out :)
Still, as I said I found about 20 frames in the escape alone, none of which really came from hard optimization, just ~10 came from improving simple jumps by a frame each.
And I also found some relatively easy room strategies he missed while doing the any%. Like shooting the door early in the ridley E-tank room and not middle landing on the door platform in the acid dip room that came right after but instead jumping straight through the door by releasing dash for a few frames.
Anyway We'll see how the rest turns out once the run is started.
Boy that Super Metroid run JXQ did a year ago sure does suck, right? Missed frames everywhere, all these easy™ strategies ignored, maybe even on purpose(!), a far cry from perfection no doubt.
I guess this is the price I pay for not restarting my run every time someone else figures out a new trick. Personally, I hope you improve the hell out of my run. I know I'm never getting to it; I'm sick of this game and I'm sick of the community around it.
Try to enjoy the attention it will bring; I sure didn't.
<Swordless> Go hug a tree, you vegetarian (I bet you really are one)
Joined: 11/14/2005
Posts: 1057
Location: United States
Don't take it too personally man. People just feel good when they can find improvements in a run made by a respected TASer such as yourself. Most of the people on this thread are not assholes, but some can come across that way when trying to achieve 'perfection'. I am not a perfectionist by any means, so I cannot relate to any of these tiny frame battles. Hell I know the run I am about to submit is far from optimized :P I don't mind if people pick it apart.
Boy that Super Metroid run JXQ did a year ago sure does suck, right? Missed frames everywhere, all these easy™ strategies ignored, maybe even on purpose(!), a far cry from perfection no doubt.
I sure hope you didn't take it that way... It's still one of the greatest runs on the site.
I just don't hope that I have to say that things I did not find that hard to be hard just to make you feel better.
Also I naturally would not call anything I have done or will do to be perfect since I don't think it is or will ever be.
Has anyone tried getting into walls like you can in metroid? I don't think the doors ever close so you can't do it that way, but in metroid you can also do it if you freeze 2 or 3 enemies in the right place under a ceiling. You then use some morphing tricks to climb through them an into the ceiling. You can also use the blocks that reappear after you shot them away.
Does anyone know if similar morphing tricks can be done in super metroid, like morph then unmorph to climb 1 half block, or add a jump at the the right time and you climb a full block. Without tricks like this you can't climb inside a wall.
Then is there anyway to cheat and test to see what potential shortcuts there would be if you could find a way to get through a wall? A map would be nice, but that doesn't tell you everything, because warping through a wall to get somewhere can take you some place unexpected using 2D logic.
"Great spirits have always found violent opposition from mediocrities. The latter cannot understand it when a man does not thoughtlessly submit to hereditary prejudices but honestly and courageously uses his intelligence."
You can get stuck in the walls in Super Metroid (which is actually how you go about accomplishing the NBMB run), but it isn't beneficial from a timesaving standpoint at any time, and certainly not from an entertainment standpoint.
but it isn't beneficial from a timesaving standpoint
Actually X-ray climbing makes a 00:15 not sound very far fetched.
I think he ment it as an everyday timesaving trick. Not as a gamebreaker.
EDIT:
Omg. I just improved torizo skip by like 80 frames or so. It is not even optimized and I fear it can be improved more. Check it out here.
All the best.
Another EDIT:
Apparently the trick is to pick them up at Y position 189 65535, so this means it's improvable by alot more. Expect it to become seriously awesome.
Hah, that's truly awesome, my TAS took exactly two days to get obsoleted :)
Also it actually makes very much sense, I wonder why no one has though of it before.
Hah, that's truly awesome, my TAS took exactly two days to get obsoleted :)
Wrong. Everyones Crateria segment was heavily obsoleted by me and Taco. Nothing else. Your run is still very high quality to be done over a period of 3 weeks. That is what surprised me the most o.O
Well at this point I think that lacking a trick that saves ~1.5 seconds is enough to consider a run at least somewhat outdated.
I'm fiddling around some in the room to try and make it even faster, so far I have been able to bring the jump down to holding for 5 frames but for some reason I don't get enough momentum at the pickup to put me far enough away from the torizo when landing thus lacking 1 pixel to make it through the door.
I'm fiddling around some in the room to try and make it even faster, so far I have been able to bring the jump down to holding for 5 frames but for some reason I don't get enough momentum at the pickup to put me far enough away from the torizo when landing thus lacking 1 pixel to make it through the door.
I have been doing that. Anyway I got it but only with a 9 frame jump so far. Surprisingly though it only saved 36 frames. (although it is maybe not directly comparable, I used my any% run and started from my entry there, and that entry might not be optimal if doing the room this way)
what's this NBMB run?
X-ray climbing? is that on the tricks page, where is the tricks page, I can't find it?
"Great spirits have always found violent opposition from mediocrities. The latter cannot understand it when a man does not thoughtlessly submit to hereditary prejudices but honestly and courageously uses his intelligence."
but it isn't beneficial from a timesaving standpoint
Actually X-ray climbing makes a 00:15 not sound very far fetched.
I think he ment it as an everyday timesaving trick. Not as a gamebreaker.
Right. As a gamebreaker, the general consensus is that it makes the run more boring than watching paint dry, but it does allow for impressively low completion times.
Also I think it does bring some entertainment from the level of glitching, the breaking in to tourian and skipping motherbrain by abusing a save station practically screams "stop raping me".
