Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
amaurea wrote:
Here is an, unfortunately soundless, comparision of cpadolf and herooftheday's last any% runs. herooftheday's are orange for the old and yellow for the new, and cpadolf's are blue for the old and white for the new. This movie compares ingame time, the result would be different in realtime mode.
http://folk.uio.no/sigurdkn/sm_ingame.avi
The hitboxes indicating the other runs were generated by my lua script for super metroid, which is found in the snes9x lua thread.
Wow, thats a very cool idea from what it looks. Though for some reason the download gets cancelled after around 50% on my side despite multiple tries. Is there a way to upload it to a stable host like Megaupload or Rapidshare, so that everybody could DL it without cancels?
Or better yet, can you post a detailed guide on what to do in order to playback multiple smv with the combination of the needed lua script to produce something as cool as this? I would be interested to compare a few other runs that way as well.
EDIT:
Oops, I somehow missed the sentence about the Snes9x lua thread before. Let's see if this guide will work.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Cpadolf wrote:
I think the big problem here (although I personally would like to keep the 14% run too) is that 14% now is neither lowest, nor faster than the glitched lowest, and the glitched lowest is not so boring that you could say 14% is objectively more entertaining. You already get entertainment from the any% runs and the 100% run, and it would simply be to arbitrary to allow a low% run that is not the lowest% possible. Because the way I see it, low% means to complete the game with the least amount of (in this case) items using all means.
My point was that this run is a "glitched" low% / NBMB one. Yes, it achieves everything as fast as possible (ingame time, realtime, lowest % etc.) but it's not a legit, clean run like all the others. The 14% is still the lowest classic / legit low%. As of entertainment, it's of course a matter of opinion, but because of the extremely long and boring X-Ray climb, I think that a 14% is still somewhat more entertaining to watch overall, because it just shows more action in the end result. And as said, many tricks only applicable in a 14% will be missed if we erase it.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Awesome and totally unexpected run from you two!
The optimization level is just ground breaking in this one, the ammo and health management simply stunning (especially in the X-Ray room), and the run shows alot of new room strategies no other run would ever have the posibility to present. Seeing that you can even skip the Speed Booster really surprised me, as I didn't expect you could have enough momentum to still manage to get stuck in that statue door deep enough. Awesome find, hero!
As of the highlights of that run, to me it's definitely the Big Metroid Skip. I did some tests skipping it without Speed Booster around a half year ago, and you have no idea how hard it is to manipulate his move patterns in a way that allows you to reach the exit as fast as possible. I have to recognize you guys did this part so much faster than I did in my test, it's almost unbelievable. Much respect on this, man. X-Ray, although way easier to optimize, was also stunning to see.
As of obsoleting the 14%, I'm not sure whether this is a good decision (the any%'s are out of question anyway, being the most entertaining run forms of this game). This run is a heavily glitched NBMB one, which has a completely different category than the glitchless ones, to which the legit low% / 14% counts as well. One thing is for sure, the 14% (no matter whether Speed or Ice route) will show alot of new speed tricks and entertaining stuff no other run would allow you to present (I did alot of tests for this run some time ago), so if we erase it completely from our list here, the watchers won't have the pleasure to see this new and unique tricks. I think it would be best to give this run it's own "glitched" category, as it just can't be compared to any other run here. Just something to consider for the judges.
Anyway, I am highly impressed of this run and vote a big YES for it. Extremely good job Cpadolf and Hero!
Very nice find. Since the momentum is kept when you take damage while being in morphball, it was actually to expect that there would be a way to transfer it even to a standing position if the timing for unmorphing is correctly done. I think I might have even executed this long long time ago by accident, but didn't realize at first since it's hard to see your momentum without the memory watcher, which I don't use. Good job in any case, this will sure be of use in a few rooms.
NameSpoofer wrote:
You got it. ;-)
Video demonstration for those who can't playback smv files:
WS Gravity Spike Tech
Saves a good half second compared to the fastest techs known there so far (or 13 frames faster than Taco's demo / 25 frames faster than moozooh's).
Once again good job guys for finding it yourself.
moozooh wrote:
That's known, I showed it to Cpadolf and hero some time ago. I did it slightly differently, though.
Mind telling us the true date you discovered this?
Cpadolf wrote:
Found some semi interesting things while testing the NBMB run.
A new way to acquire bluesuit, very useless, and very pleasing. And a dead end glitch, it seems to work only if the statue has already sunk, and it occurs if you try to trigger the elevator doing anything but a normal jump (bombjumping might get you through too, I dunno).
Very interesting finds, Cpadolf. Too bad we can't get a use of them in a speedrun.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
NameSpoofer wrote:
Decided on a completely new route after I realized this cold truth.. re-recording from the norfair elevator, and had a crazy idea that involves some Kejardon-type shit.. Taco thinks it won't work tho.. but yarr.. still worth a shot!
Well, since I already tested everything there is to test route and strategy wise for any run, I'm pretty sure I could tell you right away whether what you try is possible or not. Though if you plan it to be a surprise in case it works, and therefore don't want to spoil, I can perfectly understand it.
Cpadolf wrote:
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
NameSpoofer wrote:
Now who's got some easy way of sparking these fucking pirates? Saturn? =P
If you plan to kill both Pirates with one speed-echo, then you will have to sacrifice "alot" of time in any case to get the Pirates in the right position. Because of this, I wouldn't recommend that. The more rational strategy is to kill the first pirate with 3 Supers, and the last one with a spark, just like I did in my 3rd RBO demo, or slightly different like hero in his any%v2. I'm pretty sure that refilling the additional 3 Supers would be faster in the pre-Ridley room anyway, if needed.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Yeah hero, although I knew about this trick for a longer time already, I too thought it would be slower at first. Only the recent discussion here made me start to doubt, so I went ahead and tested it just now. I certainly didn't expect it to save so much time before. Sorry if this comes a bit too late.
DK64_MASTER wrote:
Saturn, why don't you like moozooh?
Just mentioning the obvious for those who wait for his run without realizing the truth about it.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
The Gravity Statue Skip trick is almost a half minute faster than the usual way. Although it requires early X-Ray, additional ammo for the CF and therefore some sacrifices to refill them here and there, it's definitely faster to use it in a 100% run in the end (both, in-game and especially realtime wise).
Here is a video demonstration of this trick used optimally:
http://www.youtube.com/watch?v=l39lHJvueZ0
Enjoy!
Oh and NameSpoofer, welcome to TASvideos. I'm glad you decided to work on a low% run, since moozooh won't finish his one in the next 5 years if he continues it in this pace. You do fast and promising progress already, and I'm looking forward for your run. Good luck!
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Oh my god inichi, you are the greatest.
The glitches you find are just sick and far beyond my understanding (at least on the first look). I really wonder how one can find the correct position to appear in Lavos mouth while being completely blindly and apparently without any clues at all while just running around in the end of time room. The way you break such a popular game like CT and do with it whatever you want is something I haven't seen in any game of such a level before. Just crazy, man.
Now it seems that there are indeed chances to break the 2 hours barrier when using all the skip-glitches and therefore completely getting rid of any forced leveling ups. Looking forward for your final run.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
My classic 100% run is now available on YouTube. Slightly modificated special edition at the end of the run. :-)
Check it here:
http://www.youtube.com/view_play_list?p=BD09D57D25B503F3
Also looking forward for more progress, evilchen.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Wow, this was a awesome run Cpadolf! I'm really glad we now have a complete TAS of SMI done by one of the best SM-TASer here. Thank you for having the patience to finish this pretty hard and challenging project. I'm sure you got even more SM-TAS experience of it due to the many difficult TAS conditions you had to face (just like I did with my RBO), which will for sure be of use in any future runs you plan to work on.
The run was very entertaining to watch due to the many new tricks you would never see in a original SM run, as Tub mentioned. I especially liked the creative use of the Spring Ball and the Grapple Beam to cleverly open doors much faster or when exiting the long grapple room after Draygon, and the hells run to Speed Booster. The route used in this run was also very optimized, and the ammo and health management once again extremely well handled, which just emphasizes your excellent item drop refilling skills one more time. The boss fights were also very good, I especially liked the Draygon fight. The probably hardest boss of the hack got pwned so badly that it almost looked unrealistic at first. Once again thanks to the TONS of Missile refills you got there, that allowed such a quick kill.
The only mistake I noticed is the shinespark in OTES during the escape. You lose a few frames by sparking through the last platform like you did (due to the longer spark distance) instead of crashing into it, since the time you need to open and enter the door is pretty same in both cases.
As for improvements, even with the now possible Torizo skip (thank you Taco), I can't see a 0:47 being reachable so far. I think even a 0:48 wouldn't be a easy task, since your run is really well optimized already. One thing is a bit funny though, how can one be so lazy to cancel the run on the first possible frame instead of somewhere in the credits? ;-)
Even though this run likely wouldn't be accepted here, despite that it's a pretty popular hack and that there is a concept demo section that would fit for such a run perfectly, you could still try to submit it to see what feedback it gets from people. I'm pretty sure that entertainment wise it's one of the best TAS runs we have so far, and that many people would enjoy watching it. In any case, this is quite a piece of work (more than 200000 rerecords speak for themself), and I will definitely go ahead and encode a high quality AVI of this run and post it somewhere for those who can't watch smv files.
Great job, Cpadolf!
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
hero of the day wrote:
Fair enough. Though if you are simply copying your old run and adding a few new techs, shouldn't you be able to finish the entire run in a short period of time?
If I would have the needed freetime and motivation, I could finish one in a month or so. But as of now, I lack both of them (especially the former). Besides I promised to work on SMR again since it was on hold for almost a whole year by now. And cancelling it again for another run would be a bad choice, I think. It's also more interesting for me to TAS, since there is alot of new stuff I have to test.
Tonski wrote:
I wonder how can you claim to beat almost all of the frame counts of Kriole and Taco with no smv proof.
Superjupi summed it up well, nothing more to add here.
Kriole wrote:
Yaaaaay!
Brinstar improved by 16 frames.
- 1 frame at the early super missile room door
- 2 frames for an improved pipebug spike before the charge beam
- 2 frames for an improved charge beam pickup
- 11 frames for an improved green brinstar pipebug room; 9 for the mockball, 2 for the door transition.
Smart guys, without doubt. :-)
The mockball before the gate in the big green Brinstar room was pretty obvious though (to me at least), since you can keep all the momentum before the jump on the small platform that way, pretty much like I did in my 3rd RBO demo. I realize that there are some limitations to shoot the gate due to the lacking wave beam though, which eventually could slow you down slightly.
Tub wrote:
Kejardon wrote:
We're just playing around with their engine, we're not actually creating anything ourselves.
I disagree. Would you also say that Beethoven never created anything, as he was just playing around with notes?
Any kind of creation, artistically or otherwise, is based on previous works. Even just combining previous works in a clever way can be a noteworthy achievement. That's what enables a society to evolve.
Very well said. It's of course not comparable to Beethoven in terms of achievements and the amount of work put into it, but the general principle is the same in both cases.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Kriole wrote:
Haha, you are so cute, you just make me wanna dropkick you :D
Oh, what a pathetic one you are. Seems like it's hard for you to face the facts without flaming somebody for no reason, especially knowing that it's Taco's work from which you have benefitted so much in first place after joining him. Take it easy.
hero of the day wrote:
Saturn,
So you plan to just sit back and wait for them to finish their run, then copy 99% of their input for your's? That sounds familiar to me... kinda like the way you talk to moozooh.
No, I plan to just copy 90% of my old any% run into the new one, but with the newest techs added in it in a few rooms. It will already be enough to obsolete any kind of run, don't worry.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Cpadolf wrote:
Well what about having a 30 frames faster torizo fight compared to JXQ, you where holding that for a long time. (released it first when moozooh prefomed the torizo skip) Or having a ~50 frames faster phantoon fight than JXQ.
Yeah, I didn't provide a smv in the cases you mentioned, but to repeat, my main point was that "all" my claims are based on real achievements (which was proven with the Torizo-fight video I posted later), not on mere assumptions or thoughts. It really shouldn't be that hard to understand.
Kriole wrote:
I sincerely urge him not to do so, the Powerrangers have been secretly working on an any% which we decided to make official so that people don't waste their time just to be obsoleted.
That's cool. Glad to hear that, since competition never hurts. :-)
I understand that you wouldn't like me to work on a any% as well (who wants to see his run being obsoleted after all), but don't worry, it would take a good while for me to finish my version anyway (being busy with SMR and other things), so there is enough time for you to finish and present it until then. Good luck with it guys.
Comicalflop wrote:
There is not a single SM TASer who has benefited from his arrogance and bullying.
Incredible how some people still don't get it, despite that the facts about how much I contributed to improve the current SM-TAS standard here were repeated over and over again. If that's not trolling, then the memory seems to be pretty weak.
gstick wrote:
Wow, very impressive one, gstick. You certainly show to have alot of knowledge of SM-TASing already, the more surprising for me that we have never seen you around here before. Keep it up.
Taco wrote:
Nice find, Taco. But as Cpadolf said, the refills in that room are very important and almost unavoidable, so it will be pretty hard to implent in almost any kind of run. But maybe it's still worth to sacrifice some time to refill this items somewhere else but use the mockball here. In any case, it will be pretty close in the end, so it needs to be tested carefully to know for sure.
Everybody knows that the only claim I did on something that is possible without showing a smv was my Ceres escape time before it got matched by Taco. Otherwise I have always provided a smv for them. And just to repeat what was said many times before, all my claims (no matter whether a smv is provided or not) are based on real achievements. I'm not that kind of guy who relys on simple thoughts, despite being right in 99% of cases in them anyway. It's just not enough.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Larkin:
I don't deny moozooh has good TAS skills (almost as good as his talking skills), but the past has also showed that he often made surprisingly clueless claims that were proved wrong later. And that's also the reason he doesn't provide a smv-proof like everybody else. There is a huge difference between talk and real achievements, and that's something he really should realize someday.
hero of the day wrote:
Twelvepack wrote:
doesn't it have several obsoleted room strategies? What is special about an un-published, obsoleted run?
Yes, it is very out of date by today's standards. In fact, his new any% demo that got leaked a few weeks ago is also out of date. There have been countless new room strategies since he started his original run a few years ago. I would be hard pressed to call it special, but maybe he has a few things in the run which have yet to be discovered. I cannot even think of a reason for him to post the original run, since he started a new one anyway.
My old any% is still over the current standard in terms of quality overall, so I wouldn't call it "very" outdated. With "special" I actually meant the new run I plan to make soon. You will know why when it's finished.
hero of the day wrote:
I am also a bit disappointed to hear that Saturn lied to moozooh about something as petty as a microscopic room strategy.
As said before, I didn't want to disappoint him finding out that his time was slower in that room. And since he didn't know about the mockball strategy by the time he posted his latest WIP (as he said), there would be no way for him to match that time without it. So after he asked me for my frame counts, I was basicly forced to lie for good sake. Hope this clears things up a bit.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
moozooh wrote:
Indeed, optimizing three mockballs after managing a Torizo skip would just break my will and make me hate the game for life.
Well, if the mockball bombs would be so obvious for you to find, then it's really strange why you didn't use them in your latest WIP, especially when you say that you already had the idea to use them there. Seems like it wasn't the case after all.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Regarding the Dachora room, the reason I avoided to tell moozooh my true frame-achievements in it was simply because I didn't want to discourage him finding out that his version was a couple frames slower than mine. And since my time in that room was only achievable by using this complex, very subpixel optimized mockballs, it would be very unlikely he could manage to execute them correctly on his own, and therefore failing to match that time. I just didn't want him to get stuck in this room with a unsatisfied feeling to not be able to match my time. And spoiling that strategy would be a bad idea as well, since it would surely be a part of the surprise for the final run. To that time I just couldn't imagine that a SM-TAS machine like Taco could exist, who finds this sort of tricks. If I knew that back then, then I would probably show moozooh my mockball strategy anyway, so that he could simply copy it into his run. But as you see, this would be pretty boring and yet again kill all the uniqueness of his run in that room, and I so like to see own and individual work, no matter whether it's slower or not (Tonski's SMI TAS is a great example).
okaygo wrote:
I might release a WIP, but I'd rather you figure this one out for yourself.
Yeah, that's for the better. No need to overfeed the copycats who have nothing else to do than to wait exactly for that over the past year. Figuring stuff out on his own is the only way to gain knowledge, which is required in order to ever finish a full SM run that meets the current standard here.
Larkin:
Yeah, it will be special. Don't be misled by other people who try to tell you the otherwise without ever seeing my run.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Glad to hear that.
And I see you still use the good old german names for the bosses. For those who wonder, "Vulgor" is the german name for the giant bug boss "Thraxx". I too had my problems to get used to them when making TAS progress reports back then. ;-)
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Yeah, I obviously had to add a few frames to my real accomplishments in this rooms to not spoil the tricks I used in them before the final run. Don't worry, you will get your proof when I post that run someday, and you will also know that the run is legit because you will see the exact same moves and refill drops like at Botwoon (as seen from the various speed demos on my YouTube channel), which would be impossible to duplicate again.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
evilchen wrote:
btw you knew the skip i do with the dog bug? it saves much time because i can skip giving the dog a name and skip the whole 1st dialog with fireeye + her first alchemist spell(should save muuuuuch time i guess)
The dog skip is well known of course. As said before, I just wanted to make a fully legit run that doesn't use this kind of skips, since it would badly destroy the game-story, which I wanted to fully show (since it was a 100% run) in any case.
Btw, do you already have a impression on how well the Snes9x 1.51 version works against desyncs in this game? Was is absolutely stable in your run so far, or are there still a few desyncs happening from time to time? And if so, did you notice in which locations they appear mostly?
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Kriole wrote:
Time to WIP our RBO =).
Weeee!
Compared to Cpadolf, we are 177 frames ahead (the comparison is fair since no difference has yet occured, Brinstar being 59 frames faster).
Incredible run guys. I checked your run more carefully, and it's unbelievable how well optimized it is. To be honest, this is the very first run that is even slightly faster in a few rooms compared to my old any% so far, and there are also many rooms that are exactly same fast, which yet again shows that they must be frame perfect since we never compared our room strategies before, yet achieved the same frame counts in them.
And as everybody can see now, there were no further Zebes-improvements from the landing room till the pre-Torizo entrance except of the mentioned 2 falling rooms due to the new falling tech, and the brand new door entrance trick when entering the pre-Torizo room. Just as it was expected.
As for the route, the only questionable thing so far is that you collected the Charge Beam and it's Missile pack early. It will sure be a bit faster to get them later when you already have PBs to save some frames when destroying the bomb block above the Charge Beam. But maybe early Charge will save you a couple of important Missiles while killing some enemies on the way like you did with the cactuses, so strategically viewed, it might be a not so bad decision after all. Like Cpadolf, I would also be interested to know what exact route you guys are planning to use though. Maybe we can even suggest some more route-improvements in it that way. If you plan it to be a suprise though, then it's surely better to not share it to avoid spoiling.
In any case, this run is totally awesome and I'm looking forward for more progress. Keep it up.
moozooh wrote:
Hah, and Saturn didn't believe me when I said that using mockball when planting bombs could speed up the Dachora room. :D
How could I not believe it when I actually used it in my old any% as well? :-)
It was a pretty obvious tech to be honest, I just avoided to talk about it at first, because I was curious whether anybody would figure the benefits of it out by themself. If we look back at what happened so far, it wasn't that surprising that it would be the Powerrangers to perform this tech in a benefitting way besides me. In fact, they even improved the bombing part by one frame, as they managed to lay 3 mockballed bombs in a way that it saves 1 frame each, where I only could mockball bomb on the first 2 blocks, despite trying hard on the 3rd (I just couldn't find the needed subpixel for it). :-(
hero of the day wrote:
Is the RBO run going to be in-game oriented? The trick used to kill the cactus earlier causes ~30 frame realtime delay due to the door scrolling. The trick only saves 4 in-game frames.
Nope, it actually saves only 2 frames. :-)
Although I'm often for in-game time as well, losing whole 30 realtime frames for them seems to be too much of a exchange. I too would rather sacrifice this 2 frames in that case, even if I would do a in-game oriented run.
Cpadolf wrote:
I don't think they will use pause abusing though, that's just stupid.
I'm also strongly against any kind of pause abusing. There is absolutely no entertainment at all watching a run that has 20 min of continually pausing and unpausing sequences in it.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Seems like we have another promising SM-TAS talent here. Very nice finds gstick!
Also nice find to keep full momentum during the Torizo exit, Taco. The only way to do that was yet again to combine the turn around door trigger with the aim down animation. The SM engine acts really strange in some situations...