Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
SNES without question. Harder to TAS than NES in most cases, but the movies often look more entertaining due to better graphics and more variety the complex games of SNES offer, making it worth the effort in the end.
NES is second though. :-)
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Yeah, I played CT alot in the past ~10 years and have a wide knowledge of this game when it comes to 100% runs. I'm glad you are interested in that run, inichi, and hope it will help you finding new strategies for your any% TAS as well. :-)
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Damn nice critical values you got there. No wonder why you could manipulate the important critical of Shiva Edge in the Lavos Core Battle during the New Game+ run that well. :-)
Also very impressive battle skips. Especially the one in Reptite's Lair will save alot of time. Very nice findings once again, inichi.
inichi wrote:
No. Berserker actually doesn't affect the Crit Rate at all, though the fact would still remain that it is the most useful accessory in a CT run because it allows an equipped character to make an instant and massive attack. It's one of Hero's great findings, I think.
I actually used the Berserker alot on boses in my 8-segmented unassisted 100% run, which was done a year before hero did his any% TAS here. I think he mentioned to get this idea with the Berserker from my run at first.
In case you wonder, this 100% CT run was my last major one I done on ZSNES unassisted before joining here and switching to Snes9x. I think it still has some useful boss strategies to show, at least for a 100% that finishes all sidequests, such as Robo's perfect Crisis Arm strategies that saved alot of time in the later boss fights for example. If anybody wants to see that run and learn something new in CT, get it and the needed emulator here (first run in the list).
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
You mean the bats in Red Brinstar before Spazer? No, they only drop Energy and Missiles often, and Supers only in very rare cases. Except the drainer, there are no enemies that drop PB's before Norfair to my knowledge. There are a few other enemies that do drop PBs, but like with Supers from the bats, only in very rare cases (Rio - the big yellow flyer from Brinstar for example).
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Looks pretty good. Nice new strategies in the gauntlet and during the Wave Beam gate before pink Brinstars E-Tank.
The only major mistake you did is not killing the 4 drainer all with one Missile shot in the small room right before the Etecoon E-Tank. You would get 4 very important PB's for only ~20 frames of delay. That way you would also save a ton of time using a PB for the bomb block barrier in the long pink Brinstar room connecting the mainstreet room and Spore Spawn's area. If you're not too far in Norfair yet, I would advice to redo the run from that small drainer room to get the 4 missing PB refills, as they would definitely save you a couple more seconds in Norfair as well. Though I know that redoing a pretty big part like this sucks, so if you are willing to accept losing some time and generally have more troubles in Norfair with PB's, then you could still try to continue.
Other than this, the quality of the run looks really good so far. Keep it up.
Kriole wrote:
It indeed was expected; if you check the jump before the pickup in frame advance, you can see me blindly going back and forth to get into the right position. I also considered "Torizo skip" to start after the Morph Ball Bomb sign had disappeared, so I was rather confused. (Since the torizo skip does not have a frame window of success)
Yeah, thats exactly what I meant before. We should try using more descriptive methods when mentioning frame differences in areas that aren't obvious to everybody.
EDIT:
I see now. Well, then you should definitely have enough energy to get this 4 PBs. They will save alot of time later.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Very nice work gotcha, thank you. The features, especially the memory watch adresses that now can be loaded like in 1.51 are very handy for every SNES TASer who can't run your memory watcher tool.
I have a question:
May I ask for a example of what "Movie play/record improvements" you did there? Changes in the movie engine often caused unexpected desync issues in the past (especially in ZSNES), so I just want to check whether such a possibility exists here. I hope not. :-)
The only missing feature I could think of is to save all Snes9x settings into a .cfg file (like in Snes9x v1.51 already) instead in the registery. That way you could for example have multiple copies of Snes9x with individual settings for different purposes each, and therefore aren't forced to change them every single time. Do you plan to add this feature in one of the next versions?
Other than that there is nothing else left to implent IMO. Good job!
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Wow, thank you Vandal! It indeed looks much smoother that way. I actually tried some interlacing methods long time ago, but I didn't know that you could merge 2 frames in one to create a pseudo 60 fps effect that looks that well. I will be sure to use this method for my next game video, no doubts.
moozooh:
I tried to convert the vids to 20 fps to maintain the blinking effect, but as you guessed right, it looks horrible jittering that way. Vandals method is much better to use, and I don't really care if the blinking isn't 100% accurate that way. Getting smooth movements in the vid is what counts most. Everything else is not that important.
pirate_sephiroth wrote:
Hey, youtube recommends Divx MPEG4.
Maybe they tested it only on 30 fps, in which DivX should work fine. However I'm pretty sure x264 is not a worse choice in any case here. They probably just didn't test x264 that well yet, since it's not as popular as DivX, but in every aspect superior to it when it comes to quality/filesize.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
It would be great if a solution for this exists. I actually tried some 2:3 fps overlapping during convertion, but didn't got a satisfying result back then.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
I tried to convert videos to 30 fps manually (and even resized them for YouTube standards, which didn't change anything either), but after comparing, the movements were almost twice as jittering compared to a directly uploaded 60 fps one. This applies to videos done through VirtualDub at least.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
It actually doesn't matter at all which codec or bitrate you use as long as the picture quality of the video doesn't suffer too much.
The only problem we have with YouTube when uploading game videos is the frames per second rate of the video (fps). YouTube is designed for 30 fps videos only, while most games have 60 fps, so after YouTube converts them back to 30, the video movements aren't that smooth anymore due to a halfed fps rate.
The reason why x264 works best is because YouTube can handle this format better than most other ones when converting it. There is still a unavoidable quality loss, but not as much as with a format like DivX.
Another option would be to convert videos manually to 30 fps before uploading them, but in most cases the quality turns out more worse than letting YouTube do the convertion job entirely.
If YouTube would someday support 60 fps, the problem with game videos would be solved. Until then we just have to accept the slight quality loss there.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
That's right. Surprisingly doing a turnaround is more than 10 realtime frames faster than the crouch exit (while same fast in ingame time). Though I have some doubts moozooh meant the difference in realtime frames when he always went for ingame time before. But yeah, this could be the only possible explanation for that statement.
Feel free to correct us, moozooh.
EDIT:
For those who wonder, we solved the misunderstanding:
IRC wrote:
[23:57] <moozooh> Saturn_: i said the skip itself was slower, not the escape alone. i just double-checked and i still don't see anything wrong. the starting reference point i chose was landing in torizo's room, since everything before that was too different and thus difficult to compare. the ending reference point was x=500 in the pre-torizo room on the way back. everything is correct.
[23:58] <Saturn_> so you gained 2 frames "before" collecting the bombs
[00:00] <moozooh> yes. the jump is part of the skip.
Moozooh saved 2 frames before actually grabbing the bombs, which was expected. I just didn't consider this as a "part of the skip" at first.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Ok, after some testing I noticed a wrong statement by me that I want to clear up now to avoid confusion later:
Saturn wrote:
The reason Kriole's exit is slower is not because of the turn around skip, but because he jumped off from the ledge in the next room much earlier instead of running through it till the end and falling with more momentum. He would end up same fast as Cpadolf's TAS if he would do just that. Still it's only a tiny 1-2 frames loss, so it's not a big deal.
This statement is actually not true. I just tested this sequence carefully, and in fact, Kriole did the entire Torizo skip same fast like Cpadolf and everybody else. Although he jumps off the ledge much earlier and with less momentum, he gains the loss back by landing on the main ground much sooner and thus can run at full speed for a longer time. Just like using turnaround or not to exit Torizos room, falling or jumping off the ledge in the next room is same fast in the end as well.
The reason I first thought Kriole's strategy was slower is because of this wrong statement which I at first believed blindly:
moozooh wrote:
Well, that Torizo skip was 2 frames slower than the one I did in NBMB WIP...
I'm pretty sure moozooh either confused something, or just didn't mention that the gain is either somewhere before getting the bombs, or after them somewhere in the mainstreet room, but definitely not in the area between the Bomb collection and the exiting of the long room with the bats right after it.
So please moozooh, if you make statements like that, try to at least make them truthful. You really confused me and some people unnecessarily with that.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
I vote for Super Mario Kart.
Although qualitative improveable, what Huffers showed in every single course was extremely entertaining and most of all unexpected, even for a pretty experienced SMK player like me.
Both Metroid runs come as a close second though.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
I wasn't much into NES TASes lately, but from what I have seen it will be Darkwing Duck for me. This TAS was simply amazing, with both, excellent move optimization and incredible luck manipulation to get the Gas refills without slowdowns. The result was a huge improvement I never expected to be possible to the previous run.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Very nice analyses, Kriole. When TASing Maridia during the RBO, I found out that 7 and 8 frame UWWJ's work best with HJ-Boots, but I never made such tests to fully confirm that. It's good to see that my numbers were correct. :-)
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
A few small errors I noticed:
In the big green Brinstar room with the Gate, it's faster to stay more away from the gate when it opens, and then run through it with much more momentum. The jump under it allows you to pass it a bit faster due to a reduced hitbox, but when you land you have to gain all momentum from scratch again, which is at least 10 frames slower than the strategy I mentioned above.
When opening the green door to the sidehopper room, you can save at least 2 frames by shooting the door on the first possible frame while aiming up instead of the usual horizontal shot that requires you to wait a few more frames until you reach the needed height to shoot the door. JXQ did the same mistake in both of his TASes there too. See my 2nd RBO teaser for a example on how to do it correctly.
And I strongly recommend to redo the sidehopper room using the strategy Cpadolf showed. I think I even did this room a couple of frames faster in my RBO run, but it required the Wave/Spazer Beam which you don't have yet.
Otherwise it looks pretty good so far. Keep going.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
A hard decision, but I have to go with hero, simply because he not only made a stunning any% Super Metroid TAS (which had a very complicated and hard route that required a lot of luck manipulation) on the very first attempt already, but also because he discovered plenty of new and innovative tricks (although some of them were known, but not public) and routes for this game in that year.
Experienced Forum User, Published Author, Player
(209)
Joined: 2/18/2005
Posts: 1451
Kriole wrote:
I figured if you exit the room in it's lowest possible position and sitting facing to the left, you'd avoid unnecessary airtime in the next room.
No. As hero suggested, just check Cpadolf's submission here. You can jump exactly same low like you did and still manage to exit with turn around by using hero's release trick.
And yeah, 49'05 is a great time for the first serious try. You already pwned every existing TAS here with that time. :-)
Hero:
The reason Kriole's exit is slower is not because of the turn around skip, but because he jumped off from the ledge in the next room much earlier instead of running through it till the end and falling with more momentum. He would end up same fast as Cpadolf's TAS if he would do just that. Still it's only a tiny 1-2 frames loss, so it's not a big deal, especially not in a entertainment run which is Kriole's speciality.