I have watched and enjoyed every single of the runs you linked, but didn't realize until now that all of those great runs were his.
a new usergroup called Ninja/Coder seems fitting.
JXQ wrote:
Also, his name is very ninja-like. :)
If that helps, I'm going to register a new account named Adminja or something :p
woah, this whole discussion turned into something.. well.. something not smooth.
So I'll just add one question.
Saturn wrote:
I'm glad you guys like my idea and the run so much. This gives me an enormous amount of motivation to work on it further, thank you all!
If kind words can motivate you to work faster, how much kind words would it take to have you finish the run in.. say.. one week? I'm prepared to type all night if needed :p
Xaphan: nope, the second visit was to refill my health, the blatant mistake costs a lot more time than that :P
going back into your house for 20 rupees?
I'm not sure if the fairy should be visited at all. bombs and hearts can be obtained by killing weak enemys.
No idea how much time you spent on the regular enemy battles, but they seemed improvable. There were also flaws in your per-screen-route planning. If you wish me to point them out I'll do, if you postponed the detail-optimization to the final run I'll spare me the time to type it. ;)
I don't see why the number of players that made the movie should make a difference about what the movie obsoletes. The number of characters, maybe, but how many players there were shouldn't matter.
seconded. IMHO publishing should depend on the movie itself, not on the process of making. The amount of time or work that went into creation, rerecord counts, tools used etc. has never been an issue, and I don't see why tool assistance or even bot assistance is fine while buddy assistance is not. ;)
I watched the 1-player run back when it was released, now I watched the 2-player run. It's certainly interesting, some glitches were used, and I liked the battle tactics using a killer and a runner.
No idea if it's improvable like Boco said though, or how the general execution compares to the single-player run.
I'll vote yes on publishing, because it was entertaining.
I'm not sure what I'd vote on obsoleting the 1-player-run, but I'm not asked about that anyway.
There are quite obvious situations where you waste time, you noticed at least some of them yourself. The movie is far from perfection and doesn't stand up to the quality of the runs on the SNES version.
second objection: IMHO it's a bad choice to run different versions of a previously TAS'ed game, unless you can show off new interesting routes or tricks, which you didn't.
Your route changes made the run slower, and even if it's because a shortcut isn't available on GBA, slower routes aren't really interesting for the viewer.
It's quite ok for a first attempt at a TAS, but nothing publishable yet. There's much to do in terms of route planning, and your run execution has to be more polished. Savestates are your friend, don't be afraid to use as many rerecords as you need to try different ways, then pick the fastest.
ahh.. I wasn't sure what to vote on guybrushs run. It was an improvement, but parts of it were sloppy.
now this one clearly deserves a yes vote, despite minor flaws (which I couldn't spot anyway). That is, if genisto agrees to releasing his 'unfinished' version. I'll wait to see if genisto posts here before voting.
Indeed, as evidenced by stuff like "Master my ass!"
You did NOT call the best mage to ever visit a video game after your rear body parts!
Meet me at dawn by the woods. You choose the weapons. No timescale or quicksave!
;)
I've had some weird moments with the german FF7 "translation" as well. Later I found out that I just had to translate everything literally to english, then I could understand them just fine.
no other funny moments except for the usual story mishaps. ff7 was full of them and I still don't understand why many consider it to be the best ff game yet.
Isn't an ISO just supposed to be an exact copy of the data in the CD?
yes, but you can choose to omit certain data, like empty tracks at the end of the disc, subchannel data and whatnot. Then there are a bunch of different iso formats. Try Wiki for a short and incomplete overview.
So, a CRC over the whole iso file won't work. A CRC over the CD's user data might, unless there's some cute copy protection generating read errors.
If they wouldn't accept save games from each other, different game versions might be the cause, either that or the emulator wasn't set up right to use the same saves for both games
the emulator was set up correctly, and all the isos worked fine on their own. But why should there be two incompatible english FF7 releases? Anyway, I didn't find any information about FF7 releases and the isos are deleted, so we'll never know ;)
this should be the smallest of our worries. It's more important to have consistent isos. I've been unable to rip my FF7-CDs or play from CD (I don't own a PSX, but after playing a copy of FF9 I bought the FF-games), so I had to download isos for the emulator.
I ended up with broken savegames because all the isos were from different sources, and wouldn't accept savegames from each other. No idea if they were different game versions (at least all of them were english) or if it was a problem with the iso. I managed to get another CD1 from the same source as CD2, replay CD1 from the beginning, then switch to CD2, now I'm stuck at the end of it (and in fact deleted all isos a few days ago)
anyway, I'm afraid crc checking won't help much here (rips from the same cd will end up with different crc values depending on ripping program, target format and other settings), and having equal game versions across players, verifiers and watchers isn't easy.
but then again, we also lack a working open source psx emu. epsxe seems to stay closed for a while, and there isn't much of an alternative yet. Galtor said in 2001 they'd "surely" release the sources when they stopped working on it, but right now they claim to be working on it without any information or releases. weird.
btw, the current epsxe rerecording doesn't match the rules of this site: the movie files aren't input data and can allow undetected cheating.
no.. ehm.. uh.. too obvious.
and walljumping won't help, as was demonstrated in world 1.. right.
Thank you for your reply JXQ, I didn't know that Racoon Mario was slower when flying. Just watched the run again, and knowing that, it looked better. :)
The category you choose isn't the most entertaining for me (because of deaths and damage taken), but you did an excellent performance on it. I'm glad it's already enqueued for processing as I couldn't even give you a yes vote.
Looks much better than your previous submission.
Seeing that you only used 171 rerecords I suspect you can still shave some frames off, but I'm not going to spend time with that game to prove it.
was this really optimized for speed?
I noticed that you killed some monsters while walking past others. Sometimes you walked back and forth instead of just going forwards. At least once you didn't take the fastest path through the menu. And it looked like you stabbed a lot of air in the run.
I don't know the game though, so you might have reasons for what you did. Can you please explain them?
damn, you'd waste 3 minutes of peoples lives!
remember, those are gamers. 3 more minutes of wasted time won't matter. ^^
I'd vote yes on the new version if I could.
edit: now that the movie has been accepted, I can finally vote yes. :p
that was entertaining to see some new tricks done on SMB3, even if they're done on a different game.
I'd vote yes. As I decided not to spam for voting rights, I might be able to vote in a couple of months.
Most of all, this TAS still needs a lot of research. It's unbelievable how much time has been cut off the deku tree alone.
IMHO when a TAS is done, it should aim for perfection, and not just for better execution of existing routes.
Well, if Acryte continues at that speed, he should be done sometime around 2027, unless he loses interest before that. Some help is needed, but it doesn't seem like anyone here is devoted enough to go ahead and fool around with it.
however, the person who programmed the video plugin should die a horrible death.
I'd like to see you program a better one and live.
as for the run, I watched the test run and the published run side-by-side. It seems like you gained some time until the castle door, then lost that time again while reaching the first course. (Everything after that is difficult to compare)
Long story short, I disagree with the star. The run is well done, but 3D games are a whole new playing field, and neither are the tools elaborate enough to produce perfect angled input, nor do we have other 3D TAS to compare it against.
anyway, pretty well done, I know you spent a lot of time on route-planning. But the tools don't seem ready yet for perfection.
you might also want to update the URLs in the Thread linked from the EmulatorHomepages-site.
I tried them today and noticed they were down, then discovered this thread.
I'm confident I can make it have a CRC32 of 0x42 as well.
oh no, another geek smartass :p
well then, md5-value of 0x42 for added challenge.
Nach wrote:
It would require doing something which might be frowned upon a bit...
recording from a save state, obviously.
using 2 different emulators and using the inevitable desync to split both players is another possibility.
I still think that 2 different games are more interesting than 2x the same game.
