Posts for Derakon


Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Isn't it possible to have multiple video streams in the same container? Then when you play them back they'd get shown in separate windows or something like that. If you made each screen have its own separate video stream, then the user would be able to resize them as they see fit.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Weatherton wrote:
Derakon wrote:
"no breaking the game's intended max speeds"
Unfortunately, I think that would be impossible to define and overly restrictive.
Overly restrictive is one thing, but I think this is quite easy to define on an "I know it if I see it" basis. It doesn't have to be a rigorous definition.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
There was someone working on this awhile back, but I guess either they've stopped or they are deciding to keep quiet about it. I know they started before the HSWK was discovered, so the original rule was "no BLJ", but it'd make sense to adapt that to "no breaking the game's intended max speeds" as a more general restriction.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Um...that MKV I downloaded has massive sound sync problems (example: when Navi is introducing herself to Link, the sound from the Hyrule Town drawbridge nightmare is being played). The video also appears to be massively accelerated, though the sound plays at normal speed. The MP4 appears to be fine, though. Thanks for the encode, Aktan!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Well, good luck. I'd still recommend checking out the existing runs for movement techniques and the like.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I haven't watched your WIP, but you should be aware that we had a 1P run of Gunstar Heroes published before, but it got obsoleted by the 2P run. You should compare your efforts against that run to ensure that you're faster than it is.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Well, thanks for making it, Scepheo! And good luck on your next TAS.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
It's in the Gruefood as far as I can tell. Rockman L is still on the workbench though, but that's a different hack.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Anyone looking for consistency in enforcement of the rules is going to be disappointed. They aren't hard-and-fast rules; much like English grammar, there's always at least one exception. The rules are things that TASers should keep in mind if they want to have a better chance of their run being accepted on this site, but ultimately if one of the judges thinks you've made a good run, then it'll get accepted (assuming you submit it). If we tried to make the rules not have any exceptions, then either we'd end up rejecting good runs, or they'd be a gargantuan tangle of legalese and everyone would always be complaining about how complicated they are. It's much simpler to leave the rules as 90% accurate and then let the judges handle the grey area. That is, after all, why they're there: to judge. If the rules were perfectly accurate, then any run that met the rules would automatically get accepted and any that didn't would be rejected, and we'd have no need for judges. Usually the judges' job is to see if the TASer obeyed the "is it entertaining" rule (and, if not, whether or not to ignore that rule), but sometimes they have to judge on the applicability of other rules as well.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
How long does the Vile stage + Zero sabre cutscene take? It'd have to be pretty lengthy to make up for not being able to oneshot all the boss refights and Sigma's forms.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Give the encoders time. N64 encodes are always a pain to make, as I understand it, and this one's long to boot, which means that all of the processing takes longer.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Ehh, I've a feeling that blazing through all of the puzzles at lightspeed would be much more dull than blazing through just one puzzle at lightspeed.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I don't really have an opinion about the TAS, but I am a bit sad that so many people don't seem to know much about Myst. It's making me feel old, and that really shouldn't be happening to me for several decades yet! Not voting.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I'm inclined to agree with sgrunt's analysis. The bosses appear to have different weaknesses, but they have exactly the same patterns and give the same weapons. The levels themselves are fairly clever about tweaking the use of existing MM2 enemies and changing their sprites, but there aren't any actual new enemies. Plenty of sprites imported from other Megaman games though. Voting meh; seemed decently-played but the hack could be a lot better. The gameplay seemed solid to me (speaking as a non-expert in the world of Megaman TASes), but it also seemed pretty similar to the existing MM2 run. If you're looking for a Megaman hack to TAS, Rockman no Constancy seems like it could be a good choice.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
The Castlevania 3 runs also use the Japanese version because it has better music, though I don't know if there was ever a published English run.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Ah, my mistake. Carry on!
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
boct1584: I believe the reason this is a fringe case is that OoT runs are in very heavy demand and yet are also difficult to make; thus, we have to take the runs we can get instead of the runs we'd like to have. Ideally (for us) abeshi would have made his TAS to the site standards, but he was making it for nicovideo, not for us, so he didn't have any reason to follow our guidelines.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Hacks often aren't accepted. Have you cleared the hack with the site administrators? I watched the first two levels of your TAS on YouTube; you seem to have a good grasp of how to TAS, but the hack itself didn't impress me. The levels are filled with enemies but don't have very interesting design, and the bosses don't seem to have changed any beyond graphical tweaks.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Man, what the crap? How's that glitch work? I think your decision to complete the glitchless run is entirely reasonable, and there's plenty of precedent for completing runs without using glitches that are discovered in the course of making the run.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Tuppenfall: it depends on how friendly the game is to hex-editing. That in turn depends primarily on how the RNG runs. For example, if you have an RNG that advances once per frame, then any delays will cause a desync as soon as you reach your first luck manipulation after the inserted delay. However, if your RNG advances only when it is used, and buying shoes doesn't involve any randomness, then you might be able to insert the extra purchase without getting a desync. Other things can also cause desyncs -- as a totally made-up example, if the game's random encounter generator depended on how much gold you were carrying, then any inserted change which modified your gold count would change when the next random encounter occurs, which would assuredly desync you. But the RNG is the main factor.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
mz wrote:
Well, this was made by Konami: http://www.youtube.com/watch?v=yjdVflSHxN4. Compare it to this piece of crap or to some of the tracks from that shitty port of Dracula X. :P
Amusingly, your first link has a repeat of this discussion in the comments. I think the takeaway lesson here is that you can make good music on just about any platform. Hell, the NES is replete with awesome soundtracks. It's all a matter of understanding your tools and making good use of them. If you don't do that, then no amount of awesome technology is going to save your music from sucking. (Also, I rather liked several of the tracks from SCIV and Dracula X)
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
I guess you could sum up the two stances as such: one stance wants to compare individual still frames, and the other wants to compare the total visual experience. Is that accurate?
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Per the Rules, we prefer the use of the English version unless other versions have glitches or game features that aren't present in the English version. Faster text is not a good reason to switch versions. That said, given the demand for an up-to-date OoT TAS, the admins might be willing to bend the rules this time if the TAS is good...
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Are you sure the published data you referred to is measuring from time of input to time of hit, and not time of starting the animation to time of hit? Perhaps the game inherently has a two-frame delay between input and action. In that case the emulator would be behaving properly. I could see that being desirable to e.g. make it easier to perform actions that require the player to hit up without jumping.
Pyrel - an open-source rewrite of the Angband roguelike game in Python.
Experienced Forum User
Joined: 7/2/2007
Posts: 3960
Tub wrote:
Slowking wrote:
It seems obvious to get more rupees in the first cycle, where time doesn't count.
doesn't clock-time stop advancing during area transitions and dialogues and stuff? I liked the WIP :)
The point is that you have to use up 12 hours one way or another. Certainly the fastest way to do that would be to just heal the Great Fairy and then wait at the clock entrance, since that minimizes time spent without the in-game clock advancing. However, then you're just wasting a bunch of time standing around in south Clock Town. By using the extra time to collect rupees and get the Adult Wallet, you get those items for "free" (actually, they only cost the time spent in transitions/dialogs that you wouldn't otherwise make, which is very little compared to the time it'd cost to get them in later cycles).
Pyrel - an open-source rewrite of the Angband roguelike game in Python.