Posts for moozooh

1 2
150 151 152
229 230
Posted: 8/4/2007 8:36 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
ShadowWraith wrote:
Amiga games emulate just fine. Unfortunately, the best Amiga emulator right now (WinUAE) has no re-recording/frame advance features. Only savestates.
No, that wasn't what I was talking about. Here's the thread.
Posted: 8/4/2007 8:32 PM
Post subject: Re: Request: Super Frog Speed Run
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Neillithan wrote:
Someone should do a speed run of Superfrog or other Amiga related games.
There are some difficulties in emulating Amiga games; they were discussed a couple months ago but I don't remember the exact thread. Right now it's pretty much impossible.
Posted: 8/4/2007 5:06 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I think Xkeeper's point was more about how you dismiss hacks altogether, and not just ignore them or count them as plain another games you've never played. In other words, if the game is unfamiliar to you, you usually ignore it. If the game is a hack, you're openly negative about it.
Posted: 8/3/2007 11:52 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
What if you redo it on Snes9x 1.51? (I haven't watched it yet, btw.)
Posted: 8/3/2007 9:01 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Saturn wrote:
You did a slight mistake in your test though. In order to get the Lava Missile you have to morphball into the hole. You can't do it without, since it's too low placed. When you get it, you have to unmorph in the lava again (otherwise rolling out of it will be even slower), and since the morph/unmorph animation costs some time, doing so in lava will take you some more damage than what you did.
Well, that's not fatal. Lava damages at the rate of 3 energy units per 4 frames, heat damages at 1 unit per 4 frames. So, according to this smv, morphing will amount to staggering 5 units of additional damage…
Posted: 8/3/2007 5:28 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Sub-37 100% run, go! :D
Posted: 8/2/2007 7:01 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Energy tank to get Speed Booster varialess, which makes it 8% in total.
Posted: 8/1/2007 10:07 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I once again repeat that it isn't about ingame time or loading or anything else like that, it's about differences in timing and behaviour that are clearly observable. If you don't believe me, open my 1.51 Draygon battle and watch the turrets shooting about TWICE as fast as on 1.43. Sounds like not-really-100% correct emulation from one of them, huh? What about Ridley with his rain of fireballs whose degree of intensity on 1.43 isn't even close to that on 1.51? You're still going to tell me that there's no difference and it's all about entering rooms on different frames?
Posted: 8/1/2007 9:10 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Saturn wrote:
Man, I really don't get what problems you have with v1.43+.
I was making a run aimed to be a reference for console, and 1.43 doesn't behave fully like console. It's forgivable for a TAS since most of the difference between it and unassisted run comes from using tools, so it's not like it would be comparable anyway. There are differences in timing that might affect boss strategies (Phantoon or Draygon for instance; it's been already confirmed that conditions are not the same between 1.43 and 1.51), or something like that. What would be the point of a reference that behaves differently from what it's actually supposed to reference? If I could, I'd do it on bsnes, since it's by far the most accurate SNES emulator in existence (followed by ZSNES, which has even worse desync issues and thus is out of question).
Posted: 8/1/2007 7:53 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Yeah, on a related note, I remember that I have planned to make a reference run for you, but it didn't quite worked out as I hoped. It was when I got to Norfair, I suddenly realized it desynced back at Torizo (yes, I did it on Snes9x 1.51 for obvious reasons). Since all the tricks had to be performed in realtime (I decided not to use slowdown except in situations where the keyboard was an obvious obstacle), I had to use a lot of rerecords and thus couldn't avoid the desyncs. So, unless/until rerecording is fixed in 1.51, I won't be able to devote my time to it.
Posted: 8/1/2007 5:11 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I don't remember the exact mechanism, but I have a feeling that Samus needs to be moving at a certain (0.24576 px/f?) speed at the moment of shooting to influence the projectile's acceleration. I usually test this each time separately, it doesn't take much time to determine the optimal button pressings over the span of 2-4 frames and then repeat them accordingly.
Posted: 8/1/2007 4:20 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Any glitch that enables the player to finish the game in as short time as possible should either be used to full extent or not used whatsoever. Similarly, it's common to have both conventional and glitched any% runs, a category for 100% run can be defined by entertainment value of glitched vs. conventional. Also, don't forget the Yoshi's Island or Illusion of Gaia situations, where a 100% run is so similar to the any% that it effectively renders the any% redundant, thus allowing it to be obsoleted by a full run.
Posted: 8/1/2007 4:15 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Cpadolf wrote:
Anyway what is the slowest you are allowed to run, 1forward 3 nothing right?
4 forward 1 nothing for continuous running, 1 forward <any> nothing for discrete motion (like you usually only need to be moving for 1-2 frames to successfully apply doppler shift to missile's acceleration; what happens inbetween is irrelevant).
Posted: 8/1/2007 1:16 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
AngerFist wrote:
I dont see how this game is different from Ice Climber or Donkey Kong.
Ice Climber and Donkey Kong had different levels, while this game didn't, IIRC.
Posted: 8/1/2007 12:38 AM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Baxter wrote:
Gelactic Gravity Golf
Haha, level 25 was awesome, I got over 3.6 millions on it alone! [EDIT] Damn, 16+ millions on level 32, with over 1300000 freefall score! This game rocks. [EDIT2] 14+ millions on 33, and ended with 61+ millions total.
Posted: 7/29/2007 2:54 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Out of curiosity, what was the solution you found?
Posted: 7/28/2007 10:09 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Wait, how is tech rating even relevant here? I used it to make a totally different point, which never emphasized the rating's reliability, nor a way for it to regulate anything.
Posted: 7/28/2007 8:25 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
laughing_gas wrote:
There's always snes9x 1.43v10, which IMO is one of the most reliable of all emulators, not just snes.
That's right of course, but it's less accurate compared to the aforementioned emulators, and it's well noticeable in the game. And it also has that awful messed up scanline below the statusbar in Super Metroid.
Posted: 7/28/2007 8:20 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
That's sad to hear. :\ I believe you should rather delay Super Metroid for later and continue working on less desync-prone projects (or at least emulators), until desync issues are solved. Seriously, though, why does it have to be like that? It looks like every good SNES emulator we currently have around has to have some severe issue. ZSNES has totally unusable interface and desyncs on flat ground, Snes9x 1.51 randomly desyncs and has issues with sound, bsnes has no TAS tools built in whatsoever (not to mention its hardware burden). Almost like if the current trend in SNES emulation is "break the movie support and fail to fix it".
Posted: 7/28/2007 5:06 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
ZeXr0 wrote:
By the way, the model are not of a great quality. Like the model of Tom Hanks... http://www.zbrushcentral.com/zbc/showthread.php?t=048910 That's a little more quality.
You still don't get the idea. The video isn't about the quality, it's about getting a decent and commercially usable result out of a single photo, with nearly no handwork involved, which can drastically diminish future production time and costs.
Posted: 7/28/2007 4:29 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
That makes sense only if the successor movie will definitely be finished, which can't be predicted reliably until the fact of submission. At the same time, a submitted improvement already has high enough chances to be published, especially if it has been known to be in production (most of the sub-par movies come either anonymously or without any prior notifications, so that no-one is able to judge the quality beforehand). Marking a movie suboptimal likely isn't a very wise thing to do, mainly because nearly all movies contain known improvements, and some of those improvements become known even before the actual submission, and it isn't well definable at what point to consider a movie improvable. For example, it's known that by the moment of writing this paragraph, JXQ's Super Metroid TAS contains over 30 seconds of confirmed improvements throughout, yet it has a very high tech rating and isn't going to be improved any time soon. Should one mark it as suboptimal? What about Genisto's SMB3 which by now also has known improvements and the new run is known to be in production (but who knows when and if Nitsuja is going to finish and submit it)? The same goes for Sonic 2 and S3&K runs, 96-exit SMW run, LoZ, Megaman 2 and so on. What to do with them? It's difficult to draw a line between "clearly suboptimal" and "good enough", especially with floating optimization standards coming from unavoidable ignorance and a bit less unavoidable game favoritism. For me, having a substandard movie on the site doesn't indicate that the site doesn't keep its standards, rather the opposite: it indicates that the standards are rising and only those successor movies that comply to them would be able to pass. Besides, it allows for higher dynamic range in tech rating.
Posted: 7/28/2007 2:38 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
ZeXr0 wrote:
On new keyboard, there's right-click button.
Boco wrote:
You can also try using the "context menu" button on your keyboard. It's between control and alt (assuming you're in the US and aren't on a Mac).
Posted: 7/27/2007 9:32 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
Cpadolf wrote:
Using my new strategy for recharging missiles in the diagonal room I improved 2 frames over moozooh. With this strategy it's possibly improvable by more though.
<NesVideoAgent> New reply by Cpadolf (SNES: Super Metroid): http://tasvideos.org/forum/p/137869 <moozooh> :D <moozooh> Cpadolf is my dad. Damn, I really thought it was impossible to hit that geemer and grab the refill in time! Thanks for teaching me the folly of my ways, hehe. Actually, it seems that with just a bit more experience, you'll excel all of us in no time. :0
Posted: 7/27/2007 7:43 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
I made some tests recently, that showed that an AVI with 450—500 kbps for video, coupled with very heavy duty encoding settings, looks considerably well in all the difficult scenes. If no-one volunteers, I'll make one (assuming I'll also be able to make a logo beforehand, haha).
Posted: 7/27/2007 7:32 PM
moozooh
Experienced Forum User, Senior Moderator
Joined: 8/4/2005
Posts: 5789
Location: Away
If you want to see a TAS of a bullet hell game, watch the current Perfect Cherry Blossom WR score replay by "HASEGA". It's not a TAS literally, but it's about as close as you will ever get. For the record, some time ago I tried to actually TAS Perfect Cherry Blossom (massive artificial slowdown, no rerecords obviously) and, although I managed to pull off some inane grazing, I failed rather quickly — died on one of Yuyuko's later spellcards. The replay is really pathetic, though, as I rush into barrages employing no scoring strategies at all, don't abuse supernatural borders, die and counterbomb a lot (!) and do other stupid things that don't look even remotely as awesome as the replay linked above. (Actually, my accomplishments on Lunatic are about the same as Maur's. :P)
1 2
150 151 152
229 230