Posts for Aktan

Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
DarkKobold wrote:
I would say that the use of continues was the main reason, however.
I had no idea! Well then, shame on you DarkKobold! Wahaha
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
X2poet, I should remind you the reason why this run was rejected was not solely due to the fact you used continues. It may have helped the rejection, but there were other reasons listed by DarkKobold as to the rejection. You should read (or reread) those to make sure it does not get rejected again.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Yes with Avisynth:
a = # Put source here
b = a.crop(0, 0, -0, 192) # Top screen
c = a.crop(0, 192, -0, -0) # Bottom screen
last = StackHorizontal(b, c)
return last
Post subject: How to playback in fullscreen with point resizing in MPC-HC
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
I know some of you like to playback videos using point resizing, so I should mention this feature is available in MPC-HC. Just go to Options, under Playback->Output, under DirectShow Video, choose VMR9 (renderless), set the 3D surfaces (recommended) *** for Surface, set Nearest neighbor for Resizer, both these sets are under VMR7/9 (renderless) and EVR CP settings ***. That's it! All fullscreen will now be resized using point resizer. Here is an image of that option: Note: For Vista and Win7, it should be EVR custom instead of VMR9.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
moozooh wrote:
I'm not, and was not, considering continues as a way to beat anything, period. Fast, good, doesn't matter. I've got a game over, that doesn't change, the fact that I buy myself a way to continue playing doesn't change the fact that I've got a game over.
So, any game over for you is considered bad. I guess you didn't like OoT much either, huh. I should mention, getting to a continue in Metal Slug needs planning too, if you think about it. You can't just get a continue on the whim via dying over and over at one spot and still save time. The deaths do need to be planned.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
moozooh wrote:
Why do you think that's new or clever or whatnot is beyond me. I thought it was common sense not to use them because it cheapens the whole experience completely and invalidates achievements, which is accepted in arcade communities worldwide. That's because everybody can get through any game with unlimited continues, no matter how hard it is, because the challenge in any non-puzzle game consists in the fact of having (extremely) limited resources that require skill to put to good use. If you can't see how making them unlimited in this way is equaled to cheating, I don't know how to explain it anymore.
So wait, if you were new to speedrunning, you would consider using continues as a good way to beat the game fast? Heck, if you have not seen so many TASes use death to beat the game, you would immediately think dying is a good way to save time?
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
moozooh wrote:
Some of his points also fail to take into account that, while speedruns in general (and TASes in particular) are ideologically allowed to warp the rules imposed by developers by being unexpectedly precise and arranging circumstances in a way to have additional leverage, they can't warp the definitions and principles of beating the game fairly, such as not using passwords, not using cheats, not using officially documented level skips (unless for a specific reason), and generally not making things easier for the sake of making them easier. His status as a game designer doesn't change that. (In fact, it bears no relation to the subject at hand whatsoever.)
Actually, I think his point (which I think you fail to realize), is that the original intent of continues was not to save time. Those examples you just gave are obviously known to save time. Continues on the other hand, by your definition of an arcade continue above, does not. Hence what x2poet did was out of the norm. I should clarify that what I mean by "save time", I meant in speedrunning, which is the goal of this TAS.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
So after talking to my cousin who is a game designer, I found some of his points interesting: cousin: "Well, is it a problem if you have 40? It defeats the point of having lives in the first place. The point which I thought was to use them wisely where it gives the maximum benefit, not to use them everywhere possible just because you can always get more on a whim by abusing the most banal and frowned upon way to gain them. " cousin: this statement looks a bit silly to me cousin: it seemed like one of the core elements of a tas is to ignore what any game element may have been designed to do cousin: "the reason the designer put a wall there is for the player to move around it... not to glitch through it" cousin: who cares what the designers intended when they added lives - i can attest that the number of lives you get is arbitrary until you take into account the money that players feed into the machine cousin: " In my opinion, making game easier by tapping into an infinite resource left open by developers for a very particular reason of aiding the incapable goes against the very notion of superplay, whether it's done for speed, for score, or for being awesome. " cousin: could this not be said of having infinite primary gun ammo in megaman? cousin: again i fall back to "who cares what the designers may or may not have intended?"
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
X2poet wrote:
But there have not this glitch. That using method is different way. It cannot be regarded as this one.
Oh, I know, I was just posting a semi off-topic theoretical question.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
moozooh: A question for you. Let's say in this game, using a continue in level 1 can warp you to the end level. Obviously this would be a "glitched" category, but would you then say the use of a continue is acceptable?
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Nahoc wrote:
I will try to at least do the SD encode (since I got this movie to sync well with Glide64).
Since I can't seem to grab you on IM or IRC, I'll post it here. I hope when you make the SD encode that you can capture at a higher resolution so you can do some AA.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Wyster wrote:
I tried reinstalling with no results. And according to every online source i could find VLC does not nor has never supported Lagarith playback?
I am apparently wrong. I meant MPlayer instead of VLC. My bad.
Wyster wrote:
btw, when i feed the dumped video to x264 i get the message "resize [error]: input colorspace Unknown format is not supported". Any ideas how to fix this?
Again, this is related to the codec and weird why no DirectShow players can play the file. When you reinstalled, did you reinstall VLC or Lagarith? I meant to reinstall Lagarith.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Warp wrote:
Does the visual quality really degrade that much in a high-resolution N64 game (which is quite different from games using pixel graphics)? (And even if some color shade gets changed just so slightly, is it really that catastrophic? We are not visual arts purists...)
Contrary to popular belief, the SD N64 encodes are not high resolution. Aka 320x240. Also N64 is a mixture of 3D and pixel art. The amount of pixel art is minor as it is usually the HUD, but it's still there. Since the capture is at a larger resolution, the color loss is less noticeable, but once re-sized to 320x240, it may still be quite visible. Then there is also the problem of H.264 hacked into the AVI container, regardless if there are b-frames or not. But that's a different issue. You could be around 1-2 frames off in sync due to this.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Wyster wrote:
Nope i didn't realize it might actually be a correct result after all, i'm gonna test this now. Used VLC for playback btw.
VLC didn't playback the Lagarith file correctly? That's interesting. I would really recommend you try reinstalling the codec. It sounds like something screwed up.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Wyster wrote:
I have no idea how to "correctly" dump this video but using the modified .kkapture guide i was able to dump without problems(it seems). However i used the lagarith lossless codec and x264 didn't seem to like that plus i can't playback the movie in any program, i.e i have no idea if it actually produced anything good. But at least it didn't crash. (I was able to dump a perfect quality video using FFV1 but it was too slow framerate and totally out of synch with sound)
That's weird that you cannot play it back at all. VLC has Lagarith available to decode so you could try that, but even VDub doesn't work? If not, it is why x264 is failing. You might want to try to reinstall it.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
More notes: This build of .kkapture also requires having a fast computer (unverified). If the capture sound is choppy, you may need a different build. All I did to change from the original .kkapture was set some variable lower so that it will capture faster. You can technically not use the modified .kkpature and still capture fine, but the capture rate will be VERY slow. As you may have already noticed, this modified version already captures pretty slow, now imagine the capture rate 50-100 times slower. Please let me know if the sound does become choppy (not from slow video playback).
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Mister Epic wrote:
I'd like to try, but the only problem is, even with 580 GB of free hard drive space, I'm not sure it's enough space.
This is why it is HIGHLY recommend you use Lagarith to capture, as all the duplicate frames will take almost 0 bytes. It should NOT take 500+ GB to capture at 1920x1440 res. Note: I have added this to my queue of things to encode, but it will be a while.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
A couple of other things to note: 1. The capture may have an extra 20 pixel border at the bottom. It should be cropped out. 2. Do not change the frame rate from 120 to like say 60. You may be missing frames in the area the video is really ~60 FPS. 3. The audio plugin adds a audio delay of around 192 ms. This may be related to the sound card, so I'm not sure this is accurate. 4. Due to having so many duplicate frames, it is highly recommended to publishable encode be dedup/deldup.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
VLC isn't the most efficient way to decode the video on Windows. You should be using MPC-HC + ffdshow + Haali media splitter. Just google MPC-HC to get that. As for the rest, follow the information here: http://tasvideos.org/forum/viewtopic.php?t=9964&start=40 Sounds like your laptop is a better bet to play it back than your desktop.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
PikachuMan wrote:
I have problems watching the cel-shaded video. I think the video freezes in about 1 minute while the audio still plays.
Are you streaming it (I doubt this)? What player are you using? It does require a pretty fast computer to play at that resolution. Or you need a player that is very very optimized.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
I'll be trying in a bit, but I have a long TODO list...
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
I use a custom build of .kkapture that I modify myself. It's a very slow capture, and requires a certain sound plugin, but it avoids a lot of the problems. The instructions on exactly how to do it is not easy either. I really should write it up, since I've been asked this a couple of times already.
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
Mister Epic wrote:
I'll leave this to you Aktan, and you can just modify the Archive collection I made since I put it in the Speed Runs archive.
I'm not sure if I have access to that..
Experienced Forum User, Publisher
Joined: 4/23/2009
Posts: 1283
feos wrote:
You were meaning that, weren't you, Aktan? XD
While I know you were joking, I do really plan to encode it eventually, when I don't have like 5 million things going simultaneously.