The latest most-officialist site for Doom TASes is now down. Being that TASVideos holds many enthusiasts of this type of speedrunning, I wonder if this site may create and host Doom speedruns?
Not only are there the official four IWADs (Ultimate Doom, Doom 2, Final Doom (TNT and Plutonia); still available for like $10 retail or online), there's also other commercial games derived from Doom (often little or no support in source ports, however), and full total-conversion PWADs available (eg, Batman Doom or Alien).
Source ports already exist with extensive tool-assistance capability, PrBoom and PrBoom+ are the ones most commonly used. Not to mention they're very portable, you can make (and replay) TASes no matter what operating system you're using (unlike the many emulators restricted to only Windows or Windows+Linux)
There's only one thing that might have to be done -- Doom TASes have traditionally always been distributed as LMPs, which are very small files with record the movements of the player, instead of distributing pre-rendered video files like this site often does. You've got some talented programers, however, I don't doubt AVI recording would take long to implement.
I published TASes of Doom at this site in the earliest days of this site. Later however I removed them because of disk space issues; I had to prioritize.
It is still an option we can consider. However, I am not familiar with the newest records, their update times, whether they're updated level-by-level or episode per episode, etc... And the software I used no doubt no longer works, so it would have to be completely re-investigated.
Joined: 4/21/2004
Posts: 3517
Location: Stockholm, Sweden
I think its a great idea. If the suggestion would go through, this along with the Doom community would expand even further along with several other positive aspects. But I think the procedure would be too complicated. Some issues that must be resolved are:
How do the Doom community feel about this?
Who would run the site?
New url?
What would happen with Bisqwits ads?
I can go on and on. In my opinion, there are too many issues :(
Nitrogenesis wrote:
Guys I come from the DidyKnogRacist communite, and you are all wrong, tihs is the run of the mileniun and everyone who says otherwise dosnt know any bater! I found this run vary ease to masturbate too!!!! Don't fuck with me, I know this game so that mean I'm always right!StupedfackincommunityTASVideoz!!!!!!
Arc wrote:
I enjoyed this movie in which hands firmly gripping a shaft lead to balls deep in multiple holes.
natt wrote:
I don't want to get involved in this discussion, but as a point of fact C# is literally the first goddamn thing on that fucking page you linked did you even fucking read it
Cooljay wrote:
Mayor Haggar and Cody are such nice people for the community. Metro City's hospitals reached an all time new record of incoming patients due to their great efforts :P
You say disk space issues, did you have AVIs or something? LMPs tend to be just as small as the emulation movie files, I don't see you having too much trouble hosting those.
Anger: I don't think you fully understand, I meant to include Doom TASes like the console TASes. Were there too many issues when Nintendo 64 runs were getting started? Actually I'd say there's more issues with N64, the state of its emulation and almost every game hadn't its engine source code released like Doom was (yes, tracking back a few source ports, PrBoom is based off of the original... DOS DOOM!).
Besides needing a mechanism to record AVIs, I don't think there'd be much trouble at all with this.
You may want to post about this in the DoomWorld demos forum as well, if you haven't done so already. :)
Regarding AVI making, there's at least one hack of prboom floating around that allows this (made by Dashiva), but I don't think it's been released to the general public, and the version I got is rather buggy, too. Still, Peroxyd (aka VinceDSS) has made a couple of AVIs (of regular speedruns, not TAS runs) with it that are available on my tracker. Note that these are recorded with motion blur - I wouldn't be surprised if this increased the size of the AVIs.
That being said, most of the DOOM speedrunning community seems to be focused more on regular speedruns rather than TAS runs, but unlike in other circles, the two groups are on very friendly terms. In fact, saying "the two groups" is actually misleading, since for the most part, they're one and the same.
BTW, the recent Doom 2 TAS by RamboBones (in 14:02 — 35 seconds faster than the previous record by entryway) was just awesome. And it was entirely frame-advanced by our terminology (or "built", by theirs).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Note that these are recorded with motion blur - I wouldn't be surprised if this increased the size of the AVIs.
Why would it?
1) You probably mean that it would reduce the quality of the AVIs. MPEG4 is always (well, in practice at least) encoded with a certain bitrate, so the end result will always be of a certain size (plus/minus a small variation), and if the input is harder for the encoder, it would just cause the quality of the result to decrease (not its size to increase).
2) Why would motion blur be more difficult for MPEG4 than non-motion-blurred video?
Note that these are recorded with motion blur - I wouldn't be surprised if this increased the size of the AVIs.
Motion blur actually helps modern MPEG 4 encoders, since all of them heavily rely on the motion vectors and the amount of overall movement on the screen (hence the sudden bitrate drops in static scenes and such), etc. The difference may be much greater than you may think (I've experienced up to 30—40% bitrate drop on some videos using motion blur with x264).
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Motion blur actually helps modern MPEG 4 encoders, since all of them heavily rely on the motion vectors and the amount of overall movement on the screen
Can you explain the exact technique in MPEG4 which makes benefit of motion blur?
I know that MPEG4 searches for shapes in the input which move on the screen and tries to outline those shapes and store just the difference in the shapes and their content. One would think that blurring the image just makes it more difficult to find such shapes. But of course that doesn't mean that MPEG4 doesn't *another* technique which is well-suited for motion-blurred video (or that the shape-searching isn't so smart that it couldn't find the shapes even if motion blur is used).
I have also noticed that MPEG4 uses more bits for the higher values of the spectrum than the lower values, which means that sharp changes in the image sometimes can even look better than smooth transitions (which may in some cases present visible artifacts due to low amount of bits used to store the info).
mwl wrote:
Whoa, I didn't know about the 14'02" D2 run. Thanks for the heads-up.
Can you explain the exact technique in MPEG4 which makes benefit of motion blur?
Well, I have a few guesses.
First, motion prediction and compensation. The more blurred the outline of an object in the temporal domain, the easier it is to estimate the vector. Blurred frames introduce an additional prediction basis that allows the encoder to spend the bits more effitiently.
Second, the fidelity factor. As can be witnessed with most other transform codecs (be it audio, video or picture compression), it's easier for them to encode gradients/clear tones/etc. than edgy transitions/noise/etc.. Blurring smears all the sharp edges which is a huge bonus for an encoder.
There could be more reasons for that, but I don't know them.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Bisqwit: You used to have Doom TAS's on your site too? Wow, didn't know that, very cool.
Honestly, adding Doom TAS's to this site would be wonderful. Like the icing on the cake.
Second, the fidelity factor. As can be witnessed with most other transform codecs (be it audio, video or picture compression), it's easier for them to encode gradients/clear tones/etc. than edgy transitions/noise/etc.. Blurring smears all the sharp edges which is a huge bonus for an encoder.
Yet when you start decreasing the bitrate of an MPEG4 video, the smooth transitions are the ones usually showing visible artifacts first while sharp edges maintain their integrity longer. I guess this is because, as I said, MPEG4 by default assign more bits to the higher end of the frequency spectrum than the lower end.
Just because a JPEG benefits from blurring doesn't necessarily mean that MPEG4 does too, even though both use the same fourier transform principle. I'm not saying that MPEG4 does not benefit, I'm just saying that it's not *automatically* so, and that my own experience has shown some signs of the contrary. I could be wrong too, though.
Encoding issues aside, I would just like to say that I support the idea to publish Doom movies here.
Doom is a huge game and has a complex physics engine, but I don't know if all these facts are compiled somewhere, just spread out within its community. A trick list here perhaps? I don't know if we should have a board here, since all Doom TASers seem to be gathered on doomworld.com already.
Doom speedrunning has expanded to a very big selection of categories; max (100%), fast modifier, Tyson (only hands and gun), pacifist etc. Also since the standard levels are pretty much optimized it has expanded to a selection of total conversions and extra levels.
I don't think we should publish any of those. Instead concentrate on these five:
- Doom 1 episode 1
- Doom 1 episode 2
- Doom 1 episode 3
- Doom 1 episode 4
- Doom 2
I'm a bit skeptical of including Final Doom.
Only NM difficulty would be the obvious choice, but the version mentioned above plays UV...
Only full runs, no single levels (unless it demonstrates something very cool that cannot be done during a normal full run).
Second, the fidelity factor. As can be witnessed with most other transform codecs (be it audio, video or picture compression), it's easier for them to encode gradients/clear tones/etc. than edgy transitions/noise/etc.. Blurring smears all the sharp edges which is a huge bonus for an encoder.
Yet when you start decreasing the bitrate of an MPEG4 video, the smooth transitions are the ones usually showing visible artifacts first while sharp edges maintain their integrity longer. I guess this is because, as I said, MPEG4 by default assign more bits to the higher end of the frequency spectrum than the lower end.
Just because a JPEG benefits from blurring doesn't necessarily mean that MPEG4 does too, even though both use the same fourier transform principle. I'm not saying that MPEG4 does not benefit, I'm just saying that it's not *automatically* so, and that my own experience has shown some signs of the contrary. I could be wrong too, though.
Well, since most if not all, MPEG4 codecs use JPEG image compression to produce the keyframes, anything which decreases the size of a JPEG will automatically decrease the size of an MPEG4 encoded video.
Furthermore, because of how MPEG4 generates non-keyframes using motion detection/prediction algorithms, anything which helps those algorithms to detect motion will result in fewer keyframes necessary, also reducing the filesize of the MPEG4 encoded video.
Motion blur helps on both counts.
How fleeting are all human passions compared with the massive continuity of ducks.
I could possibly think of one or two: a pacifist nightmare MAP02, because it has never been done unassisted, and nomonsters MAP07, where you normally need to kill everyone to lower a platform.
I don't think we should publish any of those. Instead concentrate on these five:
- Doom 1 episode 1
- Doom 1 episode 2
- Doom 1 episode 3
- Doom 1 episode 4
- Doom 2
Andrey Budko currently holds all of the Ultimate DooM TAS records. He posts on Doomworld as "entryway," and one could request permission to publish his work here.
Well, if you decide something on the inclusion of his works here, I could contact him. It'd be much easier for me because Russian is the primary language for both of us.
Warp wrote:
Edit: I think I understand now: It's my avatar, isn't it? It makes me look angry.
Joined: 4/8/2005
Posts: 1573
Location: Gone for a year, just for varietyyyyyyyyy!!
Truncated wrote:
I could possibly think of one or two: a pacifist nightmare MAP02, because it has never been done unassisted, and nomonsters MAP07, where you normally need to kill everyone to lower a platform.
Oh, true. Pacifist or Tyson MAP30 would be awesome too.
Edit: Or pacifist full game run.
Well, since most if not all, MPEG4 codecs use JPEG image compression to produce the keyframes, anything which decreases the size of a JPEG will automatically decrease the size of an MPEG4 encoded video.
In a typical MPEG4 video keyframes are very rare (unless there's a *lot* of change in the video), usually spaced hundreds of frames apart. Reducing the size of keyframes will not reduce the size of the whole video a lot. On the contrary, if we assume that blurring actually hurts the in-between frames (I don't know if it does, but let's just assume) then it will hurt the whole video a lot because over 90% of frames will be in-between frames.
Furthermore, because of how MPEG4 generates non-keyframes using motion detection/prediction algorithms, anything which helps those algorithms to detect motion will result in fewer keyframes necessary, also reducing the filesize of the MPEG4 encoded video.
But the question remains: Does motion blur help the encoder finding the shapes and their motion vectors or not?
I don't know how it actually works, but I can think of at least one argument pro and another con:
Pro: Motion-blur makes it easier for the encoder to find matching shapes because blurred images reduce the need to find very exact matches (IOW. more imprecise shapes will resemble each other in content and form thanks to the blurring).
Con: Motion-blur makes it harder to find matching shapes because the image is fuzzy and it's hard to guess what has moved and where. Almost anything could match almost anything else, making the motion vectors erratic and the contents of the moving shapes change a lot. (Also given that MPEG4 by default - probably - assign less bits for the lower end of the spectrum, the blurred content will show artifacts more easily than sharp content.)
I can't say which one of those is true.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Doom was released for the N64, Doom 64, so a TAS of it is feasible by our emulators, even though it was intended for PC. I don;t know hoiw well it runs though.
It's a totally different game. Different graphics, weapons, enemies, levels etc. It's also dark as shit. I had to turn up the brightness level to max in the plugin settings.
