Sign in to follow this  
Followers 0

Fastest Possible Time: Classic Doom?

Marijo Sedlic's slowmotion-assisted Doom II run on Ultra Violence skill was clocked at 16:17, and Yonatan's runs through all 4 episodes of Doom I, plus the top times for each secret level listed on COMPET-N (since he left them out), totals 17:08.

Combined, that would make 33:25 for the entire official Classic Doom collection. From watching Yonatan's runs I could see quite a few precious seconds that could be shaved off, while Sedlic's was almost flawless. Does anyone think it might be possible to run through both WADs in 30 minutes or less, on Ultra-Violence skill and tool-assisted (saves and slowmotion ONLY)?

Share this post


Link to post

What about people running through TNT and Plutonia? They're a part of the Classic Doom set too. But probably, yes, somebody most likely could.

Share this post


Link to post

i don't mean to sound silly, but what is this slowmotion that people are referring to? i know very little about demos.

Share this post


Link to post

Slowmotion = running at reduced speed, which allows dodging and maneuvering to be much easier. All of a player's actions in slowmotion could be done at regular speed as well; it's just more difficult.

I did not include Final Doom since they were not designed by id Software; only marketed by them. I'd like to stay on topic though.

Share this post


Link to post

You might instead want to use the 16:05 Doom done Quicker as the reference for Ultimate Doom. Combined with Sedlo's run, that makes a total of 32:22 for the two games.

What about improving that?

Sedlo's run of Doom 2 has the technical perfection, but the routes used in some of the levels are old and have been improved upon since that. Which means that there's the odd 10 second improvement in a single map to go for, but most maps are already optimal or at most one second away from being so.

DDQr doesn't have the technical perfection you'd get from TAS, but it uses somewhat up-to-date routes. If any major time-savings are to be done they'd probably have to be done here. Some of the long, "linear" runs (and there are plenty of them in Doom) are at least 5 seconds away from optimal in DDQr.

So could the total be improved down to 30:00? I'm really not sure. 2.5 minutes DOES seem like an insurmountable amount, but there's a lot of maps to spread out the improvements over.

Share this post


Link to post

I just watched the DDQ and it doesn't include any secret levels. Also, due to the no-saving during levels rule Yonatan's overall time is faster. (16:13 without secret levels)

At which website are the new routes being uncovered? COMPET-N hasn't been updating much in the Classic Doom division.

Share this post


Link to post

Cameron Prosser stated here that less than 13 minutes is possible for a TAS run through Doom2.

Speed movies don't need to include secret levels - the reason why Doom2 speed movies normally include them is that the secret exit can be reached far more quickly on map15 than the normal one, and map32 is a useful fill-up level (though recently people have been tending to skip it).

The standard convention for Ultimate Doom episode runs is not to include the time for ExM8.

BTW, compet-n is constantly updated - new demos appear here.

Note that Sedlo's map07 trick would already take about about half a minute off his 30uv1617. There was also some relevant discussion here.

TAS sites are here and here.

Share this post


Link to post

Cam Prosser's 13 minutes time is based on his claim that he has lots of new tricks for most of the levels. Now maybe he does, or maybe he's being somewhat optimistic, personally I'd tend to think the latter. For example he says map03 is possible in 12 or 13 seconds, but that he can't get the trick to work. Obviously I don't know what this new trick is, but if it's what I think it is, then in my opinion is really is impossible, even in a built run. I haven't been into speedrunning as much as I used to, but my own guess for the optimum D2 time would be about 15 minutes.

Share this post


Link to post

Exactly. What new trick? He didn't say anything about it in either topic Grazza linked to.

I believe secret levels should be included for completeness.

Share this post


Link to post
BBG said:

What about people running through TNT and Plutonia? They're a part of the Classic Doom set too.

so are master levels. but none of those were made by id so no one really cares. they're just like pwads that were sold.

Share this post


Link to post

Cam Prosser's 13 minutes time is based on his claim that he has lots of new tricks for most of the levels. Now maybe he does, or maybe he's being somewhat optimistic, personally I'd tend to think the latter.


I'm with you on this one. Speculation and proof are different things, and he never showed us any proof.

I haven't been into speedrunning as much as I used to, but my own guess for the optimum D2 time would be about 15 minutes.


That's a good estimate for a TAS run. 36 seconds can be saved on MAP07 alone, and there are a few others. [rough estimates here] MAP10: 1 sec (glide), MAP12,13: 4 secs each (keygrabs), MAP14: 10 secs (RJ), MAP31,32: 13 secs (32 can be skipped), MAP24: 6 secs (RJ). That's 74 already which makes our time 15:03.

As for DOOM, here are the route improvements over Yonatan's TAS. [rough estimates here] E1M4: 1 second (jump), E2M7: 22 seconds (RJ), E2M8: 4 seconds, E3M2: 3 seconds (no plasma needed), E3M3: no change (elevator trick but BFG needed, evens out), E3M6: 1 second, E3M7: 4 seconds (exit jump), E3M8: 11 seconds (BFG), E4M3: 19 seconds (jump), E4M4: 1 second (jumps), E4M5: 2 seconds, E4M8: 4 seconds (RJ). 72 seconds, makes our time 15:00.

These are just route improvements alone. With a time of 30:03 already under 30 minutes is surely possible, mainly by shaving extra seconds off Yonatan's DOOM TAS effort.

Share this post


Link to post
sargebaldy said:

but none of those were made by id so no one really cares. they're just like pwads that were sold.

An odd claim. Anyway, there are TAS speedruns on Plutonia (UV and NM) and Evilution (UV) by Istvan Pataki, and Vile has done both 30ev and 30pl unassisted (see the compet-n site).

Regarding Cameron's 13 minute claim: maybe he has found some ways to trigger the noclip bug??

Share this post


Link to post
Jonathan said:

For example he says map03 is possible in 12 or 13 seconds, but that he can't get the trick to work. Obviously I don't know what this new trick is, but if it's what I think it is, then in my opinion is really is impossible, even in a built run.

Heh, jumping off the elevator to skip the blue key? That is indeed impossible.

Share this post


Link to post
Grazza said:

Regarding Cameron's 13 minute claim: maybe he has found some ways to trigger the noclip bug??

Huh?

Share this post


Link to post
Fredrik said:

Huh?

See Ledmeister's Blackbug.lmp. Is it known/understood exactly what triggers this bug?

Just a thought.

Oh, and on reading Cameron's posts again, it seems he's talking about a built run, rather than "just" a TAS run.

Share this post


Link to post

See Ledmeister's Blackbug.lmp. Is it known/understood exactly what triggers this bug?


Yes. cph has found it out a while ago. It is quite complex and can't be done on purpose (hell I don't even remember the explanation).

Share this post


Link to post

Might have to do with random gaps in the collision code, like when fireballs fly through walls, but that's just an outsider's speculation (in other words i'm clueless).

Share this post


Link to post

If you left out E4 it could easily be done in under 30 minutes.

Share this post


Link to post

BTW, Altima's av08-204.zip contains a noclip bug demo. He wasn't able to (or chose not to) exit in that one, but instead spent a while walking through walls and stuff.

I'd be surprised if it is related to fireballs flying through walls - isn't that just due to the speed of the fireball, the thickness of the wall, and the moments at which the game engine happens to calculate the fireball's position?

Share this post


Link to post

Thanks for the list Adam. Hopefully someday the demos will be redone, either by new Doom fans or veterans that have decided to return, although it is rather unlikely without motivation.

I do not consider exploiting bugs as legitimate, and after all, Led has said that it does not work on all computers.

Share this post


Link to post
Grazza said:

the thickness of the wall

All walls in Doom are infinitely thin :P

Share this post


Link to post

mlin said:
I do not consider exploiting bugs as legitimate, and after all, Led has said that it does not work on all computers.


Most bugs work anywhere. As for the no-clip bug on Altima's demo; it isn't a record demo, it's just accompanying the valid demo as an amusing thing Casey decided to share with other demo viewers and players.

Share this post


Link to post

If the fastest verified time for every level was combined into a single demo, its total length (Doom I & II) would be 30 minutes, 29 seconds.

If all of Adam's yet-to-be-verified (I say this because I did not find any demo, TAS or otherwise, demonstrating them) time reductions were done and combined into a single demo, it would be 29 minutes, 58 seconds.

Hence, if Adam's estimations were accurate, it would be possible to complete both Dooms in under a half hour. Amazing, isn't it? That's an average of 28 seconds per map.

All details and calculations:
http://www.angelfire.com/blog/bloggg/cdoomfpt.html


(Key: Y = Yonatan's 16:13 Ultimate Doom; C = Individual COMPET-N records; D = Ultimate Doom Done Quick; M = Marijo Sedlic's 16:17 Doom II)

Share this post


Link to post

If all of Adam's yet-to-be-verified (I say this because I did not find any demo, TAS or otherwise, demonstrating them) time reductions were done and combined into a single demo, it would be 29 minutes, 58 seconds.


I did lv14x010 and lv24x030 nomo runs a while ago, with modified WADs (I added a rocket launcher and a mega somewhere on the route). MAP14 could be done even faster with the newer RJ variant used in lv14-020. I don't usually pull out these times from thin air ;)

Share this post


Link to post
Grazza said:

An odd claim.

well maybe that's just me. but i really don't see how TNT or Evilution is much different than a Community Chest or Icarus, except that it was packaged by id and isn't available for free. there's no new endboss, or plot, or weapons, or coding.. then again i hardly think doom2.wad had anything going for it either.

Share this post


Link to post

sargebaldy: It was the "no one cares" that I found odd, especially as you seemed to be suggesting that no one had put much effort into speedrunning through Final Doom.

mlin: Most speedruns involve exploitation of bugs (e.g. wallrunning, grabs; straferunning could be called a "software design error" too), though not generally quite such obvious ones as the noclip bug.

Share this post


Link to post

just a couple points of clarification:

1) I've never known Adam to "guess" at anything. :)
2) This is a c and p of cph's response:
"Subject: Re: manormis.lmp (YADB)
Date: Sun, 01 Jul 2001 13:16:19 GMT
From: cphipps@doomworld.com (Colin Phipps)
Newsgroups: rec.games.computer.doom.editing

In article <20010630215620.10186.00003553@ng-md1.aol.com>, Led wrote:
> Anyone feel like trying to nail down the All Ghosts bug (all monsters and
> players suddenly become ghosts)?
> Opulent and some others here should know where to find the Doom 1 example of a
> player encountering the bug during a Nightmare game on map E1M3.
> Some Doom 2 Master Levels examples of the bug are here:
> http://m1.aol.com/ledmeister/manorbug.lmp
> (for Manor.wad, most illustrative demo of that bug on my site)

Looks like it's causes by an intercepts overflow; on that demo it SEGV's in
linuxxdoom-1.10:

Program received signal SIGSEGV, Segmentation fault.
0x80645df in PIT_AddThingIntercepts (thing=0x40323720) at p_maputl.c:671
671 intercept_p->d.thing = thing;

(interestingly, it's actually the sergeant shooting at the player when this
occurs on manorbug.lmp)

extern intercept_t intercepts[MAXINTERCEPTS];
extern intercept_t* intercept_p;

As soon as Doom exceeds MAXINTERCEPTS, intercept_p gets overwritten and it
starts trashing memory. Fixed in Boom (without demo compatibility, but it's
kinda hard to be demo compatible with that kind of lunacy :).

--
Colin Phipps <cphipps@doomworld.com> http://www.cph.demon.co.uk/"

oh, and Prosser is certainly one of the very few players skilled enough to manage some of these tricks.
Lastly, the blacktwr run is possible, because he dies in the lava. You cannot interact with the linedefs once you are in that state(ie. switch a switch or open a door).

Share this post


Link to post

Thanks for digging that up. It certainly clarifies things nicely. ;)

Opulent said:

Lastly, the blacktwr run is possible, because he dies in the lava. You cannot interact with the linedefs once you are in that state(ie. switch a switch or open a door).

So, it can only help on maps with a "-10/20% health, end level/game" sector?

Share this post


Link to post
Grazza said:

So, it can only help on maps with a "-10/20% health, end level/game" sector?


exactly. :)

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0