What is the quickest possible time humanly possible to beat Doom/Doom 2/Final Doom?

As the title states, I'm interested to know if anyone has any information on this. I was watching Zero Master's 30pl1958 speed run of The Plutonia Experiment and his Doom 2 Nightmare run (30nm2235) and he basically just makes Doom his bitch, ha. But it got me to thinking/wondering: what is the minimum time needed to beat X, Y or Z WAD/what  have you? I know that people do TASs of Doom/Doom 2/Final Doom/etc. and that the focus of those are on getting perfect movements/frames/kills/etc. But obviously it's probably very close to being impossible for a non-cheating human to achieve such times, right?

So I'm just curious if anyone has done any research into the human limits of speedrunning Doom. 

P.S. People who speedrun the first Mario game on NES say that the theoretical human limit in a speedrun on that game has almost been achieved. 

2 minutes ago, Chewyninja69 said:

I know that people do TASs of Doom/Doom 2/Final Doom/etc.... But obviously it's probably very close to being impossible for a non-cheating human to achieve such times, right?

I think some might object to your equating TAS with cheating.

I suppose that TAS times are probably pretty close to the "theoretical human limit." If you want to factor in things like psychological and physiological factors -- hand muscles getting tired, effects of excitement or disappointment on one's skill, etc. -- I don't know if there is a way to determine these things.

Just now, 42PercentHealth said:

I think some might object to your equating TAS with cheating.

 

I suppose that TAS times are probably pretty close to the "theoretical human limit." If you want to factor in things like psychological and physiological factors -- hand muscles getting tired, effects of excitement or disappointment on one's skill, etc. -- I don't know if there is a way to determine these things.

I should've of worded it differently, my bad. I meant for a "non TAS'ing human"

That' probably impossible to tell at this point, since a new trick to break some maps has been discovered most recently, and it remains to be seen for which of the maps it will be applicable and how much time it will save ideally...

Just now, Nine Inch Heels said:

That' probably impossible to tell at this point, since a new trick to break some maps has been discovered most recently, and it remains to be seen for which of the maps it will be applicable and how much time it will save ideally...

I seen that yesterday on Youtube. Exciting stuff. Will be interesting to see how it pans out...

In theory, there's nothing that would prevent sufficiently skilled humans from approaching any "objectively perfect" performance arbitrarily closely. I think this fact alone is sufficient to end "research into the human limits" before it even starts.

Just now, Chewyninja69 said:

I should've of worded it differently, my bad. I meant for a "non TAS'ing human"

No need to apologize, I understood what you meant. I just think of things like IDDQD and IDCLIP as "cheating," whereas TAS-ers take advantage of things like savestates or slow-motion to make an "ideal" speedrun.

I understand the word "theoretical" to be very closely related to the word "ideal," so I'm willing to take TAS times as a suggested theoretical minimum.

1 minute ago, scifista42 said:

In theory, there's nothing that would prevent sufficiently skilled humans from approaching any "objectively perfect" performance arbitrarily closely. I think this fact alone is sufficient to end "research into the human limits" before it even starts.

^^ Yeah, what he said.

3 minutes ago, Nine Inch Heels said:

That' probably impossible to tell at this point, since a new trick to break some maps has been discovered most recently, and it remains to be seen for which of the maps it will be applicable and how much time it will save ideally...

No more map 31 in D2ALL...

So basically whoever sets the best time in a map/WAD essentially is the "ideal lowest time" until someone else comes along and beats it, I suppose?

A Doom 2 TAS should be around 10—12 minutes long. Plutonia is like 9 or 10. TNT is 17—18.

29 minutes ago, 42PercentHealth said:

I just think of things like IDDQD and IDCLIP as "cheating," whereas TAS-ers take advantage of things like savestates or slow-motion to make an "ideal" speedrun.

Technically, it's still cheating, but it's not the same kind of cheating. It's redefining the rules in order to produce a different challenge instead of an easier one, and TAS runners are anything but unskilled or lazy.

15 minutes ago, Da Werecat said:

Technically, it's still cheating, but it's not the same kind of cheating. It's redefining the rules in order to produce a different challenge instead of an easier one, and TAS runners are anything but unskilled or lazy.

This is what I meant earlier, 100%. I just couldn't find the words, but only because I work 3rd shift hours and haven't been to sleep yet, haha.

I have no doubts that there are some folk out there who use TAS tools mainly for cheating purposes or compensating for a lack of skill, but proper TASes from someone who, most importantly, knows what they are doing with the tools and how exactly to apply them, are pretty amazing.

I do wonder when we'll reach the lowest possible time. Obviously any map takes some time in order to complete, since you need to physically get from the beginning to the end switch, but with new tricks being found to this day, we still have time before we get a record that will never be beaten.

TAS runs have been used for Goldeneye/Perfect Dark to bring up hypothetical strategies a human player could use to speedrun a map, but sometimes the coordination is such that it would be a very hard/rare thing to do. If you are slightly interested, go watch RWhiteGoose's speedlore on Goldeneye, and you'll see how the world records for levels can change several times within a single week, and new strategies are found all the time even to this day to get shorter and shorter times even on levels that were thought to be routed perfectly.

idmap30

idclip