Doomworld Forums - What constitutes TAS? [split from 2 glides]

Doomworld Forums : Powered by vBulletin version 2.2.5

Doomworld Forums > Special Interest > Doom Speed Demos > What constitutes TAS? [split from 2 glides]

Pages (2): « 1 [2]

Author

All times are GMT. The time now is 10:21.

Qaatar
Member

Posts: 385
Registered: 06-09

At last, this thread finally went somewhere. I mostly agree with dew's points. Finally, someone with a slightly different perspective, can explicate this issue more succinctly and accurately than all of the pontification that I just spewed - I'm technically not a native speaker either, but whatever. Although, many of these points were brought up in many other threads and private messages before this, but oh well...at least this thread isn't entirely worthless (to quote Phml).
:)

12-15-10 21:15 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

myk
volveré y seré millones

Posts: 14422
Registered: 04-02

Qaatar said:
We're just pushing opinions around now,

As opposed to what, and objective and scientific discussion? But I do see the cause of reiteration in the discussion here. Picking apart bits and pieces of things just lets each speaker establish a parallel opinion without really debating the underlying differences of opinion. I may also have contributed to that even though I was trying to pull my arguments together to support a general idea, not isolated fragments.

To combat that fragmenting wall in the debate I'll ask a question:

Agreeing that PrBoom+ already provides some differences with vanilla, is it not reasonable to use the features PrBoom+ offers, such as the HUD, smart totals, automap overlay, high resolution and the related application of high DPI mouse sensitivity, more liberal key location arrangement, possible lighting alterations and wide screen support freely and on their own grounds?

So while I was wrong to label this thread definitively as "bullshit," most of what we've been arguing over, in my opinion, has been rehashed BS.

LOL, whatever, another of these "ok, no, but yes" comments. See my observation about fragmentation above for a what I think is really going on.

Yes, not wanting to compete in C-N has many complex factors for me. I think I outlined quite a few in the previous threads (such as not wanting to beat my Doom heroes, etc.).

From what I see, the overriding factor is the differences between the engines.

You might not care or be knowledgeable about this, but that's exactly my point: you and many don't seem to understand the mindset that I have.

I said I don't care much about Boom and that makes me ignorant about your type? I've been paying attention to demos since 1999. One would think I'd have an idea about demo optimization and perfectionism by now. My reduced appreciation for Boom stuff is because it alters the core game functionality and I'm not very interested in that. Hence when I watch a Boom demo it's more for entertainment, just like with any port demo. With vanilla demos, especially true ones, I pay much closer attention to technical and competitive aspects. It's not like high perfectionism is incompatible with a preference for vanilla. See kimo and xepop for good examples.

As a demo perfectionist, I want to optimize all demos to the best of my abilities, without TAS.

I think you can do this even more fully, without needing to hold back in some ways, if you consider what I'm arguing.

Phml said:
Absence of evidence means no evidence now? You can't be serious.

Look, I don't need you to use 320x200. The fact that I have certain ideas about this subject that clash with yours doesn't mean I'm trying to get you to do what I prefer. If you don't want to, for whatever reason, be it scientific, aesthetic, religious, impulsive, hedonistic or because it works better with a combination of LSD and pot, go ahead and use 1600x1200 or whatever, but don't expect to have me agree with your interpretation without evidence against what I've observed personally and gleaned from other sources and statements.

Go see an ophthalmologist and ask for explanations.

That would be your prerogative, not mine. You even said you tend to think 320x200 may do damage in general, even to me. I don't think so and I haven't had the need, related to the notion, to consult an expert on whether I may be damaging my eyes.

dew said:
for example the recent slew of SAV's choco/prboom demos does not qualify for the official tables despite those often being vastly superior to older records, however calling them TAS just because of different executable used would be nonsensical.

Calling them TAS compared to vanilla can make sense. That's how I see it, more or less. There are degrees compared to the use of slow motion and saves, but you're still using a third party tool with enhancements. The comparison applies to any vanilla versus PrBoom/+ demo, and not just toward Compet~n.

that'd be like saying there's just c-n type demos and TAS type demos and pure demos for limit removing wads would become impossible. or rather the minimal set of features for limit removing would either have to include the non-TAS prboom features, or dictate use of eternity... not really a feasible solution.

Your picture is incomplete. What we call TAS in PrBoom/+ is previously defined by vanilla standards. Essentially, TAS is the use of any tools to obtain a machine readable (compatible) demo, generally to optimize a route. PrBoom+ is a TAS engine, although due to vanilla traditions many exclude the strong TAS functions to record a demo that doesn't disrupt the usual flow of the game (speed changes, segmentation.) Now, to a PrBoom/+ regular, what vanilla can do pretty much fits among the options in PrBoom/+, so PrBoom/+ users are tempted to include vanilla demos as part of the fold. This doesn't apply both ways.

because of the vanilla clause i respectfully stay the hell out of them.

What I propose is that you go on and record on these as well, just acknowledge the difference and post demos in the right places (or they'll get moved). Maybe make a PrBoom+ Alien vendetta thread if there isn't an alternative one already and post there, for example. I kind of sympathize with you holding back, and ideally I'd love to see people posting more vanilla demos instead of vanilla compatible demos made with a port that adds extraneous features to the game, on any level or map set compatible with Doom or Doom+, but as long as people leave spaces open and do things respectfully, it should be possible to have both things. But that is why I put an emphasis on the difference between the engines.

aaand for the last, wads outside c-n. there are no tyrannical rules set for them on either dwf or DSDA, no golden set of instructions except a vague tip to "go as vanilla as reasonably possible". this would be where the merit of this discussion actually lies for me. c-n wads be damned, but i don't see why my achievements on scythes, pl2 or ksutra should be cheapened just because someone prefers choco or vanilla. times have changed, the old c-n crowd is 99% inactive and i'm not keen about self-imposing harsh restraints on me. it's strictly the runner's choice to record a pl2 demo in vanilla, so claiming my pr+ demo is inferior or incomparable would be unfair and uncalled for. it means applying c-n rules outside of their bounds and i couldn't care less for such pedantry. imo there should be certain sensitivity to the wad's age of origin. zdoom doesn't fit into this at all, because of different physics.

"Moving on" is a personal choice, just like avoiding the tyranny of source port features and choosing to record with vanilla, regularly or circumstantially. Just like it's the runner's choice to record with whatever he wants, even ZDoom, it's the viewer's choice to judge a recorded demo.

To me it doesn't matter that much whether a demo is for Compet~n or for whatever. Either way, when I sit down to watch a demo I have the fact that the user was using vanilla or applying any possible set of features from the port in question in my mind, and it changes the way that I see the recording. Compet~n just helps make vanilla more of an option. This goes back to my previous comment about how PrBoom/+ includes vanilla and vanilla excludes PrBoom/+, and explains why we see it differently.

In fact, especially now with the strict vanilla compatibility flag allowing ZDoom more purism, PrBoom/+ is in some senses closer to ZDoom than to vanilla, as far as the "interface and input" aspect is concerned, such as the HUD, resolution and key binding.

12-16-10 05:18 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

Qaatar
Member

Posts: 385
Registered: 06-09

myk said:
As opposed to what, and objective and scientific discussion? But I do see the cause of reiteration in the discussion here. Picking apart bits and pieces of things just lets each speaker establish a parallel opinion without really debating the underlying differences of opinion. I may also have contributed to that even though I was trying to pull my arguments together to support a general idea, not isolated fragments.

Yes, an objective discussion is certainly possible. For instance, we could be discussing how each feature works objectively, and then compare the subjective effects these features have on different players. We can, for example, compare and contrast exactly how much difference a certain feature can provide, and weigh that difference against other features across other players. Scientific? Don't think that's possible, and since we've established that no one here is trying to institute an arbitrary standard, there's no need for empirical data. All I wanted was some in-depth analysis of each feature as well as a general feel of the opinions of each player on said features.

Instead, we're just pontificating, which is fine and enjoyable, but not informative at all in my opinion. For most of these points, I can do a search on these forums, and probably find similarly constructed opinions from you and others.

myk said:
To combat that fragmenting wall in the debate I'll ask a question:

Agreeing that PrBoom+ already provides some differences with vanilla, is it not reasonable to use the features PrBoom+ offers, such as the HUD, smart totals, automap overlay, high resolution and the related application of high DPI mouse sensitivity, more liberal key location arrangement, possible lighting alterations and wide screen support freely and on their own grounds?

For one thing, we already acknowledged that pr+ has differences from vanilla, so that isn't the starting point of this debate. The starting point of this debate is HOW those other features affect other players. Why you cling onto this vanilla vs. pr+ debate, I have no clue.

myk said:
LOL, whatever, another of these "ok, no, but yes" comments. See my observation about fragmentation above for a what I think is really going on.

If I was being disingenuous and wanted to call it bullshit anyways, I would have straight up, without equivocation, said 100% bullshit. Indeed...LOL.

myk said:
From what I see, the overriding factor is the differences between the engines.

Again, we already know there are differences, so no, it's not the overriding factor. Hell, it's not even a factor. See above.

myk said:
I said I don't care much about Boom and that makes me ignorant about your type? I've been paying attention to demos since 1999. One would think I'd have an idea about demo optimization and perfectionism by now. My reduced appreciation for Boom stuff is because it alters the core game functionality and I'm not very interested in that. Hence when I watch a Boom demo it's more for entertainment, just like with any port demo. With vanilla demos, especially true ones, I pay much closer attention to technical and competitive aspects. It's not like high perfectionism is incompatible with a preference for vanilla. See kimo and xepop for good examples.

I don't care about your reduced appreciation, and I'm sure that you don't give a shit about my demo recording motivations. I never said high prefectionism is incompatible with anything, but Boom is a category that has a great many wads, and thus, demo opportunities. Unfortunately, we're stuck with a very nebulous set of rules and regulations regarding source ports in Boom, and what TAS entails within those confines. Therefore, it's important for me to learn about the usage and objective definitions of those confines, the opinions of others towards those confines, and to be as informed and transparent as possible about all of those encompassing details. As a side note, this is not incompatible with the stance on Zdoom, because the actual gameplay differences in terms of skill requirements between Boom and vanilla are practically non-existent. Between Zdoom and Boom/vanilla, it's almost night and day.

myk said:
I think you can do this even more fully, without needing to hold back in some ways, if you consider what I'm arguing.

Just to reiterate again, in my opinion, all of this has nothing directly to do vanilla. This has everything to do with the general perception of certain features in certain source ports, in this case, pr+. If the end result is something that's close to vanilla, then I think that it will be purely incidental.

I've always considered what you argue, but I'm afraid that we're not even on the same page in regards to this debate. Don't get me wrong, of course I would like to use those features and think that they're not TAS. It would cut down on recording time, and most probably, raise my skill level. But this is the problem: I don't want a gross misrepresentation of my skill level. I want to be as transparent as possible, and to make my demos as organic and genuine as possible, within limits, and to be consistent across all types (vanilla and Boom). I think we all use common sense in this regard. Most of us will probably be okay if a certain feature allows someone to enhance their skill level by, say, 1% (for example, maybe screen resolution for Artem...he's obviously just as good of a player with choco). We won't be okay if a certain feature allows that person to record something that wouldn't otherwise be possible (slow-motion, segmentation, or even the monster counter in some instances). However, I wouldn't have known most of this, and I was hoping that this thread would give me more insight into these issues. Why? If I had started recording during the golden age of C-N (say, 8 years ago), I most probably would have trained myself on vanilla. Instead, I started recording this year, when probaby 90+% of players use pr+. Thus, I went with the general majority, and started using pr+. I believe that personal opinions of every player regarding the features of source ports are heavily influenced by the overall general perspective.

By being as transparent as possible in my playing, and by looking at exactly how the general opinion of certain features are, I can feel comfortable and be proud of my demos, knowing that I represented myself well. Selfish, yes, but whose motivations aren't?

12-16-10 06:56 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

vdgg
Moderator

Posts: 872
Registered: 08-07

Qaatar said:
The starting point of this debate is HOW those other features affect other players.

In all my PrBoom plus demos:
- I never ever use K/I/S HUD during recording
- I have the same keyboard configuration as in Vanilla demos
The only potentially advantageous feature I use is higher resolution (640x480, to be exact). Two times I tried recording a demo in both Vanilla and PrBoom plus (MM2 MAP09 Pacifist and Darken2 MAP01 Max) and in these very maps higher resolution is beneficial (e.g. sniping those imps on the mountain). In most cases I think the screen resolution does not matter at all. (Why I use PrBoom+ almost exclusively then? It does not crash as often as Vanilla; For C-N stuff I use Vanilla just to show respect to my grand predecessors - then of course PrBoom plus for playback - checking kills statistics and such).

12-16-10 08:07 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

Qaatar
Member

Posts: 385
Registered: 06-09

@vdgg

I'll post all of my settings in depth when I have more time...gotta go to sleep. I think it's great that you kicked this off!

Perhaps we can sticky this to the forum if more people join in? Or maybe do something like the C-N player profiles, and modify the DSDA player templates to include these specifics? Once we do that, maybe encourage everyone to fill one out on DSDA? Just some ideas. :)

12-16-10 09:11 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

Qaatar
Member

Posts: 385
Registered: 06-09

These are my settings for the majority of the demos that I've recorded thus far:

Glboom+ 2.5.0.8
HUD (wider pov, kills/items/secrets, timer, no unequipped ammunition information)
1280x800
Gamma factor 4
Uncapped Framerate
Custom key configuration:

WASD - Forward/Strafe left/Backwards/Strafe right
M1 - Fire
Shift - SSG
Q - Turn Right
E - Turn Left
R - Rocket launcher
F - Plasma
C - BFG
1 - Fist/Chainsaw
2 - Pistol
3 - Chaingun
Alt - Strafe-on
Space - Use
Mouse sensitivity - 18

Probably will switch to 'M2 - Use' once the full 2.5.0.9 is released.

Custom key config is actually quite interesting. For example, stretching my fingers to press 7 for the BFG, in such a fast-paced game as Doom, is rather inefficient. I haven't played vanilla in a long time, so I don't know to what extent one can customize key settings? I think I have covered the monster counter extensively, so there's not much to add for me. It would be interesting to hear about the timer effects from the speedrunners like dew. Unequipped ammo info plays a big deal in plutonia-type maps that limit cells, as I stated earlier. Gamma factor 4, I've already compared to perma light amplication goggles, so not much to add there. It provides an advantage for dark maps that are intended to be played with vanilla in 320x200 (as Hitherto pointed out before). Uncapped framerate makes a big difference in my ability to play for lengthy periods of time (over 20-30 minutes), as I get nauseous and dizzy otherwise. One thing that I've noticed is that even in pr+ (not gl+), with uncapped framerate, the picture still stutters for me, and acts like the framerate is capped. Why that is, I have no clue, but it's one of the biggest reasons why I use gl+ instead of pr+. Anyways, those are just some observations and questions that I have.

12-16-10 21:23 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

dew
Feral Kid

Posts: 1702
Registered: 05-08

Qaatar said:
It would be interesting to hear about the timer effects from the speedrunners like dew.

uhm. obviously in longer runs i remember certain checkpoints where i take a quick glance at the timer. you usually feel that the run is slow/fast, but being sure is a nice gift, considering records are sometimes achieved in runs that felt retarded. :) this is mostly useless in short runs and interestingly enough it can even lead you to sabotage yourself by obsessing about the timer too much and not paying attention, or getting nervous when a checkpoint time seems awesome.

the real advantage of timer shows once you learn to quickread it after each exit, before the screenwipe to intermission. maxruns don't care about hundredths, but in speed grinds you really want to know if the run was 15.51, 15.20 or 15.03. watching the demo again with timer or running it through lmpc is a goddamn waste of time you could've spent on another 2 attempts. actually.. if the precise time showed during intermissions, timer wouldn't be such an issue to me. :p

my setup:
prboom+ 2.5.0.8 (only use gl for playback, it loads too slow for my needs)
smart hud (spoiler statistics, timer... i don't really follow the rest)
640x480 (uncapped, although i wouldn't have a problem otherwise since zdaemon is capped)
gamma one below max.. i think

key binds:
WASD, 12345VC - weapons in this order, E - use, ZX - turns, right mouse button - +strafe, / - turn180 :), space - fist, Y - quit
weird, right? cause.. Q, space - jump, ZRFGHB - CTF binds in zdaemon, hehe.

12-16-10 23:17 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

SAV88
Member

Posts: 308
Registered: 11-10

My control is mobile enough for me, even though I use keyboard only. My configuration is almost like the standard Doom controller setup, except from a few details.

Fire -- Ctrl
Run -- Shift
Strafe on -- Alt
Use -- Space
Strafe left -- A
Strafe right -- S

Up -- Numpad 8
Down -- Numpad 5
Left -- Numpad 4
Right -- Numpad 6

I play without autorun. Not only it would be a disadvantage for me being unable to simply walk, but also I'm SO used to the Shift key that I can't run comfortably when not holding Shift down :-)

To simply strafe, I use Alt. I press A or S when I implement Strafe50 (many times per each minute) or execute Strafe40-on-turn.

As for the weapons, my fingers are long enough to press standard weapon keys like "5" and "7" operatively and comfortably.

I'm used to play with normal screen size, with status bar on. In PrBoom, I turn the HUD on for a few gametics when I want to look at the timer and/or the monster counter.

Pad Enter -- HUD on
Numpad 0 -- Decrease the screen size (in order to turn HUD off and restore the status bar).

__________________
My Doom speed demos, old and new

Last edited by SAV88 on 12-17-10 at 09:47

12-17-10 09:38 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

Creaphis
I will deliberately take a contrary position just for the sake of writing incredibly long arguments

Posts: 3991
Registered: 10-05

I suppose that in Qaatar's perfect world, we would be able to gather such an extraordinarily large amount of information about players, maps and features that we could devise a statistical model to adjust the times of demos based on the features used and the maps recorded on, so that demos could be compared fairly. A hypothetical example:

code:PrBoom+ feature constants:

H (Hires) = 1.02
T (Timer) = 1.03
S (Smart totals) = 1.05
P (Pausing) = 1.07

MapX/feature interaction effects:

XH = 1.00
XT = 1.02
XS = 1.09
XP = 1.03

Demo1 features used:
Hires, Smart totals

Demo2 features used:
Hires, timer

Demo1: 905 (base seconds) * H * XH * S * XS = 1056 (adjusted seconds)

Demo2: 1025 (base seconds) * H * XH * T * XT * 0.95 (Keyboarder bonus!) = 1043 (adjusted seconds)

Therefore, we can conclude with 73 percent certainty that demo2 is better than demo1 (statistical calculations omitted).

Obviously this is absurd, and any work in this direction is absurd as well, despite (but also because of) the impossibility of gathering enough data to be able to create this ideal method of comparison.

If we want a fair way to compare PrBoom+ demos to other PrBoom+ demos, then we can do as Myk suggests and simply consider all PrBoom+ features that aren't explicitly TAS as fair game. If a new runner beats the time of an older PrBoom+ demo, partly thanks to the help of extra features, then it's the fault of the previous runner for not using them as well. If you don't like that, whoever you are, then feel free to codify your own standard which allows only a specific subset of PrBoom+'s features. Adam Hegyi is only a man after all, he bleeds when pricked, and we can all feel free to arbitrate the rules of Doom speedrunning as he once did. If we're lucky, one of these new standards will catch on, and then we will once again enjoy a speedrunning environment with a truly level playing field. New demos will continue to be produced that don't qualify for this new standard, but that's no concern, as right now all PrBoom+ demos are merely shambling about in this lawless wasteland.

12-17-10 19:33 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

4mer
Junior Member

Posts: 198
Registered: 08-10

Something I would like to see but I doubt it will ever happen, who knows maybe someone will read this and be inspired.
Something along the lines of prBoom- (prBoom+ but minus the serious TAS features i.e. slow mo) timer and monster count would be allowed.
A source port that would provide a new level playing field for demo recording and competition.
It would have some sort of anti-cheat features to prevent TAS demos (i.e. anti-tampering, encryption) and could have some sort of internet interface i.e. for linking up to demo servers for uploading new records / replaying historical ones (i.e. like ScoreDoom but with demos instead of high scores).
Players could perhaps choose a category as they begin the game...

12-17-10 21:00 #

Profile || Blog || PM || Search || Add Buddy

IP || Edit/Delete || Quote

Quasar
Moderator

Posts: 4615
Registered: 08-00

4mer said:
It would have some sort of anti-cheat features to prevent TAS demos (i.e. anti-tampering, encryption)

Afraid this is a case of "build it and they will come," "they" being hackers in this case. If you build a system like this, which will always have a way around it (and it must if you're able to play back the demos), it will at best attract people who wish to defeat it just for the purpose of having done so.

Encryption won't do anything since the key has to be in either the demo or in the source code for the engine. If you cannot modify one, then you can modify the other. Signing is useless, since again, the signing code is open, and even if it weren't (and it HAS to be because PrBoom is under the GPL), it could be circumvented by anyone that reverse engineered the algorithm.

You ultimately cannot circumscribe the issue of trust. You either choose to trust the runner's word that their runs are legit, or you do not. It will never be possible to scientifically demonstrate beyond doubt that a given run is uncheated - unless of course it's so awful so as to make that obvious (though even then it's technologically feasible to "build" a horrible demo - it's just not worth the effort).

12-17-10 21:10 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

4mer
Junior Member

Posts: 198
Registered: 08-10

I was thinking that any anti-cheat measures would only be to dissuade casual hackers.
The system would rely on trust and perhaps have a community of genuine users who would be observant to such attacks.

12-17-10 22:00 #

Profile || Blog || PM || Search || Add Buddy

IP || Edit/Delete || Quote

dew
Feral Kid

Posts: 1702
Registered: 05-08

4mer said:
A source port that would provide a new level playing field for demo recording and competition. ...anti-cheat features...

anders already brought this idea up, i remember it being discussed a lot on #nightmare cca. in january this year. he looked for some coders to do the job and got shot down, because let's not kid ourselves, a new port for how many people? 20? 40? not to mention there are people that'd break the encryption and release hacks just to prove they could.

i came with an idea then and i'll repeat it now becuase i'm not really sure how silly it is. :p
imagine competn2.exe, an utility that you run in the background during your recording session. it reads what doom/port writes into demofiles, adds timestamps and feeds it to a competn2 server. when player submits a demo, logfile of the whole recording session is made public and people versed in the works can study the hell out of it. you know, like audioscrobbler for doom. plus no need to build new ports, it might even recognize what is starting up, unless port exes start faking their identity of course. downsides are obvious: it's a logging program in your system and it needs to connect to an online server.

edit: of course it'd still be possible to fake a logfile, but it'd become a real pain in the ass to get the continuity and timestamps right.

Last edited by dew on 12-17-10 at 22:29

12-17-10 22:23 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

4mer
Junior Member

Posts: 198
Registered: 08-10

Maybe recording a demo in real-time while on-line and perhaps receiving the seed for the random number generator from a server may be enough.
Who cares if someone hacks it anyway all their demos would be quarantined / deleted once they were discovered.
There would be no point hacking the system if no-one ever new about it.

12-18-10 10:45 #

Profile || Blog || PM || Search || Add Buddy

IP || Edit/Delete || Quote

Creaphis
I will deliberately take a contrary position just for the sake of writing incredibly long arguments

Posts: 3991
Registered: 10-05

Why waste our time building an ineffective anti-cheat port that does nothing but create an atmosphere of mistrust? Even if that wasn't a stupid idea, there are no rules for an anti-cheat port to enforce as of yet because nobody's proposed a new standard ruleset. There's a sequence to these things.

12-18-10 15:48 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

RjY
anARCHy

Posts: 802
Registered: 05-02

Here's an interesting case, opinions about which I would like to hear.

Yesterday I took a long demo by The Green Herring with lots of deaths and level resets in it, and chopped out all the failed runs, leaving just the final successful exit. Ordinarily this would just make the thing desync but I was able to tweak the random number generator seed value in the Boom demo header to make it play back properly.

The question is, does this sort of thing, in your eyes, count as TAS or not?

I would say not because the original demo wasn't recorded with any TAS features besides the usual interface enhancements you get from PrBoom-Plus normally, it's just cutting out the usual die-quit-restart recording cycle and saving some time and repetition.

On the other hand I did brazenly edit a demo after it was recorded, so you could argue it was fiddled and shouldn't count as a proper demo. (If so it should not count against TGH, who played no part in my "crimes" :P and in particular didn't go into the recording session with any intention of leaving the unsuccessful attempts on the cutting room floor.)

12-19-10 09:49 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

myk
volveré y seré millones

Posts: 14422
Registered: 04-02

RjY said:
The question is, does this sort of thing, in your eyes, count as TAS or not?

If it were your demo and you presented it as an FDA, you'd be cheating, otherwise called "TAS" when presented openly. You could present it as a first exit, implying or mentioning that there were previous attempts, instead, where the hack would be pretty harmless. In the latter case it's not much different than changing the viewpoint in a multiplayer demo or something like that. At most, it could change the time stamp, if that matters.

Qaatar said:
Yes, an objective discussion is certainly possible.

I liked Creaphis' response to this.

Why you cling onto this vanilla vs. pr+ debate, I have no clue.

You might have something to learn, then. I will do my best to help here.

If I was being disingenuous and wanted to call it bullshit anyways, I would have straight up, without equivocation, said 100% bullshit.

Wanting something and being able to are different things. We do what we can.

Again, we already know there are differences, so no, it's not the overriding factor. Hell, it's not even a factor. See above.

What I see clearly is that you and dew avoid competing with Compet~n stuff while at the same time you use an incompatible engine. As I see it from a practical perspective, its a passive-aggressive measure, much like a "strike" or complaint at Compet~n for using "some crappy old engine". My suggestion is to drop that stance and simply record whatever you want with the acknowledgment that the resulting demos are on "another league". At the same time, you can still use the Compet~n as a reference point for competition, it's just that people submitting to Compet~n directly will not really do the same (PrBoom+ includes vanilla, vanilla excludes PrBoom+.) That is, you can beat the times of your old heroes without concern because we've clearly established it's being done on a different wavelength. You are expected to beat many of the old times, more or less, as you're using a somewhat more powerful set of features.

Unfortunately, we're stuck with a very nebulous set of rules and regulations regarding source ports in Boom, and what TAS entails within those confines.

There's nothing nebulous about concrete, material source ports. They're raw things. Boom is one thing, PrBoom+ another. This is just like with vanilla.

As a side note, this is not incompatible with the stance on Zdoom, because the actual gameplay differences in terms of skill requirements between Boom and vanilla are practically non-existent. Between Zdoom and Boom/vanilla, it's almost night and day.

Both PrBoom+ and ZDoom are years away from vanilla in their behavior by default. Enginewise, PrBoom+ is equivalent to vanilla in Doom compatibility mode, while ZDoom is very close in strict vanilla behavior. It's definitely not day and night there. Aside from that, the two advanced engines share user interface features vanilla doesn't have (and many that Boom doesn't have). And, enginewise, Boom itself is generally farther away from vanilla than ZDoom in strict Doom compatibility mode.

If the end result is something that's close to vanilla, then I think that it will be purely incidental.

You can't think of these things without factoring what vanilla can do because it has a heavy historical influence on the demos using its native format. Once you do consider vanilla you can rethink demos according to PrBoom+, as well.

I've always considered what you argue, but I'm afraid that we're not even on the same page in regards to this debate.

That's right, although I do know what page you're on. This is why I'm noting the differences in the pages we're on.

Don't get me wrong, of course I would like to use those features and think that they're not TAS.

Of course you don't, and you shouldn't, as you use PrBoom+ extensively. Perhaps you'd call them TAS in the sense that I described if you were a regular vanilla user. Terms take a different light depending on their context.

But this is the problem: I don't want a gross misrepresentation of my skill level.

The first place that would happen is if someone felt your demos were being compared to vanilla demos. With other PrBoom+ demos, the controversy is much lesser, if any, as most people would expect you to simply take advantage of its features, short of, supposedly, the ones historically marked as TAS by everybody (slow motion, saves, sr50 on turns).

I want to be as transparent as possible, and to make my demos as organic and genuine as possible, within limits, and to be consistent across all types (vanilla and Boom).

If that really matters to you, you could include your CFG and technical specs much like I do in my text files.

I think we all use common sense in this regard.

People tend to do what they are in the habit of doing, which is a mix of laziness and necessity, modified by occasional arguments and people pointing out a lack of proper info. It's cool to see some of you post your specs, but the data you post today can change tomorrow or may have not applied previously, if we don't forget about it.

Most of us will probably be okay if a certain feature allows someone to enhance their skill level by, say, 1% (for example, maybe screen resolution for Artem...he's obviously just as good of a player with choco).

There's little point explaining what I already know and stated expressly in a more ample manner: If they use PrBoom+ regularly, yeah, they'll be okay if one of its features is used without mention, if not, maybe not.

We won't be okay if a certain feature allows that person to record something that wouldn't otherwise be possible (slow-motion, segmentation, or even the monster counter in some instances).

It's all possible, just more or less possible. "Blatant" TAS saves time like hi res and smart totals can, just more time. This is why PrBoom+ users will set TAS apart, and why a vanilla guy will set PrBoom+ apart.

I believe that personal opinions of every player regarding the features of source ports are heavily influenced by the overall general perspective.

I'm not a fan of mass thinking. We have people with different interests here and the idea is to be inclusive, so they can each participate with their preferred modes of play, which may not be entirely compatible. Basically you're saying 90% of people use PrBoom+ so that's what really counts and that others aren't as relevant and can perhaps be ignored. Please use that only on a personal basis and with due respect to others, because if that mentality reigned speed runners should have had to go fuck off and die a long time ago, because we are a minority in the community. Just like we have a space here in general, each different person and group in the speed running community has a place here in turn.

12-19-10 21:33 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

Qaatar
Member

Posts: 385
Registered: 06-09

myk said:

That's right, although I do know what page you're on. This is why I'm noting the differences in the pages we're on.
If that really matters to you, you could include your CFG and technical specs much like I do in my text files.

I'm not a fan of mass thinking. We have people with different interests here and the idea is to be inclusive, so they can each participate with their preferred modes of play, which may not be entirely compatible. Basically you're saying 90% of people use PrBoom+ so that's what really counts and that others aren't as relevant and can perhaps be ignored. Please use that only on a personal basis and with due respect to others, because if that mentality reigned speed runners should have had to go fuck off and die a long time ago, because we are a minority in the community. Just like we have a space here in general, each different person and group in the speed running community has a place here in turn.

- Yes, that's what I'm planning on doing from now on, although I have not successfully recorded anything since my "revelation."

- lolwut? (sorry for the trolling vernacular, but this is ridiculous). That's totally erroneous, baseless, and tangential extrapolation. Perhaps you're not expressing yourself clearly, but I'm reading that as a rather disingenuous effort on your part. No one said anything about irrelevancy, being ignored, etc. I value vanilla demos just as much as the rest, if not more (although perhaps not to your level of admiration). I value the workmanship, respect the conditions under which it is recorded, and many other things about vanilla. Almost all of my Dooming heroes recorded strictly under vanilla, and I would never "defile" any of their records. Unfortunately, all of those things do not change the fact that pr+ is the dominant port being used today...and what I said previously follows from that fact.

Anyways, this, as always, is turning into a semantical debate with you, in which we take turns expressing our ideas in more and more different and...strange ways. You can have the last word, as usual. :)

Creaphis said:
If you don't like that, whoever you are, then feel free to codify your own standard which allows only a specific subset of PrBoom+'s features.

Exactly. I was codifying my own standard, and was curious about the standards of others. I think some were assuming that I was on the attack (thus strawmanning me, either intellectually or otherwise), or that I was trying to establish a unified standard. No...I was only curious for my own good.

Last edited by Qaatar on 12-20-10 at 00:23

12-20-10 00:07 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

Paska
Junior Member

Posts: 242
Registered: 07-09

Belial said:

Which works the same as ignoring possible lost monsters while playing vanilla. It cuts down the time required to record a demo of satisfactory quality, but doesn't affect said quality for the final product.

True TAS features are impossible to mimic in vanilla.

It indeed does affect the final product. Here is little example:
Let's say one aims for 1:59 max time, map doesn't matter as long as the monster counter helps a little bit in this map. So after 6 hours of work 1:59 achieved without using monster counter.

With monster counter one can achieve the time in 4 hours of work and if keeps working 2 hours more he gets 1:58.

Just an example. I'm not exactly sure what you meant but I think this points out something.

With save/loads one does not affect the final product either, it "just saves time" because after 100 (or so) times more work one will achieve the same as using few saves.

Last edited by Paska on 12-20-10 at 00:44

12-20-10 00:37 #

Profile || Blog || PM || Search || Add Buddy

IP || Edit/Delete || Quote

dew
Feral Kid

Posts: 1702
Registered: 05-08

myk said:
What I see clearly is that you and dew avoid competing with Compet~n stuff while at the same time you use an incompatible engine.

well, duh. i avoid c-n because of the incompatible engine, not just coincidentally at the same time.

As I see it from a practical perspective, its a passive-aggressive measure, much like a "strike" or complaint at Compet~n for using "some crappy old engine".

it isn't a "strike" by any means since i try to not make ripples, which i would totally be doing if i was protesting against something.

My suggestion is to drop that stance and simply record whatever you want with the acknowledgment that the resulting demos are on "another league".

tssssk. this advice is entirely useless to me. i already recorded a decent list of prboom+ demos for czech-n on doom2.wad, none of them compatible with c-n or pdang. i acknowledged the result by posting them in that another league and not here.

12-20-10 03:06 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

kimo_xvirus
Junior Member

Posts: 211
Registered: 03-08

I don't think High Resolution matters much. If anyone beats a vanilla record with prboom+ it's a well deserved record.
A simple test you can do is launch epic2 map01 and from the start location aim at the wall and try guessing what piece of wood you'll shoot in high and low resolution then try it from different locations.
It will be just as hard in both resolutions.
Although when high resolution sometimes had more visible outlines for the pieces of the wood you might call it slightly easier.

Most important is to mention what video settings you use, I assume all PrBoom+ demos that don't mention it are recorded in software renderer, high resolution and uncapped framerate and when mentioning GlBoom-plus in txt, I assume they are recorded in OpenGL renderer, high resolution and uncapped framerate

Also I'd like to clear some differences between prboom+ and vanilla with tools, some were never discussed before

Vanilla
180 turn only with "9" sensitivity (dms015)
able to turn vertical sensitvity on and off during game (dms015)
unable to walk while using the always run hack (joyb_speed 29)
menu pausing desyncs demo
**can use cheats like IDDT and IDBEHOLDL during game (they don't get recorded in demo)

PrBoom+
180 turn with any sensitvity
unable to turn vertical sensitivity on and off during game
able to walk by pressing the run key when having autorun
**you can menu pause

** imo should be considered cheating, the vanilla one is obvious and the PrBoom menu pausing was stated by dew in The Two Types of Glides thread.
If you want to pause in the middle of the map (which I don't like), use the pause break key not the menu

and also to clarify about that 180 turn in vanilla thing now that I bothered to try The Doom Mouse Spinner.
The Doom Mouse Spinner txt says

code:
IMPORTANT!!!!!!
     Make sure you maximize the mouse sensitivity setting in Doom.
Anything less than full sensitivity will mean you'll turn less than 180
degrees.

the max sensitivity he talks about happens to be 9 which is the maximum possible to set from the doom menu, any other value makes the key you set turn something other than 180 degrees (varies). So it would be unfriendly for vanilla mouse users.
First should we ban the "other than 180 degress" turns caused by having different sensitivity than 9?
Second how should mousers use turn 180 degrees while using their normal sens in vanilla?

Giving All I said
My PrBoom+ demos are all recorded in software renderer, high resolution (640x480) and uncapped framerate and no prboom hud
except my old dmonfear and 2002ado runs where I mentioned in txt I used 320x200 resolution and capped framerate
I used the PrBoom advantage of being able to walk while having autorun in these 2 runs p230-113 and d201-221 which I didn't mention in their txts.
I switched to the prboom hud with k/i/s counter only at the end of these runs gra1-622, gra2-421, ua09-721 and ua10-1210 which I mentioned in all their txts except gra2 and ua10.
I will supply all these information for any future demo I make

Maybe we should make a new sticky for PrBoom+ Rules, explaining all the prboom+ features and config variables to determine a fair standard. Also So that someone who never tried the original engine doesn't think that something like "Change palette on pain" option was in the original game.

Last edited by kimo_xvirus on 12-20-10 at 18:52

12-20-10 15:34 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

vdgg
Moderator

Posts: 872
Registered: 08-07

I very much agree with what Kimo said (and it's cool he mentioned all the technical stuff I wasn't aware of). In particular,

kimo_xvirus said:
I don't think High Resolution matters much. If anyone beats a vanilla record with prboom+ it's a well deserved record.

From my POV, the orphaned Compet-N transposed into something like a "club of players who like classical WADs and don't mind using vanilla executable". There are 16 active players now ("active" = those who submitted at least 1 demo in 2010). I obviously can't speak in the name of everyone, but I feel at least some of us think the same way - we are a minority and we cannot ignore excellent runs performed by the majority. For example, Xit Vono held a UV Speed record for Requiem MAP03 (1:29). Then skepticist recorded his non-vanilla glide trick demo (0:53), after which Xit responded with his 0:52, saying "one second faster than Vladimir". This may be very off-topic, just wanted myk to be aware of that :)

12-20-10 18:36 #

Profile || Blog || PM || Email || Search || Add Buddy

IP || Edit/Delete || Quote

tatsurd-cacocaco
Member

Posts: 456
Registered: 07-08

kimo_xvirus said:
If anyone beats a vanilla record with prboom+ it's a well deserved record.

Yes, you are right. Although people sometimes discuss the difference between prboom and vanilla, a lot of people don't look on such demos with a prejudiced eye. New records have new routes, new strategy, and so on. They mean the prosperity of speedrunning. We shouldn't stop them. Compet-N records are official records, but I personally feel that a c-n record which is beaten by a non c-n player are no value as a record. I'm not saying that I neglect old records. I like to watch old records and I'm interested in the whole story of each map with each category.

However, I think menu pause in prboom is OK. Certainly pausing without leaving pause tics is impossible in vanilla doom, but pausing itself is possible, unlike HUD, and isn't banned in c-n. Pause tics just irritate watchers. If we need to clarify that you use pause or not, I think that it is enough only by mention in a text. My runs on vanilla doom are rarely hindered by calls or something, and I really hate it. Impossible menu pose seems to be a tricky behavior of vanilla doom, like VPO. People shouldn't stop recording demos or decrease interest by leaving pause tics when unexpected situations happen.

EDIT: Of course, the change of the setting for glides or something is definite cheating.

EDIT2: I fixed spellings.

Last edited by tatsurd-cacocaco on 12-21-10 at 13:00

12-21-10 10:53 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

4mer
Junior Member

Posts: 198
Registered: 08-10

T-C It's pause not pose :)

12-21-10 12:01 #

Profile || Blog || PM || Search || Add Buddy

IP || Edit/Delete || Quote

myk
volveré y seré millones

Posts: 14422
Registered: 04-02

Qaatar said:
That's totally erroneous, baseless, and tangential extrapolation. Perhaps you're not expressing yourself clearly, but I'm reading that as a rather disingenuous effort on your part. No one said anything about irrelevancy, being ignored, etc.

It not baseless if you've been following the conversation, including especially your attempt to degrade my points by dragging our argument down as if it were "bullshit" to eliminate a branch of the topic. Later when dew posted something along your line of thought you added "oh finally someone posts something worthwhile." Degrade the whole argument when you are at a loss and then salvage your side or angle at the first opportunity. Add to that when you ignore my points and reasonings and only respond to what you feel is an accusation. I can't see how your valued "general perspective" could have ended up including things being rejected in that form of denial.

Almost all of my Dooming heroes recorded strictly under vanilla, and I would never "defile" any of their records.

I heard you, but that reeks of retroactive reasoning. Part of my point is that it would not have happened without the split caused by the different engines. You said yourself you didn't want to be taken as a cheater, and beating their times with PrBoom+ would have been considered the equivalent of doing what is not allowed, but openly, much like TAS is to cheating, from a "Compet~n" perspective. My response is based on your reasonings, but not necessarily the ones you chose in arguments such as this one, but where you just spoke of your habits and reactions to facts and events, or reacted to these aspects more spontaneously, past that shell your ego generates.

Unfortunately, all of those things do not change the fact that pr+ is the dominant port being used today...and what I said previously follows from that fact.

The fact that a good number of people use it here wasn't being debated. Our discussion, based on what I highlighted and we debated, was about how to take it in regard to vanilla.

I think some were assuming that I was on the attack (thus strawmanning me, either intellectually or otherwise),

That sounds rather oblique, but should I be surprised? You weren't "on the attack" so much as being in that nerdy form of defensiveness you get into so easily with your linear mode of thinking. "OMG what's this myk doing by insisting so much about the difference of vanilla, he's crazy, let's argue something back!"

or that I was trying to establish a unified standard.

Consciously or not, in most of your responses to my posts marking vanilla aside you were effectively minimizing those differences and this leads toward irrelevance, where the rest remains. So it ended up being and ambiguous "we agree there's a difference but we don't." You could have made this discussion much shorter by saying something like "ok myk, you have a point, from a vanilla-oriented perspective, vanilla's (user-side) standard is quite strict and exclusive when compared to everything PrBoom+ adds. Here, on the other hand, I'm just trying to make the best use of PrBoom+, on another level to what vanilla and Compet~n offer." We were close to something like that but adding your mental defensiveness to my intuitive frankness really causes long winding roads. Perhaps it unveils a lot, though.

kimo_xvirus said:
I don't think High Resolution matters much. If anyone beats a vanilla record with prboom+ it's a well deserved record.

It certainly is a well-deserved PrBoom+ record. Anyone using PrBoom+ can also take any vanilla record as a record, while a vanilla user can be more skeptical the other way around, yet obviously taking it as a reference and by all means checking and acknowledging things like new routes or evident skill.

Vanilla
able to turn vertical sensitvity on and off during game (dms015)

I'd be surprised if Andrey doesn't add this option to PrBoom+, if players start using it. He generally makes sure PrBoom+ can do any relevant thing vanilla or Boom can manage. It has become relevant especially lately due to glides, so it's no surprise if nobody cared or noticed previously.

PrBoom+
**you can menu pause

You can "window pause" in vanilla, on Windows 98, which is similar but a bit less precise because screen drawing time can eat a little bit of the time the active game time.

Second how should mousers use turn 180 degrees while using their normal sens in vanilla?

By flicking the mouse and being very steady-handed :p

Yeah, it isn't as accurate, although it creates a nice contrast between a keyboarder using the spinner and a mouse user. I used to use 9 sensitivity with mouse acceleration, but I didn't use the spinner with that. Before becoming a mouser I had used the spinner, though. It also had a bug that occasionally made you turn indiscriminately until you moved in the direction the mouse started turning to abnormally. It's true it could be of some use to make very accurate turns in some types of runs (glides?), but it's mainly a feature for keyboarders. The spinner itself was mentioned mostly for its vertical sensitivity toggling.

Maybe we should make a new sticky for PrBoom+ Rules, explaining all the prboom+ features and config variables to determine a fair standard.

Rules or just descriptive guidelines? It's really hard to keep such rules consistent. Hardly any such agreements were used for vanilla; only that it shouldn't be hacked, that 3rd party effects could be questionable, but all these things are part of PrBoom+ itself. Why not just allow people to use all the minor features, but encourage them to be informative about what they use?

Also So that someone who never tried the original engine doesn't think that something like "Change palette on pain" option was in the original game.

There you have a hack that is possible in vanilla but only through a WAD (patch, hack). It's also one of those changes that can be advantageous due to visibility but also a disadvantage in the sense that the red screen is a "HUD" effect telling you you got hit and how much.

vdgg said:
From my POV, the orphaned Compet-N transposed into something like a "club of players who like classical WADs and don't mind using vanilla executable". There are 16 active players now ("active" = those who submitted at least 1 demo in 2010). I obviously can't speak in the name of everyone, but I feel at least some of us think the same way - we are a minority and we cannot ignore excellent runs performed by the majority.

But who spoke of ignoring, aside from me and Boom (and especially because I don't have time to check every demo and thus prioritize by technical standards)? My argument is that to let PrBoom+ demos bloom, they should be treated as such. Otherwise there's always that "oh wait, this was recorded with a modified executable allowing further settings, unlike those." Continuing a vague "it's about the same" causes uncertainty and friction. Active players may, for instance, vary in how closely they follow vanilla standards on PrBoom+ depending on how much they use vanilla as well, or how much of a vanilla history they have behind them. Other than that, people will tend to use PrBoom+ more freely. I'm not referring to ignoring demos but to sorting out the differences so that each type can thrive, with feedback.

For example, Xit Vono held a UV Speed record for Requiem MAP03 (1:29). Then skepticist recorded his non-vanilla glide trick demo (0:53), after which Xit responded with his 0:52, saying "one second faster than Vladimir". This may be very off-topic, just wanted myk to be aware of that :)

I'm surprised you thought I wasn't aware :p

In that case, once xit manages to beat skepticist's demo he has the following in favor: a Compet~n record going by well-defined rules (dictated mainly by a simple executable), a "global non-TAS" record and beating a better executable with a crappier one. It's a good example to illustrate the first thing I said to kimo above...

12-21-10 23:43 #

Profile || Blog || PM || Email || Homepage || Search || Add Buddy

IP || Edit/Delete || Quote

All times are GMT. The time now is 10:21.

Pages (2): « 1 [2]

Doomworld Forums > Special Interest > Doom Speed Demos > What constitutes TAS? [split from 2 glides]

Show Printable Version |

Email this Page |

Subscribe to this Thread

Forum Rules:

You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

HTML code is OFF
vB code is ON
Smilies are OFF
[IMG] code is ON

< Contact Us - Doomworld >