Posts for Max]I[muS-X

Experienced Forum User
Joined: 3/28/2006
Posts: 14
It's true, there are a lot better Q3 vids out there made without scripts and bots. Tricking iT 2 comes to mind. However, the reason I posted this here was because of the potential that this has. When I watched that movie Anti-Conformist, and it's subsequent disappointment, I couldn't help wondering what kind of video would be possible if the same tools were in the hands of the people here at TASVideos. I guess really it's just a matter of getting those tools, which would probably be the hardest part.
Post subject: Q3 TAS'ing (sort of)
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Hey, this is an old video, and I first saw it a long time ago, but the thought occurred to me today that maybe one of you guys would want to play around with this. http://www.youtube.com/watch?v=W9YLOztR35c Basically a guy named TChouky found a way to script the bots in the game to do whatever he wanted, at certain intervals of time, and by doing so he could perfect tricks, runs, etc. This wasn't just limited to the bots but to his own character. I don't know how difficult this is, but I thought it was a neat idea and had good potential. However, he recently came out with a vid called Anti-Conformist, which was supposed to be the same concept but with a frag video, simulating insane "frags" or series of frags that would otherwise be all but impossible for a human to perform. I thought this would be amazing to see! However, the frags were poorly synchronized, and the situations weren't very believable or exciting. So anyway, this led me to at least post this for people here, and see if anybody wanted to take a stab at mastering this. There are certain maps they play, where the goal is to get from the beginning to the end as quickly as possible using any combination of these jumps, which fits the speedrun bill. I think it's also a pretty cool way of extending the TAS perfection feel to something more modern, like Q3 (old though it may be). The only issues are: I don't think he will tell you how he does this, because at the end of the video (or maybe it's at the end of the article showcasing the video, I forget) he says "don't ask" for the mod to do this. So that would mean somebody that knows something about programming or scripting would have to figure it out, which probably would be the most difficult thing. Then it would just be a matter of finding somebody here willing to do it. So, what are your thoughts?
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Do you even have any clue what this discussion is about? I'm not attacking TAS's, so you can re-read my posts and stop pretending that I am. I have no problem with advancing technology, or any of the things you mentioned which for some illogical reason or another you assumed I did. You're entire post is irrelevant. As for Graveworm, I'm not defending myself, I'm not even the issue at hand. What did you do, skim every post and assume I'm attacking anyone or anyTHING?
Also, a person who can not be proud of what he or she has done without having it confirmed by someone else really should try to start thinking a bit different.
We are talking about video games, and I've already advanced reasons why they would have that view. You ignored them and instead just stated your opinion, minus any sort of reason stating why you believe so. I'm not even sure you read the entire thread.
For me, TASing Super Metroid is 10 times more nerve-wrecking and harder, in gameplay ways than making a console speed run-through. I've never gotten tired of console running, but I've stopped working on a movie like 30 times in 4 months.
Ugh, it's not "HARDER" to MAKE the run. The frustration you are referring to is an element of patience, sure, but the only skillset you are acquiring is with the emulator, not the game. The whole point of using slowmo and small segments and everything else is that you are making it EASIER to do these things. You are breaking them down frame by frame almost, and you do this because it makes it possible to produce the optimum final result. Do you find yourself skilled at a game when you replay a fraction of a second of gameplay 200 times, as somebody else used as an example? Or is it simply an effort in tedium to produce a fantastic end result? I like TAS's, as I've said many a time before. But please realize their strengths AND weaknesses, and don't just try to argue in their favor for EVERY facet of discussion, just because they mean a lot to you. This wasn't an attack on TAS in the slightest, it was putting forward a reason why someone would lose interest in making a normal speedrun when there is a perfect TAS of a game out there already. Yes, there ARE valid reasons for these feelings, whether or not you choose to accept it.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
KDR_11k wrote:
Must I mention that some people use way over 200 rerecords for segments of less than a second sometimes? Especially when luck manipulation is involved. While the segment size is smaller there's also much higher perfection required, you're not going to redo a segment in a normal speedrun just because you lost a few frames (at least not in runs as complex as a Super Metroid speedrun).
No, you didn't have to mention that, as I already addressed that aspect.
"Max wrote:
I[muS-X"]Sure, the number of "attempts" in the TAS to equal the same amount of attempts in the time of a regular run's segment may be comparable, but remember that after every successful segment in a TAS, your progress is saved, and you can continue building the run. With a normal speed-run, you must repeatedly start over until you literally become better at the game, and are good enough to accomplish all tricks in the segment to a satisfactory degree of success. This is what I meant by "harder".
Perhaps the first sentence could have been worded better, but it's the same general concept. I should have also added "ultra-tiny" between "successful" and "segment", as it better outlines exactly why I believe creating a TAS is just less nerve-wracking than making a normal speedrun. Just to clarify where all of this is coming from, this goes back to the argument that normal speedruns are created almost exclusively with the knowledge in mind that others will view it, which is why if there's a "perfect" speedrun floating about it takes away from the fun of planning as well as the drive to create a worse looking, normal speedrun. Which is what was referenced in LeCorrier's quote, which was in the main article of the thread, which some people made comments showing lack of understanding for this point of view.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Internet discussions always tend to veer off course just a little...
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Both require mental effort and planning...
Simply having the imagination to think of the fastest way possible through a level, and discovering glitches, are indeed problem-solving skills and puzzles I most enjoy, but I think most people can accomplish this to a degree of similarity.
In this part of the argument I was mainly talking about the difficulty of creating a run, not planning. I acknowledged both the time it takes to make the run, as well as the creativity and problem solving abilities that it takes to plan it. However, these things are also required in non-TAS runs, so it's not really much of a point to compare differences. I think the main difference would be that planning is much more important in a TAS, but barring the discovery of a new glitch, the differences of times between capable runners are (or perhaps should be) a matter of seconds, if not fractions of a second.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Making a TAS is tedious and requires a lot of patience I'm sure, but you don't really think it's difficult to make a run frame by frame in slow motion do you? How many attempts does it take you to make one segment in a TAS? Hopefully not more than one or two. In regular runs, where the segments are much longer and full speed, it takes hundreds. Sure, the number of "attempts" in the TAS to equal the same amount of attempts in the time of a regular run's segment may be comparable, but remember that after every successful segment in a TAS, your progress is saved, and you can continue building the run. With a normal speed-run, you must repeatedly start over until you literally become better at the game, and are good enough to accomplish all tricks in the segment to a satisfactory degree of success. This is what I meant by "harder". Making a TAS can be self-rewarding, I never said it couldn't. To a degree, the same can be said for regular speedrunning, if it weren't for the process involved. As I've compared it to before, making a TAS is sort of like making a painting. After each careful, deliberate stroke your progress is saved and you're free to add to it at your leisure. People do make paintings for themselves, without intending to show anyone; they do it for the beauty of the final creation, and the fact that they created it. I could see someone working to make a TAS just for themselves. Making a regular speedrun, on the other hand, is not the same process. I don't have a nifty analogy to compare this one as of yet, but suffice it to say that running the same segment over and over and over and over again is much less fulfilling and I think not quite as "worth it" to most. It requires mastery and memorization of a video game to sometimes ridiculous degrees (if you consider where a video game stands in life), and the final result is less than how you saw it being perfected in your mind. Most of the fun of it is showing your creation and accomplishment to others, especially when they have to consider what it took to make it. With nobody to show it to, why give yourself so much grief? Finally, since you compared the self-rewarding feeling of completing a TAS with the self-rewarding feeling of a huge physical feat, I have to say this. Yes, they both give you self-rewarding feelings to varying degrees, but which is more meaningful? The reward should match the work put into it, and importance of the subject to the individual. Making a TAS requires patience but come on, comparing it to walking across Antarctica? It's not even in the same ballpark of accomplishments, not even close. Is making a TAS one of the highlights of your life? I hope not.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Well, my point was really to show that beating a video game should never be on the same level as some of the examples they were trying to compare it with. I agree we are all different and have different tastes, and accomplishments of the mind are pretty rewarding too. I DO think however, that a physical accomplishment will be far more rewarding than a mental one in the long run, for just about anybody. When I think back to the best memories in my life, they weren't things I thought about, they were things I DID. I'm majoring in Game Programming and I like to be creative and create things with programming also, but they don't provide me with the happiest of my memories. I also disagree that TAS's are more impressive than runs done by hand. Both require mental effort and planning, but the regular runner has to take into account his abilities as well, and then work harder for the final product. Simply having the imagination to think of the fastest way possible through a level, and discovering glitches, are indeed problem-solving skills and puzzles I most enjoy, but I think most people can accomplish this to a degree of similarity. /offtopic
Experienced Forum User
Joined: 3/28/2006
Posts: 14
You can't compare the satisfaction of achieving a great human feat like walking to the South Pole and beating a video game in x amount of time... Yes, SOME people will climb a mountain or walk to the South Pole just to prove to themselves that they can do it. But this is a PHYSICAL endeavor, and it is very rewarding emotionally. It pushes your body to it's limit. You are experiencing and accomplishing things in the environment your body and mind are "designed" to appreciate. I don't think I should have to break down exactly how doing something extremely physical is much more "reality" grounded and is therefore more self-rewarding than beating some video game quickly. For example, I don't think you'll meet many people who's life's goal is to beat Super Mario Brothers 3 in 10 minutes, regardless of who sees it. That would be pretty sad.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Unless it's for very special personal reasons, which is usually not associated with a video game, people do not usually run games only to prove it to themselves. They do it for the shared enjoyment and reactions of others viewing their time, effort, and final product. Would you spend time running a game by hand over and over to the point of madness if nobody was going to see it? That said, when there is a video of "perfect" play of a game floating around, it tends to discourage people from bothering to make their own run, especially if that video is popular. Since TAS runs are all about planning and glitches/tricks, all the fun of finding the nuances of the game is gone, and working so hard on something not many people will enjoy (outside of say, the SDA forums) is usually not worth it to most. I enjoy watching TAS runs, I have to say. I like to see a game being demolished, and they appeal to me as a perfectionist. It's very fulfilling to watch one of your favorite games played to the point of ridiculousness. However, it helps that TAS runs are stuck in the realm of N64 and below. If I had seen a TAS run of HL2, I would have enjoyed it, but I would never have bothered making one. The idea just wouldn't have been appealing to me.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Well, you'll have to ask RandomEngy or hope he reads your response, because he did the capturing/video editing etc. :)
Experienced Forum User
Joined: 3/28/2006
Posts: 14
Oh, and I just fixed the broken links. Sorry about that.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
We posted it in a number of sites, you guys are just one of them :P I posted here to share the video, not because I considered this tool assistance or because I wanted you to host the video. I wouldn't even dream of putting it in the same category, because we do not use anything close to slomo, 1000's of savestates, or anything of the sort, and it took much too much effort and practice to make these runs. On top of that, none of the tricks or glitches require the script. The script itself merely utilizes a loop function. Scripts have been accepted before, and word is it will be accepted at SDA anyway. I'm not discrediting what you guys do, in fact I admire the hard work and dedication you put into making everything pixel perfect. To me this site is more about hard work to make a finished masterpiece, rather than a show of skill, and I respect that, which is why I thought of posting this here. But I also don't consider the run we did in the same category at all.
Experienced Forum User
Joined: 3/28/2006
Posts: 14
MOD EDIT: hijacking this OP to reflect the brand new HL2 Done Quicker run. <3 samsara https://sourceruns.org/speedruns/half-life-2-done-quicker/ --- ORIGINAL OP BELOW --- Hey, we just finished our project over at www.hl2dq.net. Not sure if anyone here has heard about it, but figured I'd let the good guys at nesvideos take a look. We worked a long time on making this, over a year in fact, and a lot of hard work and effort was put into it. If you have any questions about it, the download comes with Comments, Questions Guide, and Tricks text files, and of course I'll be checking back on this thread. Also, no cvars or cheat commands were used in the making of this run. Right now our main download links are torrents, and we've had people report some real nice speeds on it. Don't use the Google Video link to watch it unless you're desperate, as the quality is butchered thanks to Google recompressing it, and it lacks the text files that accompany the video. Torrent Downloads (Insane,High,Medium): http://24.17.254.227:6969/ Game Arena (High Quality): http://www.gamearena.com.au/files/details/html/halflife-2-done-quick-1h35min Planet Videos (Medium Quality): http://nato.daddeln.de/pv/index.php?nav=files&action=detail&id=7264 Google Video (Low Quality, streaming): http://video.google.com/videoplay?docid=-7597745302270855580 Introduction to the run from the Comments.txt:
Thanks for watching our speedrun. We had a lot of fun times and painful memories making this, and we're glad we have an audience to see it. In case you think we timed our jumps too perfectly, know that this speed-run was NOT focused on timing jumps. For bunnyhopping, we used a simple macro which presses the button assigned to "jump" repeatedly, for as long as it is held in. This was done to increase focus on less tedious and more complicated tricks in the run, as well as to prevent people from doing silly things like making an in-game script with hundreds of "+jump" commands and binding jump to the mousewheel as well as 12 buttons on a joystick to stomp on. In case you're wondering, yes, the above method has actually been used before the macro was voted as an option. This run is also not perfect. We have spent a long time trying to cut down on seconds throughout various maps, but some do have their weaknesses. Sometimes we may run into a pole, walk when we could have sprinted, or miss a teleport or two. Nevertheless, we spent a large amount of time finding revolutionary shortcuts and shaving off seconds, so it should at least be quite an undertaking to improve on this run significantly. For this run, we began timing the moment we were able to move as per SDA rules, and stopped the moment we lost control. These times were the tram opening and the explosion of the portal, respectively. We clocked in for a total time of ~1:36:50, with an SDA time of ~1:37:03. We finished with 26 non-traditional segments, most of them due to the random and difficult nature of their content, for a total of 93 segments. For every level transition we included a special _2.dem demo which is automatically generated while recording during a map change, to help bridge the gap between loading a new level and the autosave which occurs a few seconds afterward. However, they do not count as their own segment as they were not necessary and were only included for video continuity. If you have any questions about how we performed the tricks shown in the video, read the HL2DQ Tricks text as well as the HL2DQ Questions Guide text.