1. Hey Guest, looking for Virtua Fighter 5: Ultimate Showdown content? Rest assured that the game is identical to Virtua Fighter 5: Final Showdown so all current resources on here such as Command Lists with frame data, Combo Lists and the Wiki still apply. However, you can expect some VF5US specific changes to come soon!
    Dismiss Notice

VF4 in my House NOW HAHAHAAHAHAHAHAHHAHA

Discussion in 'Console' started by Prince_Hatim, Jan 27, 2002.

  1. ice-9

    ice-9 Well-Known Member

    Re: [various things] per second

    For VF I imagine there's some sort of table of moves that specifies the time taken by the various phases of each move. I would guess that this is expressed in time, not numbers of frames.

    I personally think it's expressed in frames, not time. Take a game with slowdown for example. How is the engine supposed to cope with that? It would totally muck up the game if some animation was tied to external time, i.e. "real time," as in real life seconds. Whereas if it's by frames, even if there's slowdown, within the game engine, internal time as measured by frames remains consistent.

    It should be noted that the display is always *at least* one frame behind the simulation. One frame is a somewhat idealistic best-case. It's not uncommon for games to be two, three or even four frames behind. If the display is four frames behind the sim this means that four simulation loops will occur before the result of user input is seen onscreen.

    In VF, the lag time is 1 frame. I don't know about Tekken, but the reason such a ruckus over Itagaki's inflammatory comments were raised was because Suzuki gave Itagaki the algorithm that reduces the lag to just 1 frame.

    Since the likelihood of this lag being longer increases as less powerful hardware tries to cope with a similar load to more powerful hardware (e.g., VF3 on DC vs. Naomi), it is possible that a *slight* feeling of sluggishness is thereby introduced.

    If I read you correctly, this suggests that the DC/PS2 versions would feel slower than their respective arcade versions (if you buy the argument that Naomi > DC and Naomi 2 > PS2).
     
  2. Fishie

    Fishie Well-Known Member

    UK Guy , he already admitted twas a pirate version ya twit .
     
  3. gribbly

    gribbly Well-Known Member

    Re: [various things] per second

    I personally think it's expressed in frames, not time. Take a game with slowdown for example. How is the engine supposed to cope with that?

    Again you need to distinguish between simulation frames and display frames. These days most slowdown is relatively benign "frame dropping". Say a game is simulated at 60Hz and usually displays at 30Hz (not at all uncommon on a console like the N64). When the going gets graphically intense, the game might temporarily display at 20 or 15Hz ("slowdown").

    But if the game is properly implemented it will still be simulating at 60FPS. So those portions of gameplay attributable purely to simulation rate (e.g., collision detection, AI, and -- especially pertinent to this discussion -- animation playback, etc.) are unaffected.

    I think it was Mr. Bungle who used a film projector metaphor. In the case of an independent renderer/game logic model there is, in effect, a way of magically discarding frames so that the action proceeds at a constant rate no matter what the actual (display) frame rate is.

    It would totally muck up the game if some animation was tied to external time, i.e. "real time," as in real life seconds

    Not at all, as long as the simulation rate remains constant. In a game like Quake, for example, velocities for things like rockets, etc., are expressed in distance/time, not distance/frame. This is because a display frame is of an unknown length, so if you tied rocket velocity to it you'd experience variable perceived rocket velocity.

    A rocket could be more safely tied to simulation frames, which *should* remain constant. But in practice it's more usual to express such things in time since:

    a) it's more natural. Meters/second is very understandable, meters/simulation frame is not.
    b) it's more robust. Sometimes the number of simulation frames per second is changed during the course of a project (perhaps you overestimated how many processor cycles you had). If you've expressed everything in absolute times you don't need to re-config your game. There may be examples of games that change simulation rate dynamically, depending on circumstances (I can't think of an actual case of this). Also, different parts of a game might be simulating at different rates (e.g., the sound engine might only update at 10Hz to save bus bandwidth).

    Whereas if it's by frames, even if there's slowdown, within the game engine, internal time as measured by frames remains consistent.

    If you mean display times, then what you've said is exactly backwards.

    If you mean simulation frames, then why not say "internal time as measured by *time* remains consistent", since that's the point of keeping the renderer seperate from the simulation.

    In VF, the lag time is 1 frame.

    Have you got a source on this? Not because I doubt you -- it's entirely possible -- but because I'd like to know more about the implementation of VF in general.

    If I read you correctly, this suggests that the DC/PS2 versions would feel slower than their respective arcade versions (if you buy the argument that Naomi > DC and Naomi 2 > PS2).

    Well I definitely think a case can be made that N2 > PS2, but of course it depends on which metrics you choose =]

    In any case that last point (about the lag in the rendering pipeline) was pure speculation. I haven't really studied this, and it's possible that four frames is entirely unnoticeable. The game would basically play at the same speed, it would just be sort of offset a little =]

    grib.
     
  4. Mr. Bungle

    Mr. Bungle Well-Known Member

    > Well, it's a new concept, buying stuff, but perhaps it actually works

    larf. i own all the boards (save for 2.1), myself. i just don't believe that you do.

    > Heck, it's a hell of a better conversion than Vf4

    righty-o...you expect to believe anything you say when you make this sort of statement when you haven't even played the fucking thing? you're so full of shit, boy.
     
  5. ice-9

    ice-9 Well-Known Member

    Re: [various things] per second

    Ah, OK, I think I see what you're saying. Your explanation makes sense in the case of PC first person shooters, otherwise how would you compensate for the difference in FPS between two different machines.

    Still, however, in VF all indications point to frames as the measurement of internal time, not real time seconds.

    - The environment is stable and consistent, unlike PC/Internet
    - All the strategy guides out there break things down into frames, not time
    - The frame stats mode in VF3/tb (i.e. Training in the DC version) lists frames
    - Most of all, AM2's system documents speak in terms of frames

    As for the 1-frame thing, my original sources came from Neogamer who read it in a Japanese magazine (Famitsu or Gamest, one of the two). However, I have heard it repeated by other people.


    Creed:

    Not to take potshots or anything, but how can you be sure you held down guard for exactly 10 seconds? Stopwatch? Did you get 600 on the nose?

    I just used my watch. I did it three times...got 597, 600, 602.

    If it makes you feel any better, just extend the test. Do it for 100 seconds. You should get around 6000 frames.
     
  6. ice-9

    ice-9 Well-Known Member

    Re: oops

    ADDENDUM:

    Wait, actually it's still possible in the first person shooter example I used to still specify moves in frames, not seconds. Duh, like you said, the display FPS second would be different, but the internal FPS would be the same.

    Anyway, if you think about it, there's not much difference whether it's seconds or frames, since the way VF literature thinks about frames is as a measurement of time anyway, not necessarily animation per se. When a punch is 10 frames, we don't think of it as 10 frames of animation, but 1/6 of a second execution. If the punch execution was sped up to 1/10 of a second, we would just say the punch takes 6 frames to execute. So as you can see, it's basically the same thing.

    In any case, it seems kind of egregarious that AM2 would fiddle around with this stuff at the risk of messing up the balance. And it's not like anyone wants the game to be faster...correct me if I'm wrong, but we all want an arcade-exact experience right? AM2 might as well have spent the time adding more modes or improving the graphics.
     
  7. Rugal

    Rugal Well-Known Member

    larf. i own all the boards (save for 2.1), myself. i just don't believe that you do.

    Why, because you can't read and understand something that's common sense, or because you're stupid?

    righty-o...you expect to believe anything you say when you make this sort of statement when you haven't even played the fucking thing?
    How would you know if I've played it or not, retard?
    You're whole arguement is based around a childish " you don't have it" mentality. Instead of posting dribble, why don't you give a valid opinion as to why you think what you do?
     
  8. gribbly

    gribbly Well-Known Member

    Re: oops

    Anyway, if you think about it, there's not much difference whether it's seconds or frames, since the way VF literature thinks about frames is as a measurement of time anyway, not necessarily animation per se.

    We agree completely! Provided simulation rate is consistent, a simulation frame can be regarded as a unit of time.

    If -- as is entirely possible in a closed environment like an arcade or console game -- an unchanging display rate can be *guaranteed*, then it becomes possible to regard display frames as a unit of time.

    However, my original point (in answer to the original question) was that I regard it as unlikely that the designers would have been thinking in terms of "frames" when designing the game. All the exact timings would have been tweaked and tuned into place, and as a result move x ends up taking y frames. Observing the game later it seems like it makes sense to specify everything in frames, since that's the most obvious unit after the fact.

    I don't know, though. It may be that they planned everything out in numbers of frames from day one, and I'm totally wrong!

    I also doubt that AM2 (or Genki) or whoever would intentionally mess with this stuff. The timing of the moves is critical to the feel and fairness of VF. Surely your number#1 priority would be to preserve it in the port.

    grib.
     
  9. Mr. Bungle

    Mr. Bungle Well-Known Member

    > " you don't have it" mentality

    ha, that's rich. you were the one to say you owned the boards first, as if it somehow validated everything you said regarding your technical explanation, or somehow turn it into "common sense".

    > Why, because you can't read and understand something that's common sense

    your logic is so goddamn fucked:

    1) bungle does not believe that i own the vf3 board
    2) simply owning a vf3 board is the clear path to the technical understanding of game speed and monitor mechanics, which is really just common sense anyway
    3) because bungle does not believe i own the board, he must not understand my explanation - or he can't read, or is just stupid

    > How would you know if I've played it or not, retard

    i do not know, i am making an assumption, based on the fact that not one single post of yours regarding ps2 vf4 is from your perspective; you have not said ANYTHING specific about the game, except that it sucks. all you're capable of doing is using screenshots off of game sites to try and prove your points, and forcing your fucked up technical "knowledge" on others as "common sense". you dodge simple questions, you do not give any useful info at all.

    > why don't you give a valid opinion as to why you think what you do

    i already have, over and over - i have many, many hours on both arcade and home versions (of vf2 as well as vf3), and the speed difference is just simply there, and it's very clear not only to myself but the vast majority of people who have had their say in the matter, quite a few. unless we're all idiots or stark raving mad or have eye trouble, that's all i can say, i won't bother attacking with some idiot technical explanation.
     
  10. Rugal

    Rugal Well-Known Member

    ha, that's rich. you were the one to say you owned the boards first, as if it somehow validated everything you said regarding your technical explanation, or somehow turn it into "common sense".
    I posted the fact that I owned the board in response to your stupid "are you just talking out of your ass again, or have you actually played both versions over a period of time and compared them? " comment.
    Now, it's obvious you're trying real hard to be an arsehole, but hey - give it a break /versus/images/icons/laugh.gif

    your logic is so goddamn fucked
    My logic for the sync of the game is based on common sense. I only mentioned the board when questioned by you as to my source of understanding, and even then it's not relevant to my understanding.
    It doesn't take a bloody genious to understand it, but obviously it's above you, Mr Bungle.
    As mentioned, the only way VF3 could be different from the arcade is *if* they changed the animations in the game, as both titles are synced to 60 frames and have new animations PER frame. Ie, slowmotion mode is enabled by re-drawing frames on animation at a ratio to the degree of slowdown (that is, 1/4 speed is obtained by re-drawing each frame 4 times).
    If you think there's a difference in the animation, then explain how. I don't consider it reasonable to believe that Genki changed the game with 6 months to port it.

    And in regard to frame delay - all VF games (bar 1) have a 1 frame delay. All games use double buffering and while 1 frame is being displayed one is being drawn, so inputs are always 1 frame behind on-screen action. This is the same with the arcade and home version.
     
  11. Mr. Bungle

    Mr. Bungle Well-Known Member

    ehhn...you're right, i was getting a bit creaky there. chalk it up to the "arguing with idiots" factor.

    anyway a project is in the works to prove the speed issue - i hope you and jeff fucking choke on the results. in the meantime, enjoy <a target="_blank" href=http://home.earthlink.net/~foops/crap/log.txt>this</a>; laughter at other people's expense is fun.
     
  12. Rugal

    Rugal Well-Known Member

    ehhn...you're right, i was getting a bit creaky there. chalk it up to the "arguing with idiots" factor.
    It' seems that arguing with an idiot is all that I am doing.

    Here's a note from the VFhome channel transcript: I'll assume RSW is you, as he is the only one dumb enough and rude enough in the channel /versus/images/icons/smile.gif

    rugal insists that because vf3 arcade and vf3 dc both run on
    monitors at 60hz, then the only way for vf3 dc to run faster is to have more frames in animation

    If you read my previous posts in the example of slowmotion playback, you could determine that the only way to make the game run faster would to be to remove frames of animation, not add them. If you add them, then playback takes longer, and it slows the game.

    second part of the transcript:
    Myke`: well, not really an argument, but.. the reason why it runs faster on, say the the DC, is because the timings for the moves are not the same between versions

    And, my previous post:
    That means that Vf3 on DC would be identical to VF3 arcade unless the programmers changed the animations etc. All characters have animations that relate to each other, and through this the pace of the game is set (the speed).

    Animations and translation of them is the key to the game being the same speed as the arcade. If their are less animations, the game will run faster, if there are more animations it will run slower. My take on the situation is that there is no loss of speed between arcade and home.
    You'd also need to effectively change ALL the characters animations, not just a single one, which is why it's unlikely that the game is faster.
    It's more likely that a single character has LESS animations, which would lead to a difference in the TIMING of his movements and key's to execute them compared to the arcade.

    Now, why don't you go away and come back a nice little boy?
     
  13. Mr. Bungle

    Mr. Bungle Well-Known Member

    way to point out minor errors and completely ignore the many other points made. and i am certainly not alone in rudeness directed at you in #vfhome. come join sometime, it'll be a laugh.

    > It' seems that arguing with an idiot is all that I am doing.

    eheh...fabulous IKYABWAI there. weak.

    > Now, why don't you go away and come back a nice little boy?

    what, did they run out of them at your favorite thai whorehouse?
     
  14. ice-9

    ice-9 Well-Known Member

    <blockquote><font class="small">In reply to:</font><hr>

    anyway a project is in the works to prove the speed issue - i hope you and jeff fucking choke on the results. in the meantime, enjoy this; laughter at other people's expense is fun.

    <hr></blockquote>

    Sure, OK, I'm interested to see. But, make sure it's a little bit better than the "Aoi can reverse Shun's chouwan after SE!" and "Kage's DP is never guaranteed after SE!" tests, aight?

    Rich, I wouldn't be so big on the "everyone's laughing at you FREAK" thing too much. You never know who's laughing behind your back.

    As for Clopin, direct him to gribbly's post, who answers his first person shooters FPS quandary.
     
  15. Myke

    Myke Administrator Staff Member Content Manager Kage

    PSN:
    Myke623
    XBL:
    Myke623
    Animations

    Rugal, you talk about adding and removing animations as if this is a 2d game, where each frame is effectively hand-drawn and then displayed at the appropriate time for the appropriate length. Umm, hello?

    Let me try to abstract the scenario. Let's say I'm coding the dynamics of a Punch. To simplifiy things, I'm only going to consider a point on the fist. Call it point X. To drive the punch I need to move point X to some other point (call it point Y) over a certain time frame (t).

    How fast is this punch? Well it's simply equal to (Y - X) / t which is your basic distance over time relation. If I want to make this punch faster, I'll reduce the value of t, or make it slower by increasing the value of t. If I want to show the punch in slow motion (like during the replays) I'll simply double or triple the t value.

    Nobody knows for sure what time base they've used in the code. You can assume frames all you want, but it remains that -- an assumption. I just think that frames are a convenient way of presenting the information to a human. It's easier to think/talk/discuss 8 frames rather than 0.13333333 seconds for a punch's execution time. So I won't speculate on what time base they're using for the code, but I believe that this time base is what affects how fast the game plays and not the refresh/display rate. In other words, if the game was displaying at 60fps or 30fps, my so-called "8 frame punch" will still be displayed in 0.1333333 seconds.

    See what I'm saying here? While the code may have been directly ported to the DC version, so that all the attack data remained consistent, I just think that the DC and the Arcade versions implement the time base (that 't' value) differently enough for the player to notice. Well, some players I should say.

    Talking about adding and removing frames of animation is pretty ridiculous in the context of 3D.

    Let's just wait for the video demonstration.
     
  16. Mr. Bungle

    Mr. Bungle Well-Known Member

    uh oh. someone is getting the Fear.

    > "Aoi can reverse Shun's chouwan after SE!" and "Kage's DP is never guaranteed after SE!" tests, aight

    get your facts straight, freak. i _never_ said that aoi could, i bitched you out for your crap-tastic testing "method", and then the thread went off on tangents. read the fucking thread again. as for the DP thing - you just quoted something hiro said which was quoted out of the dc tb book. even if the 1P/2P thing is true, then, well...i wasn't entirely wrong.

    > You never know who's laughing behind your back.

    /me slaps forehead in shock and horror at the sudden realization that more than a few people really despise him, and that some people who he may know and possibly like talk and laugh at him behind his back. epiphany! jeff's fabulous insight and analysis strikes again!
     
  17. Rugal

    Rugal Well-Known Member

    Re: Animations

    Talking about adding and removing frames of animation is pretty ridiculous in the context of 3D.
    I agree - in terms of this game - which is why I dont think that the arcade is different to the DC version.
    Note that timescale for the arcade is identical to the DC version as they both conform to 60HZ screen display.
    Sure, you can set games to run at different speeds (ie Quake etc), but that's a matter of re-drawing existing frames or interpoleating frames into the game (ie putting frames of movement into motion where they didnt exist on a slower machine). This isn't really the case for VF3 as it is set to a standard that includes unique frames of animation for each characters move.

    BTW, I'll be in Vf home later, Mr Bungle /versus/images/icons/smile.gif
     
  18. ice-9

    ice-9 Well-Known Member

    My goodness Rich, reading your posts brings me back to middle school...

    And you were WRONG about the DP thing. Why don't you take your own oft-unsolicited advice and admit it?
     
  19. Mr. Bungle

    Mr. Bungle Well-Known Member

    okay, fruitcake. as a categorical statement, i was wrong.

    now, go on...print this out, tack it on the wall, and jerk off to it. keep it safe because there's not much else like that you can put up on the wall; meanwhile, the "ice-9 hall of insanity" - the list of absurd, stupid, and hilarious things jeff has said - grows larger and larger all the time. more laughs for the rest of us.
     
  20. uk-guy

    uk-guy Well-Known Member

    He has a pirate copy! And he admitted to it! With a photo of Himself and Jason! Very Clever, lets hope the federation against priracy don't float aroun this board.

    BTW saw someone get to Fist King (HA HA) on Friday night. All the emperors are in Tokyo for the big tourney this Sunday. They played at Kano Spo and Shibuya GIGO.

    Anyway JoeSAD Akira who is an emperor beat 5 emepror challenges in a row (including OhSu from daioh.com) to get the new rating. We talked with Kyasao and he told us the following new 'titles' exist.

    Fist King
    Wind King
    Sand King
    Evil King
    The Real Emperor.

    Also if your win rating is below 75% than you get less credible titles like Tetsujin and stuff. Kyasao reckons anyone reaching The Real Emperor is impossible because you have to beat like 20 emperors in a row, and the competiton is too tough.

    L8rs
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice