Dampierre random move data (theory)

Status
Not open for further replies.
S

Somebody

[08] Mercenary
Even though there's a thread for Dampierre's stances that covers some of Dampierre's "random" moves, I think there should be a separate thread on the proper timing for certain "outcomes". The purpose of this thread is to get down the appropiate timings for Dampierre's moves as accurately as possible. I've gone through most of Dampierre's "random" moveset and tried to get pretty close approximate times for the timing of each "outcome". My methods of getting these timings are far from perfect but it should at least give an idea of when to time a move.

Important: I guess I should clarify, Dampierre's moves are not random; the time when you input a move determines the outcome.

Update:
- Added 22_88KK
-----------------------------------------------------------------------------------------

Since I don't have any tools to precisely measure the framerate and the timer for SCV, I had to go on a standard stopwatch and my own average timing. Hopefully this will be improved in the future by me or someone else, but this should still help.


Notes:
- Every second on the timer is equal to 60 frames and most of Dampierre's "random" moves correspond to the last 6-10 frames of a second (Which is about the last 1/6 of a second).
- Only the first button (not D-pad) input matters.
- 33_99bK~PB refers to the version with the low sweep while 33_99bK refers to the mid kick version
- 44K~PB refers to the "Lord Dampierre" version while 44K refers to the roll.
- A+G (W!) refers to the wall splat version of the move.
- Again, the frame data may not be 100% correct, it's just an approximate timing.
- 22_88KK~PB refers to the "Lord Dampierre" version with the second hit while 22_88KK refers to the roll.

*There might be more frames for 44K~PB and 22_88KK~PB.

33_99bK~PB (GB) last 2-4 frames
33_99bK (GB) last 4-6 frames
33_99bK last 6-10 frames
33_99bK~PB first 50-54 frames
B+K (instant) last 2-4 frames
A+B last 6-10 frames
A+B~PB first 50-54 frames
623B~PB last 6-10 frames
623B first 50-54 frames
4B~PB last 6-10 frames
4B first 50-54 frames
*44K~PB last 6-10 frames
44K first 50-54 frames
3AAAAA... (Every hit after the first) last 6-10 frames
3AAAAA... (FP, Every hit after the first) first 50-54 frames
1K last 6-10 frames
1K (FP) first 50-54 frames
11_77K~PB (Only first hit comes out and puts you into PB) last 6-10 frames
11_77K first 50-54 frames
*22_88KK~PB last 6-10 frames
22_88KK first 50-54 frames
33_99B BE (GB) last 6-10 frames
33_99B BE first 50-54 frames
6B BE (instant) last 2-4 frames
A+G (W!) last 6-10 frames
A+G first 50-54 frames
B+G (Max dmg) last 2-4 frames
PB B_A+G_B+G ???

Notes on PB B_A+G_B+G:
PB B_A+G_B+G seems to be split up into two parts, the command input and the mashing during the move. The command input seems to determine the length you get to mash and the last mash input during this length seems to determine whether or not you get the last two hits. However, I really don't know what the timing is, but I would assume that it's at the very end of a second (last 6-10 frames) for max length and the very end of a second for the last two hits.

Additional notes:
- I haven't spent too much time on the different timings for B+K or B+G, but I'll try to figure out the different timings at some point.
- Still need to add some of the brave edges.

Any help to complete, confirm, or correct this data would be greatly appreciated.
 
I'm not trying to sh*t on your efforts, in fact I respect the work that you've put in to this, but like someone else mentioned in another thread, this all seems anecdotal.

I've attempted some tests based on your results and have been getting inconsistent and unconvincing results. Your results could have very well came from lucky random coincidences.

The only way to really prove that the random outcomes are based on the timer would be to have something programed to execute a move at the exact same millisecond every time with 100% consistent results. Otherwise human error will always be a factor that would cause doubt.

The main reason that I'm not convinced is with moves happening at the start of a round. Executing a move at the earliest possible time at the beginning of a round is the easiest way for someone to consistently do a move at a certain time with minimal human error. Yet doing so still yields random results.

So if your timer based theory were true, then opening a round with B+K, A+B, or 4B would give consistent results pretty much every time.

I was actually considering making a random move data thread as well, but under the assumption that it's random, just how Eyce_Theon did for Dampierre in the SCBD SA here: http://8wayrun.com/threads/dampierres-stances.3612/#post-148860

Basically doing a move 100 or 200 times and logging how often it went into stance and how often it didn't, and then estimating what the odds him going into stance are based on those results.
 
A video showcasing the exploit with Dampierre's 33_99B (623B in SCV) in SC: BD:


It's not very good quality, but I'll upload more videos and try to improve the quality.
 
No luck involved OOFMATIC, it can be done EVERY time. Try to actually time it for once, the easiest way is to go to training and keep resetting the time and attempt the move over and over again until you have a feel for the timing, every time you reset the infinite time it goes back to zero. After you have a general feel for the timing, try with an actual timer in 2P-Vs., then try against a real opponent when your proficient enough. Remember:
A+B = very end/ beginning of second
A+B~PB = every other part of the second
623B~PB = very end/ beginning of second
623B = every other part of the second

Until you can proficiently do these you will fail in high level play.
 
Okay, I have several issues with you, Somebody/DOCTORSTUPID (yes, I know you are one person), so let me just lay it out here.

First of all don't use that black font, myself and others use the Dark forum skin and on certain browsers it makes it very difficult to read your text without highlighting or switching to the Light forum skin.

Second, it is against the rules to have multiple accounts. Myself and mods/admins of this site know about it and one of your accounts may be deleted. I suggest you tell Jaxel which account you wish to keep or he might just delete one randomly. I'll get back to this issue in the next point.

The next point: I am against how you are constantly presenting your theories and "test results" as absolute facts, with no proof. It's misleading to new players because they might think that it's true even though there is no proof of it. And that video you posted doesn't prove anything. And the fact that you have a second account doesn't help because it makes it appear as if a second person confirmed your tests, when in fact you are alone in finding these "results".

Now, as for the theory at hand (because that's all it is, a theory) I've already stated my opinion on it in my previous posts (which contained points that you still haven't addressed) but I'll add another point:

Assuming your theory is true, please explain this to me -

Say you do two random attacks in a row. For example 623B, 623B. If you were to input the second 623B right after a non-PB 623B, it would consistently come after the same amount of time, because it would come out immediately after the animation of the non-PB 623B.

So according to your theory, if you "time" your input for the non-PB 623B to come out, doing a second 623B immediately after would lead to consistent results on the second one, because the same amount of time would have always passed.

Anyways aside from that the main reason I dislike your theory is because it can't be concretely proven, like I stated in previous posts. The human error factor of testing something like this leads to unfounded assumptions and no solid proof.

And even if it were true, it actually wouldn't be very useful in battle, because in high level play you need to be basing your moves off of the opponent, not off of a fraction of a fraction of a second on the clock. That would just be something to distract you from the opponent, and would get you killed real quick.

With Yoshimitsu it's different, because you have a whole second for your timed move to come out, so it wouldn't be as big of a distraction. But with Dampierre and it being a tiny fraction of a second, with the amount of moves he has that you claim require this, it's just crazy and unrealistic to use it effectively and consistently.

DOCTORSTUPID said:
Until you can proficiently do these you will fail in high level play.
Click to expand...

This statement is so idiotic that it made me laugh.

I'll leave this thread open to give you a chance to have your say, but after a few days I will be closing this thread, since it is misleading with unfounded and unproven claims. I will move any posts of yours in other threads regarding unproven "timed" moves to this thread, to prevent people from being mislead. If you have any issues with me about this feel free to PM me.

And if you can prove to me without a doubt that your theory is true then I will offer you a full apology and re-open this thread.
 
Alright, sorry if my post came off a bit harsh but the last thing I want is a bunch of new Dampierre players being mislead with false information.

If you can can prove to me somehow without a doubt that your timer theory is true, then I will join you in investigating it extensivley. Otherwise I don't want to bother wasting my time with testing it, because of points stated in my previous posts.
 
Somebody said:
Alright, fair enough, but all I can do is just upload more videos.
Click to expand...
The thing is, A+B and 623B are not terribly good moves aside from the damage output (which is inconsistent).

They are linear, and they are slow for something you just throw out there.

When using them as whiff punishers- you can't possibly watch the clock to know if you're hitting the timeframe or not. I can barely punish whiffs with my poor reflexes as is, much less process all of that information...
 
Since I'm not 100% positive on the timing, I would practice in versus mode and spam something like A+B and try to pay attention to the time when it works, you should eventually see a pattern. To be honest, I haven't found any reason to believe that Dampierre's moves are even pseudorandom other than that it says (random) on his move list. Still, I will upload videos to either prove or disprove this exploit. It's fair to say I don't have any real proof right now, but it's still worth a try to try and do it on your own.
 
It's random. It's easy to prove it's random.

Step 1: Local versus, pick Dampierre vs Dampierre
Step 2: Do and easy command like A+B or B+K on both controllers at the same time
Step 3: Observer how often they have different results.

Now, there will be times when it'll line up perfectly with what you say, but you're going to have to ignore the many many times that it doesn't match up for your theory to be correct.
 
Agreed. I will repost my two other points that also prove that it's random:

So if your timer based theory were true, then opening a round with B+K, A+B, or 4B would give consistent results pretty much every time. I've tested that and it isn't the case.
Click to expand...

Say you do two random attacks in a row. For example 623B, 623B. If you were to input the second 623B right after a non-PB 623B, it would consistently come after the same amount of time, because it would come out immediately after the animation of the non-PB 623B.

So according to your theory, if you "time" your input for the non-PB 623B to come out, doing a second 623B immediately after would lead to consistent results on the second one, because the same amount of time would have always passed. Doing one immediately after another does not give consistent results.
Click to expand...

Thread closed since the theory is now disproven.
 
Status
Not open for further replies.
Back