UPSB v4
Research Department Feedback / Combo transformation
-
Date: Fri, Jun 29 2012 14:20:08
New article has been released. View it here: http://www.upsb.info/wiki/index.php?title=Combo_transformation
-
Date: Fri, Jun 29 2012 15:02:17
cant understand D:
-
Date: Fri, Jun 29 2012 15:17:34
woah this is cool stuff. didnt know pen spinning will get so technical
-
Date: Fri, Jun 29 2012 15:23:09
Walkaz 9/3 wrote: cant understand D:
it is a hard read, but the technical terms were necessary to formalize the concept. the article does contain informal descriptions of the transformations as well. -
Date: Fri, Jun 29 2012 15:33:14
Unfortunately this is hard to take into account the overall flow of the combo, see HALs combos where it often looks like the pen is always moving in one direction with very minimal direction changes.
-
Date: Fri, Jun 29 2012 16:13:29
eurocracy wrote: Unfortunately this is hard to take into account the overall flow of the combo, see HALs combos where it often looks like the pen is always moving in one direction with very minimal direction changes.
im pretty sure you posted in the wrong thread here -
Date: Fri, Jun 29 2012 17:11:40
Pen spinning is starting to sound more and more like math class.
-
Date: Fri, Jun 29 2012 18:13:07
Zombo wrote: im pretty sure you posted in the wrong thread here
No, no I did not post in the wrong thread. The problem being that if you mesh together tricks from two different combos you lose the fluidity of the original combos. For example, you could end up with a LOT of direction changes that could look rather messy. That's in the trivial examples. In the Non-Trivial examples, the sheer amount of time it would take to lay out a combo that is equally difficult could result in a combo being rather bland with no peak points in it (See spinnerpeem as an example of where extremely high peaks of difficulty occur in his combos). You're talking about making the slow and fast parts of a combo the same amounts of rotations, unfortunately that can result, if it is not oscillated in a change of speed randomly, several times in a combo, or otherwise. Also, flashier elements take longer to lay out the same level of rotations as a slower trick, ending up with a combo that could be far longer than it should be. And as before it's going to take an absurd amount of time to plan out a combo that you can do consistently and has the rotations counted out. I also find it rather confusing with the explanation in the code boxes how you've done it like program code, which makes it really confusing for anyone reading it. Conditional statements are not so easily processed by the human mind. Overrall, combo transformation sounds like an interesting concept but is next to impossible to apply to a real situation, and I don't see it being applied in the future. -
Date: Fri, Jun 29 2012 20:01:24
eurocracy wrote: No, no I did not post in the wrong thread. The problem being that if you mesh together tricks from two different combos you lose the fluidity of the original combos. For example, you could end up with a LOT of direction changes that could look rather messy. That's in the trivial examples.
None of the trivial examples are transformations which require 2 combos as input. Not sure what you're referring to.In the Non-Trivial examples, the sheer amount of time it would take to lay out a combo that is equally difficult could result in a combo being rather bland with no peak points in it (See spinnerpeem as an example of where extremely high peaks of difficulty occur in his combos).
The examples are theoretical in nature. The transformation guarantees a specific property, does not mean the property is actually desirable. This is an example to show how difficulty can be flattened. The combo is then meant to be reprocessed to add peaks and valleys of difficulty, as stated in the iteration section. The idea is to take an existing combo, strip it of its highlights so that it can serve as a foundation for a new combo with different highlights.You're talking about making the slow and fast parts of a combo the same amounts of rotations, unfortunately that can result, if it is not oscillated in a change of speed randomly, several times in a combo, or otherwise. Also, flashier elements take longer to lay out the same level of rotations as a slower trick, ending up with a combo that could be far longer than it should be.
I assume you're talking about fit-to-rhythm. The transformation is not trick-wise, but sequence-wise. The problem will not occur since you're transforming sequences of tricks of roughly same length, not individual tricks.And as before it's going to take an absurd amount of time to plan out a combo that you can do consistently and has the rotations counted out.
The article is theoretical, there is no claim that the examples are practical.I also find it rather confusing with the explanation in the code boxes how you've done it like program code, which makes it really confusing for anyone reading it. Conditional statements are not so easily processed by the human mind.
This article is actually part of a larger project to formalize pen spinning such that it can be mechanized. The ultimate goal is to create combo generation/transformation programs. Combos output by such program are then given back to the spinners, who can then modify them as desired.Overrall, combo transformation sounds like an interesting concept but is next to impossible to apply to a real situation, and I don't see it being applied in the future.
Research work is not concerned with practicality. It's better to put out novel ideas first and deal with pragmatic concerns later. Even without machine assistance, some of the easier systems can be applied fairly easily. Don't forget that at a higher level, combo planning already takes a large amount of time, so using systems to guide your thinking can speed up the process a little bit. It can be used in two ways: holistic transformations can provide a coarse-grained framework for a combo, which can then be "filled in the blanks" manually by the spinner, OR a combo structure is already set in place by the spinner, who then applies fine-grained transformation techniques to refine combo sub-sequences in a stepwise manner. -
Date: Fri, Jun 29 2012 20:44:40
Oh, alright, I think I got a little confused...
-
Date: Fri, Jun 29 2012 21:04:49
eurocracy wrote: Oh, alright, I think I got a little confused...
also in this thread -> http://forum.upsb.info/showthread.php?t=11790&p=209699&viewfull=1#post209699 I talk about a new theory I have concerning stability. Combos which vary too much will appear too chaotic and undesirable. Therefore there is a right degree of variance to be had in a structure. One approach to this is the concept of "fixed point" or "anchor'. The idea is to pick one dimension of stability to be varied throughout the combo, which will serve as grounding element, and vary on the other dimensions. The example given in the other thread is to stabilize spin rotation and vary other aspects (this is the approach demonstrated by Supawit127). -
Date: Fri, Jun 29 2012 21:49:10
cool shit Zbo. I don't get why people are having a hard time understanding this...
-
Date: Sat, Jun 30 2012 03:43:54
Awesome wrote: cool shit Zbo. I don't get why people are having a hard time understanding this...
because our 1st language isnt english? x) -
Date: Sat, Jun 30 2012 04:23:02
english wasnt my first language there is still many words i :dunno:
-
Date: Sat, Jun 30 2012 21:08:09
lol it does take quite a bit of time to process for people who's first language isnt english (including me), but after taking the time to read it carefully, its understandable. just takes a bit of brain power and time @Zombo holy crap all these articles you take the time to write are freakin crazy (in an amazing good way)