|
Hi! Thanks both Andys :) and Jonathan
for that great insight!
I hope H developers are reading this, because I don't feel too comfortable posting on H forums (yet) @Andy G maybe it’s a workflow thingy. So it seems! I like to break functionality down into separate nodes and not try to do everything inside one gigantic Attribute VOP. Actually, don't you often -have to- spread-out across a few nodes with vop vs in vex ? Which can also be another rather big point, because.. that, together with mostly low level nodes therefore requiring lots of them (low level nodes also being likely more technical), is what seems to make using code (vex) sometimes simpler however that may sound... (in some ways but not at all in others) Jonathan recently posted a clip illustrating that. ___________________ On 03/27/17 9:48, Jonathan Moore wrote: https://vimeo.com/206788209 (in comparison to the spaghetti junction that VOP’s can end being for certain challenges). ___________________ a few snippets from that video :: 1m44 ... with the exception of very very few I'm not sure if anyone can call themselves "experts" because there's so much that you can do.. so robust lets just say there's -Alot- of... possibilities. I would have to agree with that! -Alot- of possibilities in which very-very few can master, like -programming- can have Alot of possibilities, yet at the expense of ... With Alex's vop net, it worked, but it seems kinda cumbersome doing math calculations and binging it all back together Promoting parameters.. it can be .. a tedious process, tedious and cumbersome Anywhere there's an opportunity to keep thing as simple as possible and in one place I think is a plus. ... It takes alot less code compared to vops that are really convoluted. Indeed, doing that in Vop(s) involves -a great number- of low level nodes, spread across a couple of top level nodes (lots of nodes and ins and outs) 7m30 These few lines of code is alot less than what was in vops. Another reason to try to use vex as opposed to vops. when Houdini users say they don't like to dive down into things to change things Any situation where you can keep everything in one place I would favor. 11m30 ... all in one node , no diving in, No need looking for different nodes to change settings on.. . He mentioned multiple times "keeping everything in one place", being "art directable" and "simple". Yet while the resulting code is relatively short, and requires much less ins and outs (which is probably the basis of the many requests about unifying all or many contexts) ... I (as well as many I'm sure) wouldn't call it "simple" or easy to understand, it can easily take just as much time to research what every bit does, and how everything relates, as it may have taken to create it, which in this case apparently involved a couple of days.. 15m38 I had kind of a hard time putting it all together searching around.. I probably spend a couple of days trying to get this simple thing put together.. but can apply to alot of different situations, and hopefully making the next guy or girl's job much easier. For the heck of it, I knew this stuff was pretty-much a no brainer in ICE , but still wanted to go ahead and do it (with some extra features as I went along) while screen recording. So here's an 8 min clip (~40 min timelapsed) you can view it online: https://www.dropbox.com/s/1nvvu5sswor6xfv/1-MoveAndWobbleAlongDirection.mp4?dl=0 or you may want to directly download the original mp4 (10 megs) for a clearer view of small text and bright connections over gray. 1-MoveAndWobbleAlongDirection.mp4 A "Bonus Round" with some speed modeling and things, which I hope, yet have yet to see if it's possible to be very fast in Houdini in general, https://www.dropbox.com/s/7otx0p61r0kpp8w/2-BonusRound-SpeedEverything.mp4?dl=0 2-BonusRound-SpeedEverything.mp4 And the scene itself Randomly Translate and Wobble along direction.scn Needless to say, it didn't involve a couple of days of searching around, or opening-up docs, - result was a single node with centralized settings, which was not only "art directable" but also "art authorable" and "art editable" with super simple to understand inner-workings made of but a few nodes, for achieving essentially the same thing (as for so many things) The point isn't to say "hey! look how much better Ice is" or to make Houdini like whatever, Just to point out that it's entirely possible for systems to be BOTH very approachable (or fast) AND very flexible at the same time in their own unique way. ICE, or XSI in general just happens to be an excellent 'living' example (or a dead but still living example), or as an at least very functional proof of that. This can also concern C4d which is completely other way around: -- making quite heavy compromises on flexibility for the sake of approachability. And I think (as well as many I'm sure) that both users and developers should constantly strive at pushing for both, as opposed to mostly (or exclusively) favoring one over the other. Thanks again for the feedback! -J On 04/07/17 14:15, Andy Goehler wrote:
|
------ Softimage Mailing List. To unsubscribe, send a mail to softimage-requ...@listproc.autodesk.com with "unsubscribe" in the subject, and reply to confirm.
