:)The last 3 GIFs are quite interesting!

-------------------------------------------------
Com os melhores cumprimentos,
Kind Regards,

[image: www.dianadavid.net] <http://www.dianadavid.net>

   -


2015-08-12 23:05 GMT+01:00 pedro santos <probi...@gmail.com>:

> @Olivier
> I haven't sent to anyone else to try it. I'm assuming is a UI generation
> issue and not GPU the UI doesn't go blank, the whole application locks up
> instead.
>
> @Matt
> For the operations I'm doing I need these many nodes. I've already done
> other bend deformers too and of course I didn't needed all this indeed (
> http://prntscr.com/83y066 ). I'm trying to get with something a bit more
> evolved but mainly it's a code blueprint. So I can test ideas faster than I
> would with code.
> The parts that are repetitive are already compounds, since again, this is
> a code blueprint I'm interested in convert those into functions.
> I've seen people using matrix to convey operations (Delta Mush compound
> was like that I think). But this operation is not very linear. Things go up
> and down the several streams of processing, so I don't know who to do
> matrix transformation would help in this case. It did help to just compound
> some sections. Just do hide out some things, and the tree started to load
> faster. Thanks.
>
> @Jason S
> Just like I said to Matt, I did that and it helped. Though I'm avoiding do
> it to everything since like I mentioned I need to pull data from many
> places and having the thing a bit more flatter helps. I also don't like to
> set value and call them locally beause this is a blueprint and the
> advantage over code is that I can immediately inspect what is being used
> where.
>
> By the way the ICE execution is very smooth. One of the reasons I moved
> out of LW after being 90% done there( http://prntscr.com/83y3rv ) , in
> ICE that is not problem. Just the UI load that was showing as problem,
> unlike LW where it loaded in 1 sec.
>
>
> Thanks for the tips. Here's the additional gifs I ended up cooking :)
> Cheers
>
>
>
>
>
>
>
> On Tue, Aug 11, 2015 at 3:06 AM, Jason S <jasonsta...@gmail.com> wrote:
>
>> I don't know, but isn't massiveness (or massive looking graphs) very
>> dependant on how elaborate processes are, and how low-level nodes are used?
>>
>> In this case the end-result looks like there's no interpenetration
>> however much 'twisted' are meshes
>> (or looks like a 'smart bend')
>>
>> And I've seen ICE trees that looked very neat at the top, while having
>> several levels of subgraphs that had considerable amounts of things at each
>> level, which I'm sure when all expanded into one tree would make-up quite
>> some trees,
>> (not speaking of execution time which sometimes seems completely
>> unrelated)
>>
>> Speaking of which, have you tried just collapsing entire graph chunks
>> into compounds (at the top level) ?
>>
>>
>>
>>
>> On 08/07/15 20:08, Olivier Jeannel wrote:
>>
>> Looks massive to me ;)
>> Have you tried on another computer ? Just an idea : Could be a graphic
>> card driver problem.
>>
>> Le 7 août 2015 19:05, "pedro santos" <probi...@gmail.com> a écrit :
>>
>>> Hi, I'm doing a bend deformer in ICE
>>> The tree has grown, but I don't think it's something too massive:
>>> http://prntscr.com/80jx9k
>>>
>>> The UI always locked up a little bit, but I could load it eventually,
>>> but my last version simply doesn't load :/
>>> I tried to use the new ICE interaction mode buttons but to no avail. The
>>> execution is actually smooth, It's the tree loading that bring it to its
>>> knees.
>>> In the last days I exclamated "Doh!" whenever I entered a compound,
>>> because when I got out of it back into the main tree the same slow loading.
>>>
>>> Any clues to what might be the problem?
>>>
>>> Thanks!
>>> Pedro / probiner
>>>
>>> Here's some demo gifs I was cooking before I got locked down...
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>
>
> --
>
>
>
>
> *------------------------------[image:
> http://i153.photobucket.com/albums/s202/animatics/probiner-sig.gif]Pedro
> Alpiarça dos Santos Animator  3DModeler  Illustrator >>
> http://probiner.x10.mx/ <http://probiner.x10.mx/>*
>

Reply via email to