Le 21/05/2015 23:39, Rainer . a écrit :
Thanks for the response so far!!!

1000000 - just a number to make it clear that I want to sort big lists.

I should have been more precise - in fact I want to sort pixel values, a huge 
amount of pixel values. I´ve done this already with sort from zexy (with 
arrays) and it worked pretty well. But it´s either the original or the sorted 
result. So a animation or a warp or at least 75% between would be nice.
...

you can have 2 array : the unsorted, and the sorted. then interpolate between 2.

 >> 1000000 is hard to handle...
 > it's more square than pixel in a 1024x768 screen!
I work with images >5k resolution (and tiling)  :-)
this is still a lot for real time in the CPU.
but if you find a way to store the 2 array in the GPU (using texture), and 
interpolate with a shader, then you will be fine.



btw. what is the meaning of "invision"???


i think Iohannes ask what is a 75% sorted list.

cheers
c

Cheers,

Rainer



2015-05-21 18:52 GMT+02:00 Cyrille Henry <[email protected] 
<mailto:[email protected]>>:



    Le 21/05/2015 18:13, Py Fave a écrit :

        you can multiply lists component wise check in (listabs ) in help 
browser

        multiply by a fraction then multiply with the goal values

    for large list, i suggest using array and iem_tab

    for a fade between A and B, x being the fade factor for 0 to 1, the fade 
would be :
    A * (1-x) + B * x

        you should look egregore by chdh (on this list ) for morphing
        and they have a live in one week on the net using the patch

    thanks for the reminder ;-)

    we don't have "lot's" of fade involve in egregore, but if you like lot's of particles 
moving, you should see the next chdh  performance : "morphist"
    (2 live in Paris next month, see chdh.net <http://chdh.net> for more 
informations)


        1000000 is hard to handle...

    it's more square than pixel in a 1024x768 screen!
    but with a good computer, 100 000 moving dots should not be a problem.

    cheers

    c


        2015-05-21 16:16 GMT+02:00 Rainer . <[email protected] 
<mailto:[email protected]>>:

                e.g. what's 75% between "1 6 8 4 3" and "1 3 4 6 8"


            Yes, that´s the question. I am thinking of e.g. 1000000 squares 
[GEM] fading
            from random position to sorted postition.

            How can I fade lists?

            Thanks,

            Rainer

            2015-05-21 15:56 GMT+02:00 IOhannes m zmoelnig <[email protected] 
<mailto:[email protected]>>:


                On 2015-05-21 15:47, Rainer . wrote:

                    Hi,

                    I´m just playing with the sort object from zexy lib, 
sorting a bunch of
                    floats. I wonder if there is a way to fade or morph(?) from 
unsorted to
                    sorted state?


                how do you invision that?
                e.g. what's 75% between "1 6 8 4 3" and "1 3 4 6 8"

                how is your result different from a generic fade between two 
lists of
                equal lengths?

                fgsmdr
                IOhannes



                _______________________________________________
                [email protected] <mailto:[email protected]> mailing list
                UNSUBSCRIBE and account-management ->
                http://lists.puredata.info/listinfo/pd-list



            _______________________________________________
            [email protected] <mailto:[email protected]> mailing list
            UNSUBSCRIBE and account-management ->
            http://lists.puredata.info/listinfo/pd-list


        _______________________________________________
        [email protected] <mailto:[email protected]> mailing list
        UNSUBSCRIBE and account-management -> 
http://lists.puredata.info/listinfo/pd-list




_______________________________________________
[email protected] mailing list
UNSUBSCRIBE and account-management -> 
http://lists.puredata.info/listinfo/pd-list


_______________________________________________
[email protected] mailing list
UNSUBSCRIBE and account-management -> 
http://lists.puredata.info/listinfo/pd-list

Reply via email to