EDIT : Tried to keep it short but I think describing in details may avoid some questions 
Hello there every1 ! I 've been playing aroung with TD for like 2 months now and I ve been following many available online tutorials and workshops BUT this is the first time i really feel hardstuck ! Btw I have little or zero knowledge in glsl .
To sum up my problem : I’m deforming a SOP Grid with an image TOP by using each pixels value to deform the grid’s points on the z axis. It works perfectly fine but now I’d like that image to be generated from CHOPs to be able to use logic elements and analysis to create a pattern where pixels interact between each others. Or something like that 
BUT when i transform my CHOP signal into a TOP it renders every value on a single row ( because the signal is on 1 channel only) AND i’d like to render it in the same dimension as my source grid.
For example : if my grid is 128128 i need to generate a 16 384 samples long channel (one sample for each point of my grid ). And I end up with a TOP being 16 834 * 1 pixel. I’d like it to become 128128 but to do that I would have to trim and merge again my CHOP signal 128 times to have 128 channels of 128 samples.
That would be quite a burden to do manually and I’ve been looking for another way to do it like maybe using python, telling TD to create X Trim and Merge CHOPs where X = number of rows or columns on the source grid. But i’m not really sure TD would work that way !
What do you guys think ? Am I missing something simple ? Could I just remap my pixels using UV coordinates ? Would going through python coding be the good idea to automate the creation of the needed CHOPs ? Is there a simple way to do it in GLSL ? Am I simply trying to chew too much than I can ? I don’t know, it seems so simple to achieve but I’ve been spending several days trying and searching for clues and now here I am haha ! 
And btw thank you to all the touchdesigner community wich i already got a lot of help from reading forum threads and following tips & tutorials.
Cheers