Open zelenooki87 opened 1 month ago
The exception is raised because rife 4.26 upgrades the architecture. (https://github.com/hzwer/Practical-RIFE/issues/109#issuecomment-2366812473)
You should not expect the limitation on tile size to be stable for the v1 implementation. The v2 implementation handles this automatically.
@zelenooki87 In order to temporarily fix the following error: "Failed to evaluate the script: Python exception: vsmlrt.RIFEMerge: tile size must be divisible by 64" on the "v1" variants until the SVP developer updates their code, you can open C:\Program Files (x86)\SVP 4\script\generate.js and replace this:
var pw = Math.floor((media.dst_w-1)/32+1)*32 - media.dst_w;
var ph = Math.floor((media.dst_h-1)/32+1)*32 - media.dst_h;
with this:
var pw = Math.floor((media.dst_w-1)/64+1)*64 - media.dst_w;
var ph = Math.floor((media.dst_h-1)/64+1)*64 - media.dst_h;
Latest Hybrid dev and vsmlrt addon. doe snot matter if I use Directml or TensorRT Input file dimensions are realy small Width : 176 pixels Height : 144 pixels
rife 4.25 and lower doesnt have this problem with such small input dimensions.