I read that people complain about the exponential complexity of shader variation compilation time.
I dread adding new packages from the asset store, if they contain new shaders.
I think I also read this might be addressed?
Is there any expectation when this issue will be addressed?
The exponential problem only applies to one shader file at a time (multi_compile). If you add a new plugin that contains new shaders to your project, you will not see exponential growth. However, you will likely see somewhat linear growth in your build times.
Unless they change how the entire shader system works from the ground up, I think we’ll always have the exponential issue with multi_compile, with ways to work around it (like with shader_feature or keyword stripping).
How do I use keyword stripping and shader_feature?
"
Compiling shader “Universal Render Pipeline/Lit” pass “ForwardLit” (vp)
896 / 10752 variants left after stripping, processed in 0.48 seconds
starting compilation…
finished in 0.03 seconds. Local cache hits 896 (0.54s CPU time), remote cache hits 0 (0.00s CPU time), compiled 0 variants (0.00s CPU time), skipped 0 variants
Compiling shader “Universal Render Pipeline/Lit” pass “ForwardLit” (fp)
[517.68s] 100M / ~415M prepared
[1121.28s] 200M / ~415M prepared
[1752.93s] 300M / ~415M prepared
[2146.39s] 400M / ~415M prepared
29568 / 90832896 variants left after stripping, processed in 2207.51 seconds
starting compilation…
[ 66s] 94 / 29568 variants ready
[132s] 187 / 29568 variants ready
[200s] 285 / 29568 variants ready
[270s] 383 / 29568 variants ready
[340s] 480 / 29568 variants ready
[409s] 577 / 29568 variants ready
[477s] 673 / 29568 variants ready
[543s] 766 / 29568 variants ready
[612s] 863 / 29568 variants ready
[679s] 955 / 29568 variants ready
[746s] 1051 / 29568 variants ready
[812s] 1147 / 29568 variants ready
[881s] 1243 / 29568 variants ready
[953s] 1340 / 29568 variants ready
[1025s] 1439 / 29568 variants ready
[1095s] 1536 / 29568 variants ready
[1163s] 1633 / 29568 variants ready
[1233s] 1727 / 29568 variants ready
[1301s] 1828 / 29568 variants ready
[1366s] 1925 / 29568 variants ready
[1431s] 2023 / 29568 variants ready
[1491s] 2115 / 29568 variants ready
[1556s] 2212 / 29568 variants ready
[1620s] 2308 / 29568 variants ready
[1685s] 2405 / 29568 variants ready
[1751s] 2505 / 29568 variants ready
[1818s] 2610 / 29568 variants ready
[1878s] 2705 / 29568 variants ready
[1939s] 2799 / 29568 variants ready
[2005s] 3028 / 29568 variants ready
[2066s] 3125 / 29568 variants ready
[2133s] 3225 / 29568 variants ready
[2202s] 3324 / 29568 variants ready
[2266s] 3420 / 29568 variants ready
[2327s] 3513 / 29568 variants ready
[2391s] 3758 / 29568 variants ready
[2459s] 3880 / 29568 variants ready
[2524s] 4009 / 29568 variants ready
[2591s] 4136 / 29568 variants ready
[2653s] 4244 / 29568 variants ready
[2716s] 4360 / 29568 variants ready
[2782s] 4482 / 29568 variants ready
[2848s] 4609 / 29568 variants ready
[2912s] 4725 / 29568 variants ready
[2978s] 4849 / 29568 variants ready
[3044s] 4971 / 29568 variants ready
[3111s] 5100 / 29568 variants ready
[3171s] 5216 / 29568 variants ready
[3239s] 5343 / 29568 variants ready
[3306s] 5457 / 29568 variants ready
[3374s] 5587 / 29568 variants ready
[3435s] 5703 / 29568 variants ready
[3498s] 5817 / 29568 variants ready
[3564s] 5954 / 29568 variants ready
[3628s] 6082 / 29568 variants ready
[3689s] 6204 / 29568 variants ready
[3752s] 6331 / 29568 variants ready
[3816s] 6458 / 29568 variants ready
[3881s] 6721 / 29568 variants ready
[3946s] 6848 / 29568 variants ready
[4009s] 6978 / 29568 variants ready
[4070s] 7106 / 29568 variants ready
[4135s] 7236 / 29568 variants ready
[4205s] 7479 / 29568 variants ready
[4275s] 7587 / 29568 variants ready
[4336s] 7687 / 29568 variants ready
[4397s] 7782 / 29568 variants ready
[4459s] 7879 / 29568 variants ready
[4526s] 7980 / 29568 variants ready
[4590s] 8075 / 29568 variants ready
[4652s] 8168 / 29568 variants ready
[4720s] 8270 / 29568 variants ready
[4786s] 8366 / 29568 variants ready
[4849s] 8463 / 29568 variants ready
[4919s] 8568 / 29568 variants ready
[4985s] 8678 / 29568 variants ready
[5056s] 8786 / 29568 variants ready
[5124s] 8897 / 29568 variants ready
[5190s] 9006 / 29568 variants ready
[5252s] 9107 / 29568 variants ready
[5317s] 9213 / 29568 variants ready
[5387s] 9326 / 29568 variants ready
[5452s] 9434 / 29568 variants ready
[5520s] 9549 / 29568 variants ready
[5582s] 9649 / 29568 variants ready
[5645s] 9744 / 29568 variants ready
[5707s] 9841 / 29568 variants ready
[5768s] 9940 / 29568 variants ready
[5836s] 10049 / 29568 variants ready
[5904s] 10151 / 29568 variants ready
[5968s] 10248 / 29568 variants ready
[6035s] 10480 / 29568 variants ready
[6098s] 10579 / 29568 variants ready
[6166s] 10689 / 29568 variants ready
[6236s] 10802 / 29568 variants ready
[6296s] 10900 / 29568 variants ready
[6361s] 11158 / 29568 variants ready
[6426s] 11290 / 29568 variants ready
[6489s] 11417 / 29568 variants ready
[6549s] 11542 / 29568 variants ready
[6616s] 11684 / 29568 variants ready
[6678s] 11814 / 29568 variants ready
[6746s] 11958 / 29568 variants ready
[6813s] 12100 / 29568 variants ready
[6873s] 12230 / 29568 variants ready
[6936s] 12359 / 29568 variants ready
[7001s] 12487 / 29568 variants ready
[7061s] 12609 / 29568 variants ready
[7128s] 12741 / 29568 variants ready
[7193s] 12868 / 29568 variants ready
[7259s] 12994 / 29568 variants ready
[7323s] 13117 / 29568 variants ready
[7388s] 13244 / 29568 variants ready
[7451s] 13376 / 29568 variants ready
finished in 7508.61 seconds. Local cache hits 792 (2.54s CPU time), remote cache hits 0 (0.00s CPU time), compiled 12703 variants (120108.14s CPU time), skipped 16073 variants"
Why it takes so long to compile URP Lit?
This is dysfunctional. A broken product.