Package Details: waifu2x-git 1:0.13.2.r171.g4313ace-1

Git Clone URL: https://aur.archlinux.org/waifu2x-git.git (read-only, click to copy)
Package Base: waifu2x-git
Description: Image rescaling and noise reduction using the power of convolutional neural networks
Upstream URL: https://github.com/nagadomi/waifu2x
Licenses: MIT
Conflicts: waifu2x
Provides: waifu2x
Submitter: fbrennan
Maintainer: eclairevoyant
Last Packager: chrhasse
Votes: 13
Popularity: 0.000015
First Submitted: 2015-09-22 13:01 (UTC)
Last Updated: 2019-02-03 07:38 (UTC)

Dependencies (20)

Required by (0)

Sources (3)

Latest Comments

« First ‹ Previous 1 2 3

fbrennan commented on 2015-10-09 10:09 (UTC)

Hello, Thanks for your report - it's because upstream finally added `make install` and `make test`, but by adding them they broke the hack that I was using previously to install. Now my PKGBUILD uses them. See https://github.com/tanakamura/waifu2x-converter-cpp/pull/9 for infos. It should all work now, I had to change his PKGBUILD patch a bit though

angerson commented on 2015-10-09 02:06 (UTC) (edited on 2015-10-09 02:06 (UTC) by angerson)

An upstream update broke the patch for src/main.cpp by moving a hard-coded path into a #define directive. It should be a quick fix; can you update the package?

YamashitaRen commented on 2015-09-24 22:05 (UTC)

Here's the CUDA PKGBUILD : https://aur.archlinux.org/packages/waifu2x-converter-cpp-cuda-git/

YamashitaRen commented on 2015-09-23 16:07 (UTC)

Yes, it might be a good idea. I think I'll do a clean PKGBUILD when I'll have time but don't count on me to maintain it :D (Your current PKGBUILD works with nvidia but it will not use cuda, which make it 2x slower on my computer) ps : I think that when sl1pkn07 asked you to replace opencl-mesa by libcl, he really meant "libcl"

fbrennan commented on 2015-09-23 14:44 (UTC)

I think it may be better, for the nvidia users, to create a waifu2x-nvidia-git package, and put CUDA as a hard dependency. What do you think?

YamashitaRen commented on 2015-09-23 12:45 (UTC)

It isn't :D Here's my quick fix : http://pastebin.archlinux.fr/1587829

fbrennan commented on 2015-09-22 14:27 (UTC)

Done, thank you for the tip :) https://aur.archlinux.org/cgit/aur.git/commit/?h=waifu2x-git&id=e6922cf8f3bae872aabf238a0c97a277bf83dcdb I would actually appreciate someone attempting to use this PKGBUILD on an NVIDIA GPU with CUDA, and seeing if my advice in the PKGBUILD is correct. I only have AMD GPU in my Arch desktop and Intel CPU in my Arch laptop, so I can't test.

sl1pkn07 commented on 2015-09-22 13:42 (UTC)

please change opencl-mesa to libcl greetings