Danbooru

Filesize inconsistencies between pixiv and other sites

Posted under General

tapnek said:

That's what I'm trying to say. Although, in those cases I mentioned, the Twitter source I uploaded was the superior source to the Tumblr one I also uploaded at the time as the Pixiv source wasn't uploaded yet.

Twitter can't be superior as long as they edit the files. Tumblr has the original art just like pixiv and everywhere else.

Schrobby said:

Twitter can't be superior as long as they edit the files. Tumblr has the original art just like pixiv and everywhere else.

Example:
Tumblr: post #2112111 (1280x1316)
Twitter: post #2150093 (1679x1726)

Tumblr scaled the image down, while twitter preserved the original image. In cases like this twitter is superior.

Tumblr downscales? I think you're mistaken:

Ars said:

post #2000942

Pixiv: http://i2.pixiv.net/img-original/img/2015/06/17/20/14/07/50945645_p0.png 853.82 KB

Seiga: http://seiga.nicovideo.jp/seiga/im4933786 853.82 KB

Tumblr: http://digitalflyer.jp/image/117928847779 853.82 KB

Twitter: https://twitter.com/un9man/status/594454266997116928/photo/1 (https://pbs.twimg.com/media/CD_tAV1VIAAK_x2.png:orig) 899.25 KB

Of course, the Twitter source is the only one that is different.

Result: Tumblr is superior since it's no third party edit.

Schrobby said:

Tumblr downscales? I think you're mistaken:

Result: Tumblr is superior since it's no third party edit.

From my understanding, Tumblr only downscales images that are larger than 1280x1920.
The image in Ars's example is 1280x847 so it didn't get downscaled.

Toks said:

From my understanding, Tumblr only downscales images that are larger than 1280x1920.
The image in Ars's example is 1280x847 so it didn't get downscaled.

I see. K, so in this special case where both files are edited the twitter one is indeed better. That doesn't change anything in general, though.

CodeKyuubi said:

Nico post #2172466 has a smaller size than its pixiv parent.

NWF_Renim said:

The pixiv version of that image appears to actually be the inferior version. The nico version of the image appears to have less artifacting.

Yeah, this. The artist simply uploaded a slightly better quality version to Seiga. I've checked other images recently upload to both Seiga and pixiv and have seen no differences.

Ars said:

Yeah, this. The artist simply uploaded a slightly better quality version to Seiga. I've checked other images recently upload to both Seiga and pixiv and have seen no differences.

That's... rather strange of the artist to do, haha.

RaisingK said:

Further discussed here.

Ah, I don't really use discord so this slipped by me.
(Christ, it's worse than I thought. Doing this to newer images is one thing but going full retroactive is just another level of fucked. The Hell are pixiv devs even thinking.)

Ars said:

Ah, I don't really use discord so this slipped by me.
(Christ, it's worse than I thought. Doing this to newer images is one thing but going full retroactive is just another level of fucked. The Hell are pixiv devs even thinking.)

Reminds me about how toranoana started employing some atrocious image compression algorithm about a year ago to everything on their site, including old articles/images. I used to occasionally find gems to upload there, now it's a graveyard.

1 2 3 4 5