This version has the exact same reslution as and a slightly lower file size (255,5 kB vs 231,1 kB) than the child post that has been flagged for deletion. Looking at both images very closely in gimp reveals barely any difference. However, the compression artifacts look ever so slightly stronger in this one.

I think the child post should stay while this one gets deleted.

Lord_Ocean said:
This version has the exact same reslution as and a slightly lower file size (255,5 kB vs 231,1 kB) than the child post that has been flagged for deletion. Looking at both images very closely in gimp reveals barely any difference. However, the compression artifacts look ever so slightly stronger in this one.

I think the child post should stay while this one gets deleted.

This is a higher-quality version. The older upload used the sample image instead of the original image as a source. The sample contains additional visual artifacts (map of pixel differences, see Idem's sourcing suite) and was thus flagged as inferior. Also see the Inkbunny section under the Sites and Sources wiki page for more information.

Songbird said:
This is a higher-quality version. The older upload used the sample image instead of the original image as a source. The sample contains additional visual artifacts (map of pixel differences, see Idem's sourcing suite) and was thus flagged as inferior. Also see the Inkbunny section under the Sites and Sources wiki page for more information.

Your "visual artifacts" link does not load for me (time out).

If you are sure about the files, I believe you. I am just confused that the supposed sample image has in fact a larger file size and appeared to me a little bit better when I looked at some areas on a pixel level. I guess, inkbunny forces a default jpg-compression to generate its sample files which ignores that the original already is a jpg, even when the compression looses effectiveness that way.

Thanks for rechecking!

Lord_Ocean said:
Your "visual artifacts" link does not load for me (time out).

If you are sure about the files, I believe you. I am just confused that the supposed sample image has in fact a larger file size and appeared to me a little bit better when I looked at some areas on a pixel level. I guess, inkbunny forces a default jpg-compression to generate its sample files which ignores that the original already is a jpg, even when the compression looses effectiveness that way.

Thanks for rechecking!

MD5 hashes do not lie at least, they are absolute of saying if the file is what it's supposed to be. This is also one of those projects which should be constantly done, but nobody has enough interest, so we end up with these cases where years later the posts are getting replaced instead of almost instantly.

JPG files do not store the compression ratio that was used anywhere. There are softwares which can estimate what it was, but nothing objectively perfect.

As such, almost all websites simply have preset values which to use globally. So what that does mean is that if user uploads JPG file which they had more compressed than the value for samples is on the website, the image is going to be saved again with lower compression, which of course means filesize raises while quality lowers.

Pixiv actually uses 100 quality for samples, so all image files which are under 1200px dimensions are actually higher filesize there with samples compared to original uploads.

There is no such thing as to much asriel on e621.

Male, female or anything in between,
Fluff really knows how to draw DAT ASS!

Asriel is one of my biggest furry crushes!