xpostml10@lemmings.world to memes@lemmy.world · 1 day agoGoogle's WebPlemmy.mlimagemessage-square152linkfedilinkarrow-up11.17Karrow-down122cross-posted to: memes@lemmy.ml
arrow-up11.15Karrow-down1imageGoogle's WebPlemmy.mlxpostml10@lemmings.world to memes@lemmy.world · 1 day agomessage-square152linkfedilinkcross-posted to: memes@lemmy.ml
minus-squareValmond@lemmy.worldlinkfedilinkarrow-up1·12 hours agoSo you have no hard proof (no critic here, I’m just curious)? Not that it’s better but that your test images has the same quality. For the rest, thank you for the links and the time but that only explains how the compression works. If you want to know you could do fourier transform and see which kind of signals are cut out in one for example.
minus-squareILikeBoobies@lemmy.calinkfedilinkarrow-up1·edit-23 hours agoQuality improvements are that you can upload/download it without getting artifacts/pixel bleeding. JXL’s algorithm ensures that it’s a 1 to 1 transfer But if I draw a stick person 512x512, there isn’t an image format that will make it anymore than it is. That’s why we look at compression
So you have no hard proof (no critic here, I’m just curious)? Not that it’s better but that your test images has the same quality.
For the rest, thank you for the links and the time but that only explains how the compression works.
If you want to know you could do fourier transform and see which kind of signals are cut out in one for example.
Quality improvements are that you can upload/download it without getting artifacts/pixel bleeding. JXL’s algorithm ensures that it’s a 1 to 1 transfer
But if I draw a stick person 512x512, there isn’t an image format that will make it anymore than it is. That’s why we look at compression