Live data from Hacker News

WebP is so great except it's not (2021)

eng.aurelienpierre.com

201–210 of 411 posts

Re: WebP is so great except it's not (2021)

#201
post #132

Earlier quoted context omitted.

I'm on a 27" 4K IPS screen here and have to squint/zoom in to see the difference the author is writing about. While it's nice some people really care for the best result I think most people aren't going to notice or care about it.

I'm guess it's also true that HN is definitely the wrong audience for this post. As the author suggests, if you spend all day in VScode/VIM, you're among the segment of computer users who looks at images the least as a percentage of time spent on a computer.

Yes, but at least there are a decent amount of font 'connoisseurs' here ;)

Re: WebP is so great except it's not (2021)

#202

Earlier quoted context omitted.

But why should it be worse than JPEG in that respect? It's a much newer format and supposedly much better.

It's just a happy accident that the way JPEG compresses things and smooths them out visually happens to be an advantage in this particular edge case.

I wouldn't call it a happy accident; JPEG was carefully designed to look good for single-frames with the limitations of the human eye taken into account.

WebP is based off of a video format, and tradeoffs there are very different.

Re: WebP is so great except it's not (2021)

#203
post #4

I opened the first two pictures in separate tabs and switched quickly between them. There is zero difference. Tried it on two different monitors, Chrome and Firefox. Same with the pictures of the guy at the end. EDIT: The last comparison is webp twice, he linked it wrong. Here is the jpg one, still no difference: https://eng.aurelienpierre.com/wp-content/uploads/sites/8/20...

You either have a bad screen or limited eyesight, it's quite funny to me that this is the most upvoted comment.

There's definitely very ugly "banding" going on in the gradients on the WebP versions i say as someone who's worked extensively with UX and interfaces.

I'm on a M2 Macbook Air.

Re: WebP is so great except it's not (2021)

#204

Earlier quoted context omitted.

> I'm sorry, but I just can't read the text with those st and ct ligatures. The whole font is bad. It looks pixilated and blurry, which can only be explained by that being the intended look. It's bizarre.

It doesn't look pixelated on my machine, it looks pretty nice - here's a screenshot at 200% zoom: https://i.imgur.com/WE1tbIB.png The ligatures are definitely an unusual artistic choice...seems like the sort of thing you'd finally get used to 4 chapters into a book, but until you're immersed in it, it's quote distracting.

The font looks much less ill-defined and irregular if I tell the browser to enlarge the text to 130%. There are still some issues at that size.

But I'm not going to judge it by how it looks after I do a manual adjustment. The way it's presented is terrible. It somehow manages not to fit into the pixel grid of my 15-inch 3840x2160 screen. These are not large pixels!

Re: WebP is so great except it's not (2021)

#205
post #91

The banding is SUPER monitor dependent, its noticeable on my 4k monitor, super apparent on a different monitor with a terrible LCD panel, and not at all visible on my iPad. I wonder if the author took that into consideration.

Back in the early 2010's I had a cheap Dell laptop with a 6-bit panel and an integrated Intel GPU. Video on that device had incredible banding, almost all the time, because as I understand it, the Linux drivers were relatively immature and did not do any dithering. A few years later a driver update enabled dithering and the bulk of the problem went away.

As a video codec developer I was a little sad about that, actually. I had to start looking closer to see problems.

Re: WebP is so great except it's not (2021)

#206
post #7

It's such a shame Google decided to block adoption of JPEG XL: it's a strict improvement over classic JPEG (you can losslessly reencode JPEG to JXL and reduce the size, due to a better entropy coder in JXL!) and JXL has various other upgrades compared to 'classic' JPEG. In the meantime, let's hope AVIF or whatever manages to pick up the slack, and/or other browsers decide en masse to support JPEG XL anyway; that woul…

I must admit, I'm not sure why JPEG XL is viewed so favourably on HN, it's not something I know a ton about, but my understanding is that the big advantage of AVIF is that you can reuse hardware decoders built into devices for AV1 for the images. It being a strict improvement over JPEG is nice for the developers not having to go back to the source image for an upgrade, but that seems like a pretty small benefit that…

JXL is an image codec - it can afford to be less efficient (it's not! Other way around, rather) and not be hardware accelerated as its typical use case is not to present 30-60 images per second like in a video codec, it will not affect the battery life of a device in any meaningful way. Also, AV1 hardware decoding is far, far from ubiquitous so many users would not benefit at all from it.

But - back to JXL vs WebP:

I think Google had genuinely good intentions with WebP, but the effort was somewhat ruined by their culture: they relied too heavily on metrics which aren't always a good proxy for image quality, because humans looking at pictures don't scale, and Google does things that scale. We now have a codec with good metrics, but looks poor.

It's based on the intraframe coding of the VP8 format - a video codec - and I think it suffers from that. Looks OK in a video, but bad in stills where you have more time to notice its warts

Most importantly, it's almost always produced by recompressing a jpeg and causing a generation loss. I don't know of any phone or camera which produces native WebP (maybe some recent pixels? Dunno), and any professional device used in RAW mode usually implies the participation of someone who cares about the finished product and will not want WebP (and will resent when it's used without their consent by webmasters wishing to tick a box in pagespeed, as the author mentions). JXL has a lossless recompression mode in which it just replaces the Huffmann compression stage of an existing JPEG with something more modern, and this results in a pixel-accurate image which is 20% smaller than the original file - this already eat WebP's claimed space saving, and then some, with no generation loss. Based on this fact alone, there shouldn't even be a discussion.

....but let's have a discussion anyway. A JPEG -> JXL lossless recompression isn't conceptually new - Stuffit (remember them?) did it in 2005, with not enough traction sadly (unsurprisingly since there were patents and licensing costs). Basically it's _still_ a JPEG - if you decompress the final stage of a JPEG, and the final stage of a JXL (or a .SIF), you get the exact same bytestream. While yet another amazing testament of JPEG's longevity and relevance, it is also concerning: How could Google do worse than that??? When basically rezipping (with a modern algo) the existing DCT macroblock bytestream of a 30 year old codec beats your new codec, you should just trash it.

Edit: ...but I forgot to answer your question. Why is JXL viewed so favorably on HN? Because it doesn't suck, and we're sad that Google decided to be a roadblock, and pushing for their own thing, which instead sucks. At least AVIF is way better than WebP, even though it's a monster, computationally.

Re: WebP is so great except it's not (2021)

#207
post #4

I opened the first two pictures in separate tabs and switched quickly between them. There is zero difference. Tried it on two different monitors, Chrome and Firefox. Same with the pictures of the guy at the end. EDIT: The last comparison is webp twice, he linked it wrong. Here is the jpg one, still no difference: https://eng.aurelienpierre.com/wp-content/uploads/sites/8/20...

At 50 y/o my eyesight began to fail and yet the differences in the pictures are freaking obvious. As in: it's impossible to not see how huge the differences are.

And many people commented the same. These simply aren't small differences.

People who cannot see the differences or who only see them after taking a close look should realize something: there are many people for whom the differences are going to be immediately obvious.

Re: WebP is so great except it's not (2021)

#208

I might be missing something because I never delved into it, but my problem with WebP is I can't save images this way from my browser. Well, I can save them, but they don't show up when I try to view them on my system (Ubuntu Mate 20.04 on RPi4).

The problem is not the format, but the software / OS you choose to use. There are OS’s that have image format libraries, and once a codec is installed, ALL apps gain the ability to use it. This was first done in the 80’s, so if your Ubuntu 20.04 doesnt support data translations, maybe its time to switch to something else.

Re: WebP is so great except it's not (2021)

#209
post #126

I know this is not constructive and I'm sorry, but I just can't read the text with those st and ct ligatures. It makes me feel like the author is trolling with them and I shouldn't take the text seriously. I know that's an exaggeration but that's what the design makes me feel.

I was going to suggest blocking web fonts with uBlock Origin, but I get them even with JS and fonts blocked. Turns out a CSS rule does this: font-variant-ligatures: common-ligatures discretionary-ligatures contextual historical-ligatures;

I didn't really understand what everyone was talking about since I use a combination of uMatrix and noscript and noscript blocks webfonts by default. But, after whitelisting his site, eh, it didn't bother me. I guess it's what one is used to.

Re: WebP is so great except it's not (2021)

#210
post #176

Outside of photographers, how many people are looking at super high-resolution images on the web? Even images that might have high-resolution versions are usually converted to a shrunken image 600px wide to fit inside the website's theme scaffolding. Is that really even worth shaving 15% off the file size? If bandwidth matters, websites should look to reduce the volume of useless stock images littering their template…

Honestly, I would have agreed wholly with you until I spend 1 month volunteering in Kiribati. 2/3G is the norm there and even few KBs would make a difference. It reminded me a lot of my childhood with 28/56k modems :/

Additionally, I believe countries like India, Pakistan, Bangladesh, ... are in similar situation infrastructure wise (please correct me if I am wrong) and so for 1/2B people would benefit from a slimmer web.

Post reply on HN