Hardware JPEG decode isn’t as valuable on today’s hardware. CPU decode with SIMD instructions is fast and energy efficient. It’s such a negligible part of battery usage that it’s barely worth thinking about for fast modern CPUs.
Even formats like H.264 are being dropped from some hardware decode engines because it’s so easy to do on CPUs now, as can be seen even on the Raspberry Pi 5. Many software stacks will skip hardware JPEG decode even when available because it’s extra maintenance and debugging overhead for such a little gain.
Progressive decode is a feature that is good in theory but rarely used in practice. If you’re concerning yourself with power usage and memory footprints, progressive decode goes in the opposite direction. Users also don’t like progressive decode as much as developers think they will as it’s often perceived as something being broken.
I think WebP made reasonable tradeoffs for the way images are actually used and delivered.
> Even formats like H.264 are being dropped from some hardware decode engines because it’s so easy to do on CPUs now, as can be seen even on the Raspberry Pi 5
CPU decode efficiency most certainly was not the reason h.264 was dropped on the Pi
It also lost much of its quality advantage over plain jpeg, given the improvements in jpeg encoders: the jpegli encoder (from google's jpegxl team) has improved plain jpeg encoding significantly.
Most photo/design apps don't seem to be using SOTA jpeg encoders, though, so I think webp looks better in comparisons like OP's than it actually is. You maintain all the ultra-fast decoding of jpg when you use a modern encoder, too.
JPEG decoding is frequently included in systems with DSP blocks because it’s easy to add to a system that is already doing similar math for video decoding.
It’s not always used when available. It can be a lot easier and safer while being fast enough to do it on the CPU. If the operating system has an image decoder path that will handle everything for you then it might be used (Apple hardware for example) but if you have to go out of your way to implement the API, common software will favor software decoding because nobody wants to maintain code testing branches for different hardware image decode platforms.
So as every time I read something decrying how webp is allegedly "annoying", the conclusion is that the problem was never about webp itself but the tooling and ecosystem to handle it
Is there a real distinction? It can be a great format, but if actually using it sucks then in practice it's a pain to use/encounter. (I don't actually have this problem personally - all my apps seem to handle it fine - but the argument seems reasonable.)
Yes, it's important to distinguish annoyances of the codec from annoyances of the ecosystem.
Ecosystem being slow to adopt it is something that can be fixed or improved, but also more importantly (given jpegxl & avif comparisons) are things that will impact any new image format.
Meanwhile annoyances of the codec tend to be things that are dramatically harder to deal with. Like HEIC being a license minefield and also slow as shit. Or that many of these newer image codecs do not have ways to work with them in a memory efficient manner like jpeg does (see eg libjpeg-turbo's sample size & skip scanline options). Or WebP insisted on being in a RIFF container for... uh... absolutely no fucking reason whatsoever afaict. Fortunately RIFF is so trivial that this overhead doesn't actually matter in practice, but then HEIC & AVIF dials up the annoying container bullshit to 11 thanks to their video heritage with ISOBMFF
Perhaps it's a matter of who hits the pain point? Ugly specs make annoyed devs, licensing problems make distributors unhappy, and ecosystem issues leave users unimpressed. (With some spillover between categories, of course.)
Yeah it's purely the fault of the tens of thousands of application devs that have been too stupid to recognize the obvious and immense benefits of webp over existing widely adopted formats /s
Firefox actually silently and automatically converts images to webp in some cases when you Save As them, with no option to revert to the format you see in the DOM. It’s fucking infuriating. I hate webp because the support is so spotty.
I don't think that's firefox doing it? I'd guess it's some CDNs doing the conversion. Or the URL says .jpg but when you look at the mimetype sent by the server it's actually webp. Extentions are meaningless for browsers.
I was amazed at how much better compression webp offered over jpg and png.
Using it in a web-browser 360 panorama viewer enabled pretty substantial latency and storage improvements.
close to 4x iirc.
ditto large "xray" style orthographic images of building scans generated from pointcloud laser scan data.
avif was even better compression for some cases, but seems to take a long time to generate, and less widespread browser support. Will look at jpeg-XL .. but Im guessing it will be a while until it has widespread support
https://squoosh.app/editor is a great app to compare how well an image compresses with newer formats like AVIF and WebP, compared to JPG/PNG.
It feels more practical now to have a fast loading page with large images now without too much effort. It was far too much of a burden before to have to provide fallbacks to older formats, on top of different images for different screen sizes/densities.
Google aren't pushing WebP. Google Docs / Slides still don't support it.
I appreciate Google isn't a monolith, but that lack of consistency makes it really hard to shake the impression certain tribes within it are at war with each other.
It's not war, it's indifference. Why should Docs give a damn what Chrome's priorities are? The only reason they would is if there were effective leadership ensuring company-wide alignment, but there is no one filling that function.
Because if they did company wide alignment, you would kill their promotion pipeline. Want harmony? Then in Google you build your own app (see N+1 chat apps) to get that feature, get promoted and dump the project on someone else.
Just as with JPEG2000, I'm still not convinced the additional complexity is worth it. There seems to be a lot of immature technologies being pushed in the past few years, when IMHO the Web should've stabilised long ago on the basics. But I guess people need things to do to keep them employed... and Big Tech needs to maintain its monopolies.
The old JPG/PNG choice had a big problem: JPG handles sharp edges, including text, badly. PNG produces huge photo files. Images containing a mix of both don't work well in either format.
By contrast, when using a modern format like WEBP or AVIF, a single encoder configuration produces decent looking results across a variety of input images.
Author here: I'm also very much looking forward to using AVIF (and also JPEG XL), but it's another tooling question. I'm really hoping that Pixelmator Pro (now owned by Apple) will support them both soon. Photoshop (which I've recently dumped) supports WebP and AVIF, but not through the "Save for Web" or "Export As" flows. Having a nice export flow where you can easily compare compression vs. quality levels is really important to me, and I suspect to many others.
When it comes to animations, is there much of a reason to use avif over webm? I suppose an avif fits into CSS and <img> without further code changes, but that's about it as far as I can think.
I just switched to AVIF after several years of using WebP. Main reason for me is HDR support, but the smaller encoding is nice. Just beware of lossless AVIF, that's a trap.
One forum I use does not embed webp links. They show up as links, period, and it's a damn waste of my time to convert. Small is beautiful unless your forum is run by two people, not having the time to update code all the time.
Many image links that work for search engines don't work for the site and I have to download and then upload to a host to get an embeddable image link.
Author here: as I wrote in the piece, I've recently dumped Photoshop for Pixelmator Pro. It doesn't currently support AVIF. Photoshop does, but not in a way that makes it easy to optimize an image for web export. JPEG XL is also exciting to me, but it's not generally available across browsers. WebP is the current best image format for my use-case, but I'm hoping others supplant it soon.
The one annoying thing about WebP is that you can no longer see from the file extension whether some image is lossy or lossless.
The other annoying thing is the name, because surprisingly we use images in non-web-related contexts as well. Admittedly, PNG technically has a very similar issue, but it’s less in your face.
Webp does not have hardware decode and does not support progressive.
I speculate the asymmetric push is a case of it solving Google’s problem (bandwidth cost) but not the end user’s problems (battery and latency).
Hardware JPEG decode isn’t as valuable on today’s hardware. CPU decode with SIMD instructions is fast and energy efficient. It’s such a negligible part of battery usage that it’s barely worth thinking about for fast modern CPUs.
Even formats like H.264 are being dropped from some hardware decode engines because it’s so easy to do on CPUs now, as can be seen even on the Raspberry Pi 5. Many software stacks will skip hardware JPEG decode even when available because it’s extra maintenance and debugging overhead for such a little gain.
Progressive decode is a feature that is good in theory but rarely used in practice. If you’re concerning yourself with power usage and memory footprints, progressive decode goes in the opposite direction. Users also don’t like progressive decode as much as developers think they will as it’s often perceived as something being broken.
I think WebP made reasonable tradeoffs for the way images are actually used and delivered.
> Even formats like H.264 are being dropped from some hardware decode engines because it’s so easy to do on CPUs now, as can be seen even on the Raspberry Pi 5
CPU decode efficiency most certainly was not the reason h.264 was dropped on the Pi
It also lost much of its quality advantage over plain jpeg, given the improvements in jpeg encoders: the jpegli encoder (from google's jpegxl team) has improved plain jpeg encoding significantly.
Most photo/design apps don't seem to be using SOTA jpeg encoders, though, so I think webp looks better in comparisons like OP's than it actually is. You maintain all the ultra-fast decoding of jpg when you use a modern encoder, too.
Is hardware decoding even commonly used for still image codecs?
JPEG decoding is frequently included in systems with DSP blocks because it’s easy to add to a system that is already doing similar math for video decoding.
It’s not always used when available. It can be a lot easier and safer while being fast enough to do it on the CPU. If the operating system has an image decoder path that will handle everything for you then it might be used (Apple hardware for example) but if you have to go out of your way to implement the API, common software will favor software decoding because nobody wants to maintain code testing branches for different hardware image decode platforms.
Only for HEIC realistically, and that only still causes it to struggle to match jpeg or webp performance anyway
So as every time I read something decrying how webp is allegedly "annoying", the conclusion is that the problem was never about webp itself but the tooling and ecosystem to handle it
Is there a real distinction? It can be a great format, but if actually using it sucks then in practice it's a pain to use/encounter. (I don't actually have this problem personally - all my apps seem to handle it fine - but the argument seems reasonable.)
Yes, it's important to distinguish annoyances of the codec from annoyances of the ecosystem.
Ecosystem being slow to adopt it is something that can be fixed or improved, but also more importantly (given jpegxl & avif comparisons) are things that will impact any new image format.
Meanwhile annoyances of the codec tend to be things that are dramatically harder to deal with. Like HEIC being a license minefield and also slow as shit. Or that many of these newer image codecs do not have ways to work with them in a memory efficient manner like jpeg does (see eg libjpeg-turbo's sample size & skip scanline options). Or WebP insisted on being in a RIFF container for... uh... absolutely no fucking reason whatsoever afaict. Fortunately RIFF is so trivial that this overhead doesn't actually matter in practice, but then HEIC & AVIF dials up the annoying container bullshit to 11 thanks to their video heritage with ISOBMFF
Perhaps it's a matter of who hits the pain point? Ugly specs make annoyed devs, licensing problems make distributors unhappy, and ecosystem issues leave users unimpressed. (With some spillover between categories, of course.)
Does Windows already display WebP files in File Explorer? :-|
Yes. https://imgur.com/a/lRSM128
It has done for some years
> the conclusion is that the problem was never about webp itself but the tooling and ecosystem to handle it
That is true in life.
EVs aren’t annoying, the charging infrastructure is.
Etc
Yeah it's purely the fault of the tens of thousands of application devs that have been too stupid to recognize the obvious and immense benefits of webp over existing widely adopted formats /s
Firefox actually silently and automatically converts images to webp in some cases when you Save As them, with no option to revert to the format you see in the DOM. It’s fucking infuriating. I hate webp because the support is so spotty.
I highly doubt that is what's happening. More likely the server is serving a webp under a .png filename or something.
I have observed the behavior you suspect many times.
I don't think that's firefox doing it? I'd guess it's some CDNs doing the conversion. Or the URL says .jpg but when you look at the mimetype sent by the server it's actually webp. Extentions are meaningless for browsers.
It's not Firefox, it's the server. Edit and resend the request with header "Accept: image/jpeg" only and you'll most likely get a JPEG file.
That seems more likely. I still hate webp.
Does it actually request the image that it already downloaded again to save it to a file? Seems rather silly?
I was amazed at how much better compression webp offered over jpg and png.
Using it in a web-browser 360 panorama viewer enabled pretty substantial latency and storage improvements.
close to 4x iirc.
ditto large "xray" style orthographic images of building scans generated from pointcloud laser scan data.
avif was even better compression for some cases, but seems to take a long time to generate, and less widespread browser support. Will look at jpeg-XL .. but Im guessing it will be a while until it has widespread support
https://squoosh.app/editor is a great app to compare how well an image compresses with newer formats like AVIF and WebP, compared to JPG/PNG.
It feels more practical now to have a fast loading page with large images now without too much effort. It was far too much of a burden before to have to provide fallbacks to older formats, on top of different images for different screen sizes/densities.
Google aren't pushing WebP. Google Docs / Slides still don't support it.
I appreciate Google isn't a monolith, but that lack of consistency makes it really hard to shake the impression certain tribes within it are at war with each other.
It's not war, it's indifference. Why should Docs give a damn what Chrome's priorities are? The only reason they would is if there were effective leadership ensuring company-wide alignment, but there is no one filling that function.
Yeah, the classic image still holds true.
https://www.globalnerdy.com/2011/07/03/org-charts-of-the-big...
Because if they did company wide alignment, you would kill their promotion pipeline. Want harmony? Then in Google you build your own app (see N+1 chat apps) to get that feature, get promoted and dump the project on someone else.
This is actually a good counter argument for 'Google is pushing X'. If they were pushing X, they'd make all their apps fall in line.
Contrast this to, say, Apple - which when decides to push something, they go all in. And that's an example of top-down decision making.
Just as with JPEG2000, I'm still not convinced the additional complexity is worth it. There seems to be a lot of immature technologies being pushed in the past few years, when IMHO the Web should've stabilised long ago on the basics. But I guess people need things to do to keep them employed... and Big Tech needs to maintain its monopolies.
At what year would you say the web should have stabilized on the basics?
As in JPEG, GIF, and PNG? Early 2000s. Maybe with a subset of SWF as a (animated)/vector format instead of horribly bloated SVG.
The old JPG/PNG choice had a big problem: JPG handles sharp edges, including text, badly. PNG produces huge photo files. Images containing a mix of both don't work well in either format.
By contrast, when using a modern format like WEBP or AVIF, a single encoder configuration produces decent looking results across a variety of input images.
Avif.
Especially for animated images. Comparing webp or avif sizes to gif is easily a factor of five to ten when animated.
Author here: I'm also very much looking forward to using AVIF (and also JPEG XL), but it's another tooling question. I'm really hoping that Pixelmator Pro (now owned by Apple) will support them both soon. Photoshop (which I've recently dumped) supports WebP and AVIF, but not through the "Save for Web" or "Export As" flows. Having a nice export flow where you can easily compare compression vs. quality levels is really important to me, and I suspect to many others.
When it comes to animations, is there much of a reason to use avif over webm? I suppose an avif fits into CSS and <img> without further code changes, but that's about it as far as I can think.
Since all modern browsers support AVIF, I'd skip WEBP and just use AVIF (possibly with a fallback for JPG/PNG).
I just switched to AVIF after several years of using WebP. Main reason for me is HDR support, but the smaller encoding is nice. Just beware of lossless AVIF, that's a trap.
One forum I use does not embed webp links. They show up as links, period, and it's a damn waste of my time to convert. Small is beautiful unless your forum is run by two people, not having the time to update code all the time.
Many image links that work for search engines don't work for the site and I have to download and then upload to a host to get an embeddable image link.
X doesn't support animated webp, slightly annoying.
XPM works just fine though.
Why? Just skip webp altogether, an annoying footnote that caused more issues than it solved.
Go straight to jpeg xl and/or avif when the they are mature.
Author here: as I wrote in the piece, I've recently dumped Photoshop for Pixelmator Pro. It doesn't currently support AVIF. Photoshop does, but not in a way that makes it easy to optimize an image for web export. JPEG XL is also exciting to me, but it's not generally available across browsers. WebP is the current best image format for my use-case, but I'm hoping others supplant it soon.
The one annoying thing about WebP is that you can no longer see from the file extension whether some image is lossy or lossless.
The other annoying thing is the name, because surprisingly we use images in non-web-related contexts as well. Admittedly, PNG technically has a very similar issue, but it’s less in your face.
Annoyingly Google voice doesn't support it at all, either. I've installed a firefox extension[0] that saves any image as .JPG,.PNG, or .WEBP
[0] https://addons.mozilla.org/en-US/firefox/addon/siat/