Not saying you're wrong on the main message (and those examples are with filters rather than true CRT) but it's still impressive how it can make the pixel art better looking in those cases.
Not so clear in the article is that also with consoles the game pixel grid didn't match the CRT grid (e.g. horizontally the number of "pixels" is similar but not 1:1) so just by geometry you have blending of one pixel data into several CRT "pixels" (plus varying intensity even in a single phosphor patch as the analog RGB signals rise and fall in intensity).
As for CRT filters one can argue how realistic they are but they show that scanline gaps and blending at least can produce amazing perceptual improvements as shown in this video: https://youtu.be/nw2QfPREu-Q
At peak career ages, the 75th percentile household income is almost $200k and the 90th percentile is almost $300k. That’s exactly the range you need to tax. Across all ages, households earning between $150k and $800k earn half of all income—$7 trillion.
The republican program was in large part vetoed by Clinton though and the spending cuts that passed were largely offset by the tax cuts. From what I read the reduction in deficit is mostly due to the dot-com boom, helped with the "deficit reduction act" of 1993 that was pushed by Clinton and happened to hike taxes at the right time to tax the boom.
It's a great test of cognitive abilities. There are many claims that current LLMs surpass humans in cognitive abilities so it's noteworthy that they underperform on that test.
Letting the model execute a chess program (that it wrote) would make sense if you're measuring its economic potential, but for cognition that would be cheating just like if you let a human run a chess program. The fact that the human would have a much harder time writing a useful program is irrelevant.
It isn't. Stockfish running on your laptop can beat every human being on earth easily at chess. It's not intelligent _at all_ in any sense that matters.
Well it's not a perfect test so you need a bit of care in how you use it. If you have no idea what the subject is doing, then you don't know if you're measuring cognitive ability or something else (like cheating ability, or algorithmic sophistication or whatever). But failing the test is a pretty clear sign of certain cognitive abilities being poor.
> The fact that the human would have a much harder time writing a useful program is irrelevant.
Why?
There's a box.
You give it a problem, and it comes up with a solution.
Why does it matter to you if the box is strictly a LLM, or if the LLM can write code that it executes?
Even neater if the box is self-contained with a local model. You provide electricity, and it comes up with solutions. Why does it matter if it can do chess "in its head", or if it has to use scratch paper?
The question is to what end? This is a benchmark task, because playing chess, or solving other well-understood problems is more of a party trick than it is useful.
If you let the LLM write a chess program, which it can ONLY do because there are already so many chess programs out there, then the benchmark becomes about recall of popular program source code, not chess.
Neither. As I said I want to measure cognitive abilities.
Your "ability of the box" is like "economic potential" in my previous comment. If that's what you want to measure, fine. But I want a deeper understanding: what is the thing doing, how is it solving problems? I want to get a sense of its abilities that is richer than a one-dimensional scale.
I agree that it's a fascinating to crawl inside an LLM, and also to crawl inside of a human, and try to understand the processes and limitations. Like, Phineas Gage is one of the most remarkable learning opportunities we ever had.
That said, it's really weird to me when people use (and judge) LLMs one way... and won't try using them another way.
Like, to judge their utility, I think we should be open to letting them write code, and use the code they produce.
Otherwise, it's like judging a Chromebook without an internet connection. Like, this was one of the most dishonest ads I've ever seen: https://www.youtube.com/watch?v=gDy9AUQJ3Fg
This lamp, without a working power outlet? It really doesn't do anything...
>> There are many claims that current LLMs surpass humans in cognitive abilities
Where? By whom? This is certainly not (yet) the general consensus, as I understand it. Are you taking the most optimistic / untethered comments as the strawman against which you feel the need to argue?
Why the aggressive tone and the strawman rhetoric? I never said there was a consensus. Yes I'm talking more about the "optimistic" commenters and pointing out that this chess thing is a good datum to temper their enthusiasm. What's wrong with that?
Also these claims are not completely without merit, it's just that LLMs seem to excel at specific "cognitive" tasks and it's interesting to see where they fail.
The article makes a good point: JPEG XL is amazing but not specifically for the typical Web use cases, compared to AVIF. But the conclusion doesn't follow. Having an excellent and versatile format supported by browsers is very useful!
Maybe a lossless re-encoding of my website JPEGs to gain 20% size is not worth the 33% longer decode, and maybe the website doesn't need very high resolution images or more than 12 bits per channel. But my personal archives can definitely use that and it's great that such files can be viewed everywhere without specialized tooling, including served on the intranet and indeed the Web.
In summary JPEG XL easily beats AVIF on the not-so-long tail of use cases and having compatible viewers everywhere is quite useful. And with JPEG XL becoming part of PDF, it will soon be common use cases too.
As for the rest of the article: there are interesting points in the benchmark section but I think it's too early in the JPEG XL adoption cycle to draw conclusions. One thing bothers me though: There is a single picture comparison (with tag line "results speak for themselves") and I find it quite misleading: it's just picking one point on the "bytes received" line that looks best for AVIF compared to JPEG XL. Try it yourself and you'll see JPEG XL shows something already at 2KB while AVIF has nothing until 8KB, and JPEG XL looks better than AVIF after 100KB.
+1, this is super unfair to show the one point in time where AVIF progressive looks better - right after it receives its 'preview' (which as you say is 4x as big as JPEG XL's).
I value integrity, especially when communicating results. This is shameful.
“For progressive, I think it's better to show the user something that's obviously a preview, but still has enough detail to be able to know what the picture is of.
If you're showing the user something that they may mistake for complete but poor quality, that's a bad experience.”
I note you made no response to the objection about representing a temporal sequence with a single, nonrepresentative and cherry-picked, screenshot.
As to bad experience, that seems like a legit personal preference, but disturbingly un-nuanced and absolutist if intended to apply to everyone.
When on a train with spotty wifi or frequent tunnels, I would actually prefer to see whatever arrived. Someone on prepaid data might want to truncate at ~100 KB, no matter what the website owner or browser PM thought.
For progressive, I think it's better to show the user something that's obviously a preview, but still has enough detail to be able to know what the picture is of.
If you're showing the user something that they may mistake for complete but poor quality, that's a bad experience.
Depends on the user. Some users are more sensitive for flashing images. Some users have very slow or unreliable network connection. Some sites have bigger images. Some sites (such as ecommerce) have higher quality and resolution needs.
Some users have uncorrected astigmatism. In the general population, the average uncorrected astigmatism typically falls between 0.50 and 0.75 diopters (D), and ~15 % more than 1.0. The high astigmatism people will see images blurred in any case and don't benefit from the fine structural detail JPEG XL can give. People with glasses and the lucky ones with low astigmatism will see the details clearly.
Sometimes passionate ends of different quality of vision seem to have a discussion on what makes sense as the web quality, with one emphasizing quality 40 being more than enough and another requesting a quality of 95.
The same kind of individualism exist also in temporal cognition and perception.
The quality of the preview layer is somewhat under the control of the user though.
And that comparison was made for a JPEG-XL focused site, and uses quite large images that are then scaled down dramatically (at least on mobile) for display which is probably not an ideal visual test anyway (but possibly necessary for there to be enough time to see any difference)
It's an absolutely ridiculous point, and the guy you're responding to (who has commented almost 40 times in this discussion) has a hilariously slithery opinion that somehow always finds a way to hold every finding against JXL.
Personally I don't like progressive rendering at all, but it's interesting that this same guy claimed elsewhere that it's super important on a very bad connection to get an early hint of what an image is. When someone points out that the "the results speak for itself" cherry picked example is absurdly misleading, and in most cases JXL is vastly superior at that case, instantly their opinion changes such that it shouldn't look like a poor quality image but it's somehow also a win that AVIF takes longer to give you anything at all to look at.
Which is incredibly weird, given that AVIF also gives you a terrible quality image, it just takes longer to show it. So it sounds like they're completely against progressive rendering? Unless they think it gives AVIF a benefit?
But it's an amazing demonstration of someone with an agenda. Utterly bizarre.
I also called AV1 short-sighted for not including an alpha channel, and corrected someone who claimed jxl-rs was single-threaded. No agenda here - I'm just correcting misinformation and stating my opinions.
Agreed. Browsers should support general-purpose image formats. I'm not interested in another “web codec” that's not good for anything but passive delivery.
> Maybe re-encoding my website JPEGs to gain 20% size is not worth the 33% longer decode
I changed all my .jpg files to .avif a few years ago. The benefit I get is much more than 20%. Now there is some loss when I use high compression with .avif, in particular via gimp - but other than that, I gained a lot of file size here, definitely much more than 20%. It also depends on the image at hand; I noticed that images with a ton of details, such as if you photograph a garden, takes more space. But for simpler images, I have easily gains of about 60% or 70% compared to jpeg, with no or almost no loss in quality. Whereas if I were to use jpeg compression, the quality loss would be insanely high. When I noticed this, I decided to abandon .jpg for my own use cases. I had to choose back then between avif and webp and while webp is fine, I found .avif was slightly better for the use cases I had.
I don't quite know how much jpeg xl performs, but you mentioned JPEG so I had to comment on that 20% statement as I found it way too low.
I also find the claim "longer decode" not correct. I have some photographs of gardens and the file size is huge, in JPG. When I compressed and changed these into .avif, the resulting page loads soooooooo much faster now when having used avif and a considerable but acceptable compression of it. It is just no comparison at all - avif beats jpg with its eyes closed.
Sorry if that was confusing, I have edited my post to clarify a bit: That sentence was about a feature specific to JPEG XL, which is that if you have JPEG files you can re-encode then as JPEG XL to save space without any loss of quality. It's nice for peace of mind, you can bulk convert a large archive and you know you still have the same pixels everywhere (later you can even turn them back to JPEG without loss if you want). AVIF cannot do that. Now if you are OK with some loss in the conversion, you can also do much better than 20% with JPEG XL.
Worth mentioning that someone else in the comments converted their large library of JPEGs to JPEG XL for exactly this reason and only realized afterwards that they lost a ton of image metadata from the EXIF tags. This probably isn't what you would assume would happen for a bulk 'lossless conversion' so if you're going to do this then make sure you're validating the metadata too (if it matters to you).
It sounds like you lossily re-encoded lossy JPEGs to AVIF, am I reading that right? If so, you're automatically suffering from generational loss there, it can't be the basis of any coherent argument.
The entire article is extremely cherry picked, and it's notably that there are a couple of people leaving dozens of comments in here each (which is odd as normally HN gates that sort of gross overrepresentation). Like >10% of comments in this 300+ comment discussion is one single person, endlessly changing positions based upon what makes JXL worst.
It's bizarre. I've never seen a format see such ridiculous attacks.
>Maybe a lossless re-encoding of my website JPEGs to gain 20% size is not worth the 33%
I was unpleasantly surprised when jxl's reference encoder dropped critical exif tags for no reason whatsoever from my personal archives. Turns out their lossless preset wasn't really lossless.
And then webp to JXL conversion also produced different viewing experience results as ffmpeg's webp decoder wasn't handling ICC correctly.
I'm now very cautious of any "lossless" re-encodes.
> Maybe a lossless re-encoding of my website JPEGs to gain 20% size is not worth the 33% longer decode,
And pray tell how much of the waiting for page is used on decoding image (vs transferring) and how much faster the image gets on the machine when it's 20% smaller on average connection?
My guess is that time saving from bandwidth decrease more than compensates for that
Taking bike.jpg and bike-repackaged.jxl in that demo, on my Pixel 10 Pro in Firefox Nightly, the size and decoding times are:
bike.jpg - 147 kB - 10ms
bike-repackaged.jxl - 126 kB - 33ms
So, I save 21 kB, but pay 23ms.
That extra 21 kB isn't part of a new request, so when it comes to download performance, we're talking about an active response. At 3g speeds, in 23 ms you can download somewhere between 20-118 kB, so at best, it evens out.
If it's a lower end device, the decode delta increases. If it's a faster connection, the 21 kB saving matters less. If the image was cached, then I'm just paying the decode overhead.
People really, really like to try to save up on the matches, so to speak. Sometimes literally: there was a short period of time during Khrushchev's rule in the USSR when the natural gas was made free for the populace. Well, the matches still cost money (1 kopeck for a box of 50 matches), so lots of people would just not turn the stove off, instead always having it burning the smallest possible flame when not in use.
All that to say, "I save 21 kB, but pay 23ms"? You'll save more if you stop sending your favourite web font over with your page and use the system fonts, and trim down your site's JS.
Yeah I didn't find that helpful. What I remember from Feynman's lectures is that photons still travel at "full speed" c between atoms, but if you look at the global progression of light as photons get absorbed then emitted it progresses slower than c.
You cannot treat light as particles in that scenario. The primary wave gets absolutely and completely delayed, with no part getting ahead. It's not some photons doing something with a certain probability and then causing a macroscopic effect once the probability goes towards 1 once you passed sufficient matter.
What Feynman does (where this confusion comes from) is that you can look at discrete wave packets (i.e. photons) and the math comes out the right way for the primary wave if you assume that only some of these wave packets get phase-shifted, and add all elementary waves together afterwards.
But still, it's photons as "wave packets" that influence the whole system, not photons as independent particles that either bounce on something or don't.
Looking it up I was actually remembering from Feynman's QED book (path integral approach). I don't have it at hand but I think it's a description based on photon particles. Anyway I agree my description with photon travel "between atoms", emission and absorption was quite bad (especially if you think absorption and emission as slow incoherent processes instead of a general way of describing interactions which is what I meant).
But still I think saying "the primary wave gets absolutely and completely delayed" is not helpful. Using a wave description as in Feynman's lecture[1] is more enlightening: the incoming wave travels at "full speed" through the medium, but doing so it interacts with atoms such that they emit an additional wave, and the sum is a slower wave.
You can say it's the same since there's only one electric field in space and so the only "real wave" is the sum of all effects. But I find it quite helpful to think that one of the components in this sum is the original wave traveling at the speed of light in vacuum, also in the space occupied by the medium.
I don't think that can be strictly true. No matter how dense the material is some photons have a chance to get through unimpeded through something like tunelling. Practically unlikely but mathematically possible.
This is only tangentially directed at your reply, but this is the kind of statement that I'll sometimes hear other parents say, and it makes me think that they've completely forgotten what childhood was actually like. Like, the idea that normal, fun activities and sports are bad because they bear some abstract resemblance to adult violence feels very contrived and misses what childhood play is about.
Yeah I'm not saying it's bad for the children, but it's unpleasant for me. It seems violence is a concept that gets deeper and uglier as you grow. For young children it's often abstract and maybe harmless. I just read a book to my child that turned out to have violent scenes that horrified me a bit but he barely thought about it and only in technical terms (like where is that artery ripped by the wolf). Same with people that had fun watching horror movies as teens, then find them unpleasant when they're older.
Maybe they did not forgot what the childhood was like. Instead, they remember their own childhood rather then yours. Moreover, they are not you, your exact copies.
It's not really that abstract. You smash the shit out of a cardboard box and candy comes out. Breaking things is not fun. I also hate youtube videos of people breaking stuff for fun.
I don't know. I broke a lot of things when I was a kid, mostly because I was excited to see how they work, but have no knowledge of their value or how to safely disassemble them. I don't do it as an adult because I now have that knowledge. So don't judge a kid as an adult.
I tried reading the English Wikipedia article "Piñata" but it is a mess of cringe information, especially the 16th century background. It seems that original piñatas were clay pots, before papier-mâché came into vogue.
I was going to say, that around the American Southwest, you can find piñatas for sale that are not only abstract, but also plenty of Spongebob and Disney and MCU characters, as well as "political figures that are not particularly well-liked" by Chicanos.
I wouldn't say "violence", rather, an opportunity to demolish something on (good) purpose. Same reason children will be glued to the window if workers outside are going at it with jackhammers, or, better yet, an excavator is crushing and scraping off the asphalt. (Yeah, I remember exactly how awesome that was.)
is play not an indirect method of practicing the physiological mechanisms designed for violence ? baby lions are both happy to play with other animals, and to eat them ... which inevitably progresses to predation only ... and humans are the apex predator ...
Those who don't lose the fun of hitting things graduate to hitting things with swords and end up in the SCA (not the coffee one).
I see the point about violence but I'm ok with it because it's based on fun and not anger. Olympic archery, javelin, or even fencing have an obvious past in warfare but today they've become sports.
I can only speak for some of the European versions of SCA-like clubs but there's a lot of pride-flag flying strictly-no-fascists fun activity with a dose of pretending to be children again.
That's the problem. Hitting a tied up animal until it breaks is clearly a violent affair, similar how hanging puppets in nooses or putting a doll in a guillotine.
I'm sure that there are kids growing up in places where bullfights are common thinking that torturing and killing a bull in a party-like atmosphere is fine.
I mean yeah, absolutely. But a child growing up in a western society is unlikely to encounter either of those. But generally I do see Pokemon as violent and I wouldn't let my 5 year old watch it quite yet. When he's a bit older I hope he enjoys playing through the original games and the OG tv show with me, but I see parents playing the TV show for their 4-5 year olds and I think that's just too young. Same with Spidey and friends and few other shows made "for children" which absolutely promote violence despite the colourful aesthetic.
Children have no concept of violence or empathy or right or wrong. Children often are extremely violent to other kids and sometimes even adults. They kick they scream they punch and they destroy objects. It's on the parents to teach them these concepts and its fun to have settings in which that behavior is allowed.
> Children have no concept of violence or empathy or right or wrong
This is not correct. I remember having a sense of shame when I was 4 when I tried to hurt another kid. Kids will also hide the things they do wrong, because they know it is wrong.
Or they know they'll be told it was wrong. Children start knowing nothing then they slowly absorb what they're told again and again, and what they see around them until they feel it. They go from knowing nothing, to knowing what the adults told them, to internalizing that and feeling it rather than just knowing it (maybe these 2 concepts fully overlap in other people's minds).
But until they do get all of that, they'll spend a lot of time plucking wings off of insects, burning ants with a magnifying glass, pulling the cat by the tail, or throwing rocks at dogs and frogs.
I can't say whether it was or wasn't true for you. But I can say your certainty is misplaced.
Most of your memories and everything you think of now about early childhood were formed long after it. They were created from bits and pieces of real memories, stories, backsplanations for things that happened later, retrospective, etc. Even if you had now an accurate debug dump of your child brain, you'd still not be able to process that the way you processed it back then.
Your kid brain was a black box, a CPU with a different architecture running code that you've never before or since. You later tried to recreate that code for the CPU you have now, based on input from others who've also never seen that original code or have access to that type of CPU. And you are certain you recreated it accurately?
How could it be possible for anyone to accurately reflect on the mechanisms shaping their emotional reaction to a specific situation when they were 4 years old? It should be transparently obvious that this is unknowable. First, by the time we begin to form memories, these influences have already been shaping us for months to years. Second, even if at 4 years old you remembered every moment from birth to now, you still wouldn't have the ability to actually reflect on all that data. Most adults aren't even capable of reflecting accurately on the influences affecting their current decisions!
I have other memories when I was 4, and talked about them with my parents years later. Another significant one is I wandered away and got lost in London, and some cops noticed me and picked me up and took me to the station.
My parents filled in some details, some of which I did not remember.
I don't doubt that you have memories from when you were 4, I have memories from that time too. And I'm sure you felt the shame you described. But I know that it's impossible for a 4-year-old to accurately decipher why they feel shame when they do!
You disagreed with sevenzero when they claimed that concepts like violence and right and wrong are actively taught by parents instead of being innate to children. Assuming you read all three lines of sevenzero's comment before disagreeing: the perspective of your 4-year-old self is not a reliable source, because 4-year-old children cannot accurately differentiate between innate or taught feelings.
I would certainly agree that one gets taught more refined and nuanced notions of right and wrong, but those are innate feelings. At least for me.
I recall seeing a TV show where they studied very young people by putting them in right and wrong situations, and they definitely knew what was right and wrong.
> I would certainly agree that one gets taught more refined and nuanced notions of right and wrong, but those are innate feelings. At least for me.
How could a person who was taught social concepts of right and wrong since birth ever hope to reliably determine which parts of said concepts in their own mind are innate, and which aren't? Human minds are infamous for being terrible at introspecting their own mechanisms at the best of times.
> I recall seeing a TV show where they studied very young people by putting them in right and wrong situations, and they definitely knew what was right and wrong.
Unless said very young people were raised by wolves, chances are that they've been taught these concepts since birth as well. You can't determine the innateness by only looking at people who share this bias.
Because we have proof that humans who aren't taught a subset of social behaviors never show any of them. It doesn't matter what we've bred foxes to do - "sense of right and wrong" is not the same as "social behavior", and innate traits in one species can be non-innate in another. You cannot analyze a cantaloupe to determine the taste of bananas, no matter how hard you try.
It is important to understand our own human limits. We cannot reason our way through our own minds, we can only accept our biases and try to work around them. But the first step is being humble enough to accept that human intuition is no substitute for actual research.
No, not all kids are a little sociopaths. They in fact hide things they have no experience of being reprimanded or punished for, because they feel bad about them. Kids do care about other kids, animals or adults in their lives. They do feel empathy - underdeveloped and sometimes weird, but they display it very clearly.
I am an incredibly non-violent person, but have a toddler has really made me work hard to understand the difference in what associations we attach to "violence" as adults.
I try to help him understand this slowly, but a lot of "play" can look violent to my eyes, and I struggle to remember that. A 3 year old just doesn't see it that way yet, and that's fine. But it's hard to remember that in the moment.
Because it's only half true. Yeah we've used phrases like those before, but if you ignore certain very specific phrases or characters for a moment, a lot of the more ambiguous things people are identifying as AI-speak comes from mixing jargon from different domains in a way humans normally don't do. Like the short punchy sentences are from marketing, not comments or technical writing like they've started showing up in.
Or at least he's being unintentionally influenced by it. Just out of curiosity I jumped a few years back in his comments looking for that pattern, and I don't really see it.
By pure luck I even landed on a comment that uses that same word with a similar sentiment, except back then he wrote it as a full non-punchy sentence: https://news.ycombinator.com/item?id=31414288
> Heh. We are now at the point where people might end up deliberately inserting typos and bad grandma into their writing, just to signal they aren’t AI.
Except what humans write will continue to be fodder for the LLM's so they'll "learn" that at some point.
The Anti-Turing Test - Prove you aren't AI becoming a thing is very... well Black Mirror.
It's been weird to see the face of a bunch of youtubers I've followed for years without any idea what they look like (mostly history/military history/engineering type stuff), it used to be video/photo's/diagrams with them providing a voice over but because of the sheer quantity of absolute shite/slop mirroring that style they've started appearing on camera as quick "proof of not AI" check.
For that type of content YT is essentially completely ruined if you didn't already follow the good ones because finding them in the absolute sea of dross has become way more difficult.
A friend showed me a video, put out by a local news organization, that started with an attractive woman, standing in the place being discussed, speaking into a microphone.
Cool. Standard local interest story.
Then, it cut to a photo montage, with “her” voice, continuing the story.
Except it was AI. It wasn’t especially successful. You could hear the “tics” from the AI.
Not saying you're wrong on the main message (and those examples are with filters rather than true CRT) but it's still impressive how it can make the pixel art better looking in those cases.
reply