It is currently quite a bit slower, but the goal is to make a codec fast enough for real time communication use.
VP9 is still about 9x slower than x264, but yields the same quality at half the bitrate. You can set VP9 to run a lot faster, but you'll lose some of the bitrate advantages. Still, VP9 is practical for a lot of applications, such as Youtube.
VP9 producing the same quality as x264 at half the bitrate is hard to believe. Do you have a citation? Which `--preset` for x264 are you basing this on?
It's based on --best for VP9 1.4.0 and placebo for x264. Generally improvement tends to be from 30-50%, based on the quality target and content (the lower the bitrate, the greater the improvement). I have objective metrics which test this at http://arewecompressedyet.com/.
Those example images are of really low quality video. Usually people who care about image quality do not care about that. Can you show some ~720p 5mb/s x264 versus 3-4mb/s VP9 samples?
Low resolution samples are the norm, I think it's because low resolution makes it so you can see how the compression algorithm works. If you a high resolution comparison, you would need to zoom in to see the difference anyway. There are, of course, compression artifacts that are readily apparent even in high-resolution (chain link fences, transparent wipes, etc) but I suspect that they are in the minority.
"VP9 encoding (using libvpx) is horrendously slow – like, 50x slower than VP8/x264 encoding. This means that encoding a 3-minute 1080p clip takes several days on a high-end machine. ... libvpx multithreading [encoding] performance is deplorable. It gains virtually nothing."[1]
1. https://blogs.gnome.org/rbultje/2014/02/22/the-worlds-fastes... n.b. x264 comparisons were taken with `--preset veryslow` which understates x264's potential performance by an order of magnitude. From the same link: "it can be fast, and it can beat x264, but it can’t do both at the same time."
This is old. libvpx 1.4.0 is a lot faster now and has multithreading. On my i7-4900MQ laptop, I get about 3fps encoding 1080p content. Still very slow, but 24 minutes for a 3 minute clip, not days.
No, generally encoding is done in software. It's just because VP9 is a much more complex format with many more different coding possibilities to search. It also hasn't been around as long as x264 to be hyper-optimized.
Comparing x264 to hardware-accelerated encoders (QuickSync, NVENC, VCE), the speed/quality/bitrate tradeoff is massively in favour of the cpu-only x264. So i think hardware encoding could help VP9, but it's not a magic bullet (unless your CPU is busy with other work simultaneously).
how exactly would you search for the specific feature I proposed? There's hundreds of image apps out there, I doubt there's one that specifically says it does that in the headline.
How would you want people to find YOUR app? You must have thought of some keywords or phrases that when googled, should bring up your app. Try using those and see what occupies the territory as of now.
Another approach is to discuss your idea with as many people as possible. Maybe you haven't heard of such an app but someone else has. This should be most fruitful if you do this with your target audience. Even if there is such an app out there but your target audience hasn't heard of it, then that's a good enough reason to try building it.
I haven't gone as far as thinking about how I would market / make money out of it... I would just want to have that feature. If it's not implemented yet then I'll do it and then maybe try to market it.
But you're right, that's a good way to think about it.
What I really don't want to do is lose a month implementing something and then realize that there was already something fairly popular out there that solved it.
I would rather not get too attached to my own ideas and tunnel vision implement them and ignore the reality that if there's already a good app for that I shouldn't bother.
You should have this in mind: Facebook was not the first social network, google was not the first search engine, amazon was not the first marketplace. So think of the market whether there is room for some more improvements and/or the market is huge enough to have more than one similar products. How can you solve the shortcomings of the similar product and be different.
Following this point, I think that it probably exist (at least others people had the idea for sure, most probably didn't do anything with it, most likely for the same reason you don't), but if it's not big enough for you to know they exist, you're free to go imo.
Facebook was not the first social network, however if you want to make a facebook clone today you may aswell throw your money right away and save your time.
Echoing the other comments, think of it not as a missed opportunity, but as a challenge to build something better than what already exists. The odds are that anything you've worked on has (probably) already been done.
Of course they would, you have to have pretty big ego problems to hate on him in the first place. Very few people in power will accept that they're not the smartest most capable guy around. Not saying EY is god and can do anything but I trust more in his potential and competence when he actually starts working on a task than in any random claimed profesional on said task.
You can not write all the POVs in HPMOR and still have 'giant ego' problems. He specifically addresses these issues (scene where harry learns to 'lose').
I think most peoples problem with him is that they hate how brazen he is in taking chances, they might think 'who is this arrogant asshole who thinks he DESERVES to talk to JKR' but the thing is he doesn't BELIEVE he deserves anything. He knows the only price for asking is hate by this group of people and the reward is far greater so why NOT ask?
> Not saying EY is god and can do anything but I trust more in his potential and competence when he actually starts working on a task than in any random claimed profesional on said task.
What are some examples of substantial accomplishments that we can look to so as to justify this faith in his ability to get things done, and done well?
I think you're taking my criticism as much more severe than it actually is. I don't doubt there is a lot of optimization that can be done in the realm of (for example) startup investing - I've seen enough of how the business world works - and I expect that EY may indeed be able to make a difference there.
But to have more trust in EY than professionals and subject-matter experts in general seems rather absurd, and is what opens up EY to criticisms of supporting a minor personality cult.
> He specifically addresses these issues (scene where harry learns to 'lose').
In which he is intended to learn delayed gratification, not humility.
> He knows the only price for asking is hate by this group of people and the reward is far greater so why NOT ask?
I'm not saying there is anything wrong with asking. I hope that he does succeed, particularly in contacting JKR, maybe even winning a Hugo awawrd (I enjoy HPMOR but I think that it is overrated). I have no problem with him seeking contacts to become an investor, talk to people about city optimization, etc. I have a lot of sympathy for what he's saying about the attitude he gets, and that is reflected on the rest of his work, because he's famous for writing fanfiction.
But simply reading his writings provides ample evidence that EY, while genuine and sincere in his beliefs, and intelligent, has a rather overly high estimation of himself and his work, and knowing that puts OP's comment into more context.
Biggest poulariser of the idea of existential risks, founder of the Field of Friendly AI research, founder of MIRI, an organisation dedicated to its research, author of a number of published articles on same. Better than most ever do but if that's all it's not enough given his ambitions.
Founder of a bunch of organizations that have done what exactly?
I like his writings, both fiction and not. At one point, I was I guess, kinda of a fan, and I wanted to look up what progress he'd made to his self-assigned goal of Friendly AI, and I couldn't find anything besides a few cute papers.
I was unimpressed. No doubt Eliezer is smart, but contrary to what he seems to think, there are hundreds of thousands of people in the world just as smart, though maybe in different ways. In the scheme of things, he's not that unique. I think Eliezer's ego would be appropriate for someone who had made some progress in those goals. Presently it's a little cringy ..... but I still hope he surprises us.
> Biggest poulariser of the idea of existential risks,
Hardly. Even EY points to science fiction as what inspired him in a lot of ways. Probably the biggest mainstream popularizer of the idea of existential risk these days is the History/Discovery Channel with the nonsense it puts out. Actually, you could probably just go with the movie/tv industry in general.
Even more scientifically, you've had worries about asteroid impacts, supernova radiation, grey goo, etc. longer than EY has been alive, and these ideas were "popular" and in the mainstream consciousness in a way that EY and his ideas are not and probably will never be. EY and MIRI are unknowns outside of a very narrow field.
> founder of the Field of Friendly AI research,
I am not really sure how much to credit him with this, but I suppose it is true that most AI research pre-EY consisted of trying to develop AI with discussions of "friendliness" being more informal.
> founder of MIRI, an organisation dedicated to its research,
An unknown.
> author of a number of published articles on same.
Articles with virtually nonexistent circulation outside MIRI and LessWrong. How many citations of EY's published articles exist outside of those communities? Being self-published is not exactly extraordinary.
Again, I don't have anything against EY. He's just simply not that significant of a figure. Maybe he will be in the future - he certainly thinks MIRI is the only organization worth donating money to because it is the only way to save mankind - but he isn't now. I would not be surprised if most AI people regarded him mostly as a crank. (I don't think that's so, but I think EY's circumstances make him somewhat antithetical to the mainstream scientific community.) To be sure, I haven't founded anything as successful as LessWrong, even, and I certainly haven't convinced anyone to pay me to think and formalize my ideas. By most measures EY is more successful than I am.
Sidenote: Don't google MIRI at work. The Machine Intelligence Research Institute is not the first result.
> you have to have pretty big ego problems to hate on him in the first place.
Why is that? This is equivalent to saying "He is rubber, you are glue, whatever you say bounces off him and sticks to you". His perceived public overconfidence is only harming him. He doesn't have to start self-deprecating, but toning it down a bit would reduce the amount of people claiming he leads a cult or has delusions of grandeur. I don't believe that the fate of the world rests on his or his organization's shoulders, and I doubt that very many LessWrong users do either. So what does he gain by asserting it does?
I am fine with him taking risks and I value his work, but that doesn't mean I have to value his level of confidence.
To get those things you need willpower and the research was about measuring willpower.
If you want different things that's your choice but for most good things in life you need willpower. You can't do what the guy from "Into the wild" did without willpower.
