Revision demoparty banned on Twitch mid-stream
twitter.com
twitter.com
Smallest demo that compressed to the largest video file
But then it might determine that the best compression would be no transmission at all, and decide to eradicate any potential viewers. Those damn pesky almighty AIs.
If you didn't know they were entirely rendered in realtime by just 4K of code, you might think you were just watching a series of beautiful, well-produced music videos.
(A few of the 256-byte ones afterwards get pretty hair-raising too.)
RTX accelerates ray intersection on a typical scene from a video game. But demos are usually not typical scenes, instead using mathematical models like signed distance fields that work differently.
There are other issues: you need compatible hardware, it is not well supported by OpenGL if at all, there may be too much overhead for 64k or less intros.
Personally I don't see how to take advantage of it in a way that opens new options, but maybe someone else will do. After all, it is something the scene is good at.
For small intros it's almost as you say -- RTX is mostly uninsteresting and not that helpful for current status quo at least on the surface. However, while SDFs are do indeed provide a relatively cheap and intuitive way to specify your scene spatially to allow ray tracing and GI approximation with varying degrees of fakiness, they confine you to a rather limited abstract geometry stlye. Anything beyond yet another repeated-rotated-cube-with-cutouts-fractaled-to-death is rather uncomfortable to pull off and is also usually very slow (see iq's incredibly realistic works). That's why most of the intros from last 10+ years look rather similar.
Also, I've been recently playing with Vulkan ray tracing extensions and turns out it's not that hard to pull off. There's a considerable upfront cost (Vulkan being verbose as it is), but it's totally manageable, around maybe 10-20k compressed. The entire PoC intro with music was about 60k, and I haven't even started optimizing it for size. It had also been compressed with "just" UPX (apparently ray tracing extensions work only in 64-bit mode, so more sophisticated compressors are out of the question). I've heard that recent work on native 64bit compressor from Ferris removed another 16k from it, so there's definitely a lot of room for actual content in 64k+rtx.
I was playing around with mesh-gen for 4k back in 2010 and you'd eat up most of your budget before even rendering, this is why SDF based raymarching reigns supreme for 4k intros (and are quite popular for larger categories as well).
And I'd be kinda surprised if you couldn't do something interesting with RT cores even in size-limited formats (possibly more in 64k than 4k though), even if maybe not quite the way nvidia intended.
You can sensibly use RT APIs together with raymarching. There is a notion of custom geometry that's specified using bounding boxes. When ray hits one, it's your responsibility to compute an exact intersection manually in the shader. This can definitely help with complex scenes with a lot of very different sdf objects that don't intersect much.
But yeah, in that case you'd still have to pay the Vulkan price first. I might experiment with what can be the size-chepest way to get vk rendering.
YouTube would be impossible to disrupt, but Twitch is ephemeral. So maybe if enough people got outraged?
This actually leads me to believe that Twitch users are probably at like a 95%+ AdBlock install rate.
And also the main point was that Twitch does a terrible job at monetizing off their ads. They have the worst advertising platform.
Ads on Twitch run for all streamers. There's very little targeting capability and they strike individual deals for all ads on their system rather than purchase from a DSP.
They're still struggling to make a profit.
Turns out their business probably needs both to work.
https://www.ben-evans.com/benedictevans/2021/3/14/do-amazon-...
SO what will disrupt Twitch is the same thing as usual: a combo of a greedy company, a user experience that gets worse and worse, and a new tech coming out.
Interestingly Youtube has 2 out of 3 - greedy company running ads every 3 minutes, and constant full screen badgering about cookies and sign-ups. I wonder what the new tech will be that will complete the set?
(For the avoidance of doubt, I think PeerTube is great technology and I hope it wins, I just don't think it has any chance of supplanting Youtube any time soon).
Much better than the current situation with YouTube's ContentID where the presumption of innocence is routinely violated, not to mention all the other issues : https://www.youtube.com/watch?v=trIn90HTqKs
(Also, the way the current ContentID might be violating European laws deriving from Article 13, but I haven't really been following that lately...)
Anything that is owned by Google won't improve but rather decline:
- search? Was at its best in 2007
- Reader? Killed after spending all the air in the rss market.
- Google+? Killed after they'd suffocated a few interesting alternatives.
- etc
(notable exception: after a decade they've actually stopped serving me insulting ads - but I do wonder if that wasn't Google but rather the advertiser who got whiff that Google sent their expensive "targeted" ads to everyone.)
- reach out to 20 biggest streamers on twitch and get them to agree to move to your platform, the same way spotify got Joe Rogan to move from YT to Spotify
- with signed agreements ready get funding
- build the platform, and all accompanying streamer tools, preferably fix all major issues that twitch has
- set the date for launch
- launch, with all your onboarded twitch streamers streaming “we moved to platform x” on their abandoned twitch channels
- profit
Note: this requires a few million USD, but it is definitely doable. In fact, YC could get a few investors and build this in a year, if they wanted to.
Anegdata, I just stopped listening to Joe Rogan, despite liking Spotify more than Youtube. I was on the fence anyway, I'm lazy and so on.
It seems to me that we're at the point where it takes the next generation of kids for a new service to get into a crowded space. Because popular people on the platform (it's always about these hubs, that you got right I think) just become older and less relatable.
1. https://checkpointxp.com/2020/12/17/highest-paid-twitch-stre... take into account that somebody likely paid for most pixels on the screen before they even start streaming
Twitch provides absolutely no discoverability to streamers. It's purely a function of big gets bigger. They have zero design or plans to help streamers connect with the viewers who are interested in them (it's not "Playing Game X"). The primary way that Twitch streamers at 100+ viewers get new ones is via Youtube Clips + Twitter!
The first company to reasonably solve this problem will capture the lower 80% of Twitch's streamers.
Weirdly, YouTube has discoverability of normal videos nailed, but not for gaming streams either. There is tremendous opportunity here.
How in the holy hell did the US get that way? It's in no way Twitch's fault, I just wonder how an entire country went that way...
Influence of prudish christian sects that preached that people should look modest and cover themselves and that any display of eroticism is a temptation to sin. Add hundreds of years of cultural "refinement" and you end up with a ban on casual nipples.
> Why the gender difference?
Well classically if a prudish women would find a man working and sweating topless and a bunch of young women enjoying the show giggling, both would be shouted at for indecent behavior and the man would be asked to cover up. It is that "cultural refinement" that created a gap by allowing the sexual objectification of men in cola commercials, while banning female nipples. Most such taboos have fallen in the 60s, but the prudes tried to make a stand on banning the eroticised display of naked female breasts and the current deal is an idiotic compromise.
Thankfully American owned companies are doing their best to keep evils such as nipples and general nudity off the screen.