Hacker Newsnew | past | comments | ask | show | jobs | submit | sylware's commentslogin

Isn't AV1 on the level of H.265? And are H.265 and the future H.266(will face the upcoming av2) free of charge forever and wherever like av[12]?

They could do the Big Tech way: make it all 'free' for a good while, estinguish/calm down any serious competition, then make them not 'free' anymore.

In the end, you cannot trust them.


Absolutely not.

I wish everyone knew the difference between patents and copyright.

You can download an open source HEVC codec, and use it for all they care according to their copyright. But! You also owe MPEG-LA 0.2 USD if you want to use it, not to mention an undisclosed sum to actors like HEVC Advance and all the other patent owners I don't remember, because they have their own terms, and it's not their problem that you compiled an open source implementation.


VP9 is more on the level of H265 really. VVC/H266 is closer to AV1. It's not an exact comparison but it is close. The licensing is just awful for VVC similar to HEVC and now that AV1 has proved itself everyone is pivoting away from VVC/h266 especially on the consumer side. Pretty much all VVC adoption is entirely internal (studios, set top boxes, etc) and it is not used by any major consumer streaming service afaik.

I guess most people forgot about x264 dark shikari's post already.

VP9 isn't H.265 level. That is the marketing spin of AOM. And even AOM members admit VVC is better than AV1.

Liking one codec or whether it is royalty free is one thing, whether it is performing better is another thing.


I'm well aware of dark shikari. While there are obvious differences technologically and subjectively, on a generational level vp9 and HEVC are both positioned as h264 successors. We all know h264 is brilliant and flexible and very capable. Many companies were looking for an alternative with better licensing; on2's vp8 was Google's first push, but it still lagged behind h264, even though it used many of the same concepts and limitations. Vp9 and HEVC were certainly the next generation and competed with each other directly, really. Among the larger consumer video services, many keep h264 for compatibility, and reserve higher quality or resolutions or frame rates for the newer codecs. tiktok eventually settled on HEVC as its preferred codec, while Instagram used HEVC for a short while before migrating entirely to vp9.

Other developers ran into a ton of issues with licensing HEVC for their own software which is still a complete pain.

Anyway, people are now looking at what's next. VVC came out quite a while ago, and AV1 more recently, but when people are looking for the current sota codec with at least some good support, they end up choosing between the two, realistically. And yeah, VVC has advantages over AV1 and they are very different technically. But the market has pretty loudly spoken that VVC is a hassle no one wants to mess with, and AV1 is quickly becoming the ubiquitous codec with the closest rival VVC offering little to offset the licensing troubles (and lack of hardware support at this point as well)

Anyway, just saying. VVC is a huge pain. HEVC still is a huge pain, and though I prefer it to vp9 and it has much better quality and capabilities, the licensing issue makes it troublesome in so many ways. But the choice almost always comes down to vp9 or HEVC, then AV1 or VVC. Though at this point it might as well be, in order, h264, vp9, HEVC, AV1, and no one really cares about VVC.


There are 2 webs.

web apps and web sites.

web apps require a [java|ecma]script whatng cartel web engine, more and more only the gogol one (blink) will "correctly" work (abuse of dominant position).

web sites are noscript/basic (x)html ("forms" and the <audio> <video> elements). Usually a "semantic" 2D table with proper ids for navigation.


... from the point where you have nothing to _remove_.

one fixed... one bazillion to go...

It should be RISC-V...

What would be the benefit?

See the FAQ on risc-v site.

Remnant of RISC attempt without a zero register.

Come on... that was a joke... this karma system...

codeberg is supposely noscript/basic (x)html browser friendly, IPv6 too, I guess. microsoft github broke slowly and surely all interop with classic browsers (now you must have a "whatng" cartel web engine to even post an issue).

(I was told that codeberg may have dropped noscript/basic (x)html interop, which would make it no more interesting than microsoft github or whatng gitlab)

codeberg people have to be careful and acknowledge the following: expect shadowpaid hackers to ruin it because you are stepping on big tech toes. 99% of the time you will spend on codeberg will have to be to protect it and to keep it available, 1% (if not less) will be forge coding.


I build my own distro, I build myself xorg (but I am writting my own wayland compositor in order to move away from x11), I was the witness of all their tantrums breaking things all over xorg.

I did appreciate a lot the revert of all their code.


kernel anti-cheat are notoriously inefficient and are weaponized by hackers.

Are they? Cheats for games like Fortnite, CS (Faceit), Rust, LoL have become very expensive (100 USD per month are not unheard of) or require you to purchase special hardware.

And I have yet to come across an anti cheat driver of the big publishers (EAC, Faceit, Javelin, Vanguard) being exploited and allow access to r/w kernel memory. It is more likely that the driver of some hardware is being exploited for, rather than anti cheat drivers.

Personally, I only remember the ac driver of Capcom ever being exploited. Compare this to the dozen hardware/av drivers which were exploitable, like the Intel LAN utility driver, ASUS IOMap64, MSI NTIOLIB or that one Razer driver. Oh, and CPU-Z and the Avast Hypervisor driver were exploitable too and allowed r/w on kernel memory. These drivers are way more likely to be weaponized than ac drivers.


I remember the anti-cheat of Valorant being exploited where basically the hackers could turn a bug in the game code into full kernel-level root access via the anti-cheat.

Video games are not engineered to withstand sophisticated hacking attacks which is fine mostly since the bad guys can only use their access to cheat - in this case they could fully compromise your system thanks to the kernel access of anti cheat


The thousands of RGB drivers from the various manufacturers that are just copy+paste jobs on RWEverything is actually disgusting and Microsoft letting that just happen is a serious problem. Ah yes you added AES to your IOCTL very secure! I'd say the only reason that these drivers haven't been exploited is because of the insane bug bounties in place. There are also other big issues in games, see the whole hack with Apex Legends lmao

In still doesn't prevent a wide array of cheating.

It is more a anti-(non-steamOS)-linux than anything else.

Open Source is a grab bag of everything and anything.

At best very lean open source software written in plain and simple C99+/assembly (including the SDK)... maybe worth saving.

You can see more and more open source devs taking (or wishing for) microsoft money with rust tough (we all know who is being the rust agenda, like openai, "open" always makes me laugh.).


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: