Every article tag can be clicked to get a list of all articles in that category. Every article tag also has an RSS feed! You can customize an RSS feed too!
Latest Comments by kokoko3k
Direct3D to Vulkan translation layer DXVK v2.5 released with rewritten memory management
12 November 2024 at 1:27 pm UTC

I really hope this is related.

Where doitsujin answered:

QuoteThis is not a bug, DXVK does not free once allocated memory blocks by design since doing so would require defragmentation, and is generally not very useful if the game needs more VRAM once again.

The BenQ beCreatus GR10 USB4 Steam Deck Dock is simply great
5 November 2024 at 1:23 pm UTC

Not owning a deck, I'd like someone to explain me why/how "ordinary hubs" slow down the handheld gaming.

Save & Sound is a big live show celebrating music in gaming coming in November
28 October 2024 at 5:22 pm UTC

A good soundtrack is 50% of the fun for me.

Portal 2 project lead wants Valve to return to single-player games
23 October 2024 at 4:24 pm UTC Likes: 2

When I play single player games, I feel like I'm doing a multi player with the game creators.

It is always been like that and I love that emotion sooo much more than playing with random angry people on the net.

Even if they were friends, the demands of family and work would make it a rare occurrence for all of us to have free time at the same moment."


I'm so grateful I'm out of the whole anticheat issue "by design".

Valve makes a big improvement for Native Linux games in a Steam Beta update
18 October 2024 at 12:31 pm UTC

QuoteSteam developers can now select which Steam for Linux runtime to use for native titles. Please consult the partner site for more details.

What exactly is a "Steam developer", one that works for Steam or they intend "Game developers publishing on Steam"?

Valve (Steam) begin a direct collaboration with Arch Linux
28 September 2024 at 5:54 pm UTC

Quoting: const
Quoting: kokoko3kHow anti-cheat relates to that?
To really secure system integrity, there needs to be a full validation chain up to the kernel (and potentially beyond). Without that validation, game devs may continue to distrust anticheat tools on Linux. We don't yet know the new API MS announced to integrate in Windows, but it's really certain Linux will not be able to provide an equivalent unless the kernel and core libraries are build and signed by a trusted entity. Wouldn't make much sense to use that APi if the user can use a patched kernel. As SteamOS uses Archs kernel images and libraries, that must be done in Archs build system, hence the speculation this is related.

To be frank, I think we will see a major shift in cheating and anti-cheat in the coming years, it will be a battle of "AIs".

Yeah, my question came from my early assumption (post confirmed by the updated article) that the signing was related to just packages, your is/remains the choice to what to install and from where.

It is and remains an optional form of protection.

Valve (Steam) begin a direct collaboration with Arch Linux
28 September 2024 at 9:00 am UTC

How anti-cheat relates to that?

NVIDIA driver with Linux kernel 6.10 causing kernel oops
24 August 2024 at 11:45 am UTC

Speculating on the fact that the open and the closed ones are causing the aame issues and given that the userspace part of the driver "should" not crash the kernel by its own, maybe there is common code between the open and the closed drivers.

Humble Games confirmed a 'restructuring of operations' with reports of all staff gone
24 July 2024 at 7:46 pm UTC

May the old staff team up and start from the ground once again.

Popular multiplayer code editor Zed gets a Linux release
14 July 2024 at 4:32 pm UTC Likes: 1

Quoting: wvstolzing
Quoting: kokoko3kWhere does your statement comes from?
I would not be that sure about that.

With all due respect, that's because you seem to be pretty clueless as to how a gpu is better at calculating pixels on a 2d surface.

QuoteMine comes from the fact that if you feel the need of accelerating the drawing of text with a modern or even aging cpu, then your doing something very wrong.

Take a look at the following articles for an overview:
https://sw.kovidgoyal.net/kitty/performance/
https://tomscii.sig7.se/2020/11/How-Zutty-works

Thank you for the efforts, but it's obvious that drawing using the gpu can be faster, if done right.
What is not, is that the implementation has to be right for it to be.

If you follow the discussion, you'll understand that I speak about inefficient coding.

There should be really no need for a text editor to use gpu rendering to draw text, ofc you can do it, but advertising that as a feature smells like the underlying code "needs" it to perform well.

Given that, they should also advertise what gpu this text editor needs to perform as they expect, if there is really something to expect :)