Live data from Hacker News

Nvidia won, we all lost

blog.sebin-nyshkim.net

591–599 of 599 posts

Re: Nvidia won, we all lost

#591

Earlier quoted context omitted.

Yeah, but it's not worth. Apparently the "gold" list on ProtonDB is games that allegedly work with tweaks . So like, drop in this random DLL and it might fix the game. I'm not gonna spend time on that. Last one I ever tried was https://www.protondb.com/app/813780 with comments like "works perfectly, except multiplayer is completely broken" and the workaround has changed 3 times so far, also it lags no matter what. Ga…

I've been running opensuse+steam and I never had to tweak a dll to get a game running. Albeit that I don't exactly chase the latest AAA, the new releases that I have tried have worked well. Age of empires 2 used to work well, without needing any babying, so I'm not sure why it didn't for you. I will see about spinning it up.

It's the kind of problem where you think it's fixed, until it's not. Someone gave an expert insight last time I brought this up: https://news.ycombinator.com/item?id=44042049

Re: Nvidia won, we all lost

#592
post #129

Earlier quoted context omitted.

AMD isn't even bad at video games, it's just pytorch that doesn't work so well.

Frame per watt they aren't as good. But they are still decent.

Did you read the article? One of the points is that frames are not a good metric anymore for NVIDIA, since they push the idea that gaming NEEDS AI-generated frames and bullies reviewers to include TAA, DLSS and other frame-boosting techs when giving framerates.

Re: Nvidia won, we all lost

#593

Read this in good faith but I don't see how it's supposed to be Nvidia's fault? How could Nvidia realistically stop scalper bots?

They did stop them for their own sales. They offered the "Verified Priority Access" program to buy their cards from them through their store at MSRP and still do.

Re: Nvidia won, we all lost

#594
post #26

Why does the hero image of this website says "Made with GIMP"? I've never seen a web banner saying "Made with Photoshop" or anything similar.

Were you on the internet in the 90s? Lots of banners like that on every site.

I do web development for the last 25 years and yes I was there before .com boom. Gifs maybe were marked but not banners like this.

Re: Nvidia won, we all lost

#595
post #363

Earlier quoted context omitted.

> I use Linux and have learned not to touch AMD GPUs The situation completely changed with the introduction of the AMDGPU drivers integrated into the kernel. This was like 10 years ago. Before then the AMD driver situation on Linux was atrocious. The open source drivers performed so bad you'd get better performance out of Intel integrated graphics than an expensive AMD GPU, and their closed source drivers were so poo…

I use the amdgpu driver and my luck has not been as good as yours. Can't sleep my PC without having it wake up to fill my logs with spam [0] and eventually crash. Then there is the (in)famous AMD reset bug that makes AMD a real headache to use with GPU passthrough. The card can't be properly reset when the VM shuts down so you have to reboot the PC to start the VM a second time. There are workarounds but they only wo…

I just read your comment today. And the good news is, someone just found the commit that triggered the sleep issue literally a couple hours ago. Fingers crossed a fix is incoming.

Re: Nvidia won, we all lost

#596
post #250

Remember when nvidia got caught dropping 2 bits of color information to beat ati in benchmarks? I still can't believe anyone has trusted them since! That is an insane thing to do considering the purpose of the product. For as long as they have competition, I will support those companies instead. If they all fail, I guess I will start one. My spite for them knows no limits

People need to start asking more questions about why the RTX 50 series (Blackwell) has almost no performance uplift over the RTX 40 series (Ada/Hopper), and also conveniently its impossible to find B200s.

Nvidia is quietly having their Intel moment.

The 1080ti can be analogized to Sandy Bridge/The 2600k: Insane performance per dollar, generous binning, and plenty of room left for overclocking. Held up with minimal concessions for a decade, still fine with mild compromises past that.

Every generation since? Gives less and less, all Nvidia can do is all they've ever done: press up against the aspect limit of the dies and lean into their massive scale. What's fun about this go round is that the lines are far blurrier between Consumer & Enterprise, a bunch of 5090s is not hamstrung from doing the things a B200 can in the way a 2600k or i9 is compared to a xeon processor. On top of this, there is an entire cottage industry dedicated to adding additional VRAM to old cards and harvesting GPUs from otherwise broken video cards and swapping them onto new boards.

Re: Nvidia won, we all lost

#597

Earlier quoted context omitted.

I use the amdgpu driver and my luck has not been as good as yours. Can't sleep my PC without having it wake up to fill my logs with spam [0] and eventually crash. Then there is the (in)famous AMD reset bug that makes AMD a real headache to use with GPU passthrough. The card can't be properly reset when the VM shuts down so you have to reboot the PC to start the VM a second time. There are workarounds but they only wo…

I just read your comment today. And the good news is, someone just found the commit that triggered the sleep issue literally a couple hours ago. Fingers crossed a fix is incoming.

Yes, I saw that! No response from the devs yet. Hopefully this gets into the kernel soon as a bugfix.

Re: Nvidia won, we all lost

#598

Earlier quoted context omitted.

I use the amdgpu driver and my luck has not been as good as yours. Can't sleep my PC without having it wake up to fill my logs with spam [0] and eventually crash. Then there is the (in)famous AMD reset bug that makes AMD a real headache to use with GPU passthrough. The card can't be properly reset when the VM shuts down so you have to reboot the PC to start the VM a second time. There are workarounds but they only wo…

I was under the impression that nvidia just didn't let consumer cards do GPU passthrough.

For the longest time they had something in their driver that threw an error and bailed if it detected the GPU was being passed through to a VM but that was easily worked around in qemu or libvirt. nvidia must have realized that that check in the driver was pointless and removed it.

Re: Nvidia won, we all lost

#599

Earlier quoted context omitted.

> This is in no way true and is quite an absurd claim. You provided no evidence to back up this very strong statement; should we just take your word for it? > especially in areas such as 1% lows, stuttering etc. Oh, if you're willing to spend $1k to improve your 1% lows, I guess your argument makes sense.

> You provided no evidence to back up this very strong statement; should we just take your word for it? Where is your evidence? You were the one making grand claims entirely unsupported by reality. Should we just take your word for it? It seems to have been your expectation given you backed it with literally nothing. My evidence would be literally any benchmark in existence and the fact I actually owned the 2080ti an…

> Take all the time you need.

Oooh, got me hard with that ending. Now I really wanna spend my time engaging with you, great job on furthering the discussion and making a positive impact. Cheers on the world you make for yourself - you're the one who gets to experience it.

Post reply on HN