Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AMD takes on Intel with new Ryzen processors for laptops

theverge.com

Post reply on HN
Print view View original

91–100 of 145 posts

Re: AMD takes on Intel with new Ryzen processors for laptops

#91

Post by paulmd »

mrchicity wrote: post #85
paulmd wrote: post #81

Earlier quoted context omitted.

Well, we don't actually know if they did fix it on TR/Epyc. Epyc is on a newer stepping (that has not been released to the consumer market yet), but the issue was only acknowledged a couple months ago. The B2 stepping may already have been taped out at that point. We don't really know because you pretty much can't get ahold of Epyc systems at the moment (at the moment, the only way is by ordering prebuilt servers fro…

Merely anecdotal, but I've run heavy multi-threaded compile tests on my 1950x for days to check whether the chip was stable when overclocked, and have never seen it segfault. That was my main purpose for purchasing the CPU. I use it for that all day every day. If I were spending millions on Epyc servers, I'd certainly test this on a reasonably large sample before committing.

Yeah, given the binning that's in play I'm not particularly worried about TR. Out of an abundance of paranoia I would recommend a 24h run of kill-ryzen.sh on literally any Ryzen/TR processor (including RMA replacements) but TR is not likely to be a problem. For that matter, the chips that are taking 24h+ to segfault are not likely to be a problem either. It's the ones that are segfaulting in a matter of seconds or minutes that are going to be problematic for normal users.

However, that's my thoughts as to why we're not seeing it on Threadripper despite it being on the same stepping. If it were a simple microcode fix (even if you needed to disable some codepath and cut a few percent off) then there is no reason not to pass that down to consumer Ryzen. So far, however, a microcode fix has not been forthcoming. So it's either TR is better binned and thus less subject to the fault, or that something about TR's NUMA layout is breaking one of the necessary conditions for the cache fault to occur.

I do think AMD stepped up their QA after the fault was discovered. I've actually heard of post-week-30 retail samples passing kill-ryzen runs, whereas essentially 100% of the pre-week-25 chips display the fault. However, there are definitely post-week-30 chips that do still display the fault.

I assume what's going on there is AMD is doing a quick run of kill-ryzen to weed out the shittiest chips, the ones that die in seconds/minutes. But since there isn't a way to deterministically reproduce this fault, and AMD can't realistically spend 24h testing every single chip for one fault, some of the only-somewhat-shitty chips are still leaking through their QA testing.

I would take a random week 33 chip over a random week 08 chip for sure, the quality is definitely up, but week 30+ is no guarantee that your chip won't segfault either.

Re: AMD takes on Intel with new Ryzen processors for laptops

#92

Post by moonbug22 »

lower wrote: post #78
Improvotter wrote: post #7

Earlier quoted context omitted.

The AMD drivers have gotten a complete overhaul. The open source drivers are amazing. The Nvidia drivers suck in comparison. Have you seen the post from the Sway developer that was basically "fuck you Nvidia"?

For reference: http://sircmpwn.github.io/2017/10/26/Fuck-you-nvidia.html

That's quite a tantrum.

Re: AMD takes on Intel with new Ryzen processors for laptops

#93

Post by flukus »

sgift wrote: post #35
turblety wrote: post #33

Unfortunately anyone who takes their security or privacy seriously will be totally unable to use this. Every AMD chip contains a likely backdoored system called AMD Secure Processor (formerly “Platform Security Processor” or “PSP”) [1]. This separate processor contains closed source, proprietary binaries that have complete and unrestricted access to the host. Despite a large petition for AMD to opensource this, they…

Looking at [4] tells me that no modern, relevant CPU supports what libreboot needs/requires, so while not a good point it doesn't help deciding between AMD or Intel - and those are, for better or worse, the only relevant CPUs for laptops/desktops.

The intel one can can finally be disabled: https://puri.sm/posts/purism-librem-laptops-completely-disab...

Re: AMD takes on Intel with new Ryzen processors for laptops

#95

Post by gnufied »

kevingadd wrote: post #53

Desktop Ryzen still suffers with thermal/power efficiency issues compared to desktop Intel cores, so I wonder if they've managed to address that issue for these mobile cores. If not, I wonder how customers will feel about paying a battery life tax for better performance... Incidentally, AMD GPUs have been at a disadvantage compared to NVIDIA's in the power management area for a while, as well. Which is unfortunate be…

Comparing power draw on full load on 2 different CPUs is not apple to apple comparison unless they are doing same thing. In benchmarks you have linked Ryzen CPUs are also more powerful.

For example - if I am compiling Linux Kernel and Ryzen takes more power but compilation is faster then my laptop can hit lower power state faster after it is done compiling. And 1500x for example is way faster than all CPUs in linked benchmark.

I do agree about Idle power draw though, but it seems very small difference.

I think in the end, we will have to wait for official benchmarks from reviewers before concluding anything. We can't draw our conclusions from AMD's marketing material.

Re: AMD takes on Intel with new Ryzen processors for laptops

#96

Post by paulmd »

mikenew wrote: post #84
lower wrote: post #78

Earlier quoted context omitted.

For reference: http://sircmpwn.github.io/2017/10/26/Fuck-you-nvidia.html

> Nvidia users are shitty consumers and I don’t even want them in my userbase. Good Lord. I get that the EGLStream vs GBM thing is controversial, but this is childish. How are you supposed to discuss anything with someone who thinks like this.

The CUDA ecosystem is the dominant one within the GPU compute world, and it's only supported by NVIDIA hardware. And up until very recently AMD's Linux drivers were total dogshit. So there is a dominant share of NVIDIA products in this market, and it will continue to be necessary to use NVIDIA products for the foreseeable future. That's just how it is.

To be honest I've never heard of this guy nor his project before, and I'm certainly not interested in trying it after his tantrum. There will be other tiling window managers out there (and I just use a normal window manager anyway). It's NVIDIA, they have a >80% share of the discrete GPU market. Someone will fill the gap. /shrugs

Hint: Linus Torvalds' tantrums are not a positive character trait worthy of emulation. Maybe they are a necessary evil when herding a dozen teams of a dozen dozen cats each, but I'm guessing that's not the situation with this guy. It looks a lot like a fanboy ranting.

Re: AMD takes on Intel with new Ryzen processors for laptops

#97

Post by roblabla »

paulmd wrote: post #96
mikenew wrote: post #84

Earlier quoted context omitted.

> Nvidia users are shitty consumers and I don’t even want them in my userbase. Good Lord. I get that the EGLStream vs GBM thing is controversial, but this is childish. How are you supposed to discuss anything with someone who thinks like this.

The CUDA ecosystem is the dominant one within the GPU compute world, and it's only supported by NVIDIA hardware. And up until very recently AMD's Linux drivers were total dogshit. So there is a dominant share of NVIDIA products in this market, and it will continue to be necessary to use NVIDIA products for the foreseeable future. That's just how it is. To be honest I've never heard of this guy nor his project before,…

This tantrum is fairly warrented TBH. Nvidia has been hostile to the linux world for as long as they existed. That they've been in a monopoly situation means that the community had no choice but to take the pain in.

Sway is a WM, and wlroots is a new library that should help in creating Wayland WMs (similar to wlc). SirCmpwn has to maintain all of that, and it turns out it's a fairly big task. Realistically, his WM will run on hardware running Intel for the most part, with AMD or Nvidia way behind. Optimizing for Intel is fairly simple, and it turns out it gives AMD support for free, since they use the agreed-upon APIs. His choice makes sense.

The thing is, people are going to always ask why that choice have been made. And I can understand this being super-frustrating, hence the post being so harsh on nvidia. But really, nvidia deserves it. Wayland wasn't designed in a vacuum, and nvidia took the wait and see approach. Once every bit were put in place, Nvidia came up with the EGLStream proposal. Not during discussion. After everything was said and done. Adding a codepath for EGLStream is really non-trivial, and I hope nobody does it because it really adds additional complexity.

Really, that AMD is getting more competitive is going to help the GPU world a lot. Nvidia has always been a bad citizen there, and oh boy if AMD ever gets in a position where they can compete on compute, that'd be great.

Re: AMD takes on Intel with new Ryzen processors for laptops

#98

Post by microcolonel »

paulmd wrote: post #96
mikenew wrote: post #84

Earlier quoted context omitted.

> Nvidia users are shitty consumers and I don’t even want them in my userbase. Good Lord. I get that the EGLStream vs GBM thing is controversial, but this is childish. How are you supposed to discuss anything with someone who thinks like this.

The CUDA ecosystem is the dominant one within the GPU compute world, and it's only supported by NVIDIA hardware. And up until very recently AMD's Linux drivers were total dogshit. So there is a dominant share of NVIDIA products in this market, and it will continue to be necessary to use NVIDIA products for the foreseeable future. That's just how it is. To be honest I've never heard of this guy nor his project before,…

> It looks a lot like a fanboy ranting.

Well it's not. He's the one who has to do the work, and if I were in his position, I wouldn't do it either. Like if you ordered a coca cola machine compatible syrup pack, and instead received a pepsi co. one, and the manufacturer tells the court that they fulfilled your order even though you are left to accommodate something other than what they claimed to provide.

What NVIDIA has done is claim to support something, while in reality supporting something entirely different and ancillary which everyone else has to do work to accommodate. SirCmpwn isn't "fanboy ranting", he is talking about how ridiculous it is that NVIDIA has pushed the onus for supporting something which is effectively their own proprietary API on the rest of the ecosystem, for apparently no technical reason, and with either no consideration, or malice of intent.

Re: AMD takes on Intel with new Ryzen processors for laptops

#99

Post by lightedman »

c2h5oh wrote: post #86
shmerl wrote: post #45

This should make life easier for Linux users who want something better than integrated Intel GPU. No more Optimus horrors and lack of proper PRIME support.

Not yet - there is no display support for that GPU in current Linux kernel, nor in 4.14 that is going to be released in 1-2 weeks. Users will have to wait for 4.15 - until mid-late January.

You need a new kernel to support a GPU? What ass-backwards garbage is this? With Windows, it's install a new driver; not update the core of the OS and possibly introduce horrible bugs.

Re: AMD takes on Intel with new Ryzen processors for laptops

#100

Post by Const-me »

kevingadd wrote: post #53

Desktop Ryzen still suffers with thermal/power efficiency issues compared to desktop Intel cores, so I wonder if they've managed to address that issue for these mobile cores. If not, I wonder how customers will feel about paying a battery life tax for better performance... Incidentally, AMD GPUs have been at a disadvantage compared to NVIDIA's in the power management area for a while, as well. Which is unfortunate be…

> AMD GPUs have been at a disadvantage compared to NVIDIA's in the power management area for a while

Correct. And here's why: https://www.realworldtech.com/tile-based-rasterization-nvidi...

Post reply on HN