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

AMD Ryzen 9 3950X Review

anandtech.com

Post reply on HN
Print view View original

151–160 of 290 posts

Re: AMD Ryzen 9 3950X Review

#151

Post by NicoJuicy »

mrandish wrote: post #23

I noticed yesterday's articles about Intel's upcoming unusually large release of security mitigations. Under serious competitive threat for the first time in a while, I'm curious if Intel may have slowed the release of some mitigations to land after this round of comparison benchmarks.

Intel has been cheating their single core performance against AMD for many years, thanks to predictive and insecure algorithms.

Reality is just catching up, AMD is already number one for a long time. Intell's cheating just didn't make it clear.

I've waited for a year to express my opinion, before I could see any evidence.

The time has come that AMD is even beating Intell's fake single core performance, so slowly we will see all of Intell's shit appear. There is no point in hiding it anymore for the consumer/gaming market, if they are not nr 1 in single core. Not if they lose trust of their business clients and profitable server market.

All those gamers that bought Intel for single core performance and OEM's that stick to Intell because of monopolies betted on the wrong horse.

https://twitter.com/damageboy/status/1194751035136450560?s=1...

This feels "nice", although I'm affected to :(

Waiting for decent AMD laptops...

Ps. Long AMD, ever since Spectre and meltdown.

Re: AMD Ryzen 9 3950X Review

#153

Post by 8K832d7tNmiQ »

kilo_bravo_3 wrote: post #64
IanCutress wrote: post #40

Earlier quoted context omitted.

It's worth pointing out that the value for Intel's 10nm is on its high-density libraries, not its high-performance libraries, which are less dense.

Are there any current products one can buy, regardless of expense, using Intel's 10nm? I know there was one product, the Core i3-8121U, a very middling performance chip which saw an extremely limited release. But it was discontinued.

That processor solely exists as PR stunt to satisfy investors. It has a faulty GPU, lower clock speed, higher temp, and should be considered as a lab sample instead. You can take a look at Charlie's article about it[1]

[1]:https://semiaccurate.com/2018/05/29/is-intels-upcoming-10nm-...

Re: AMD Ryzen 9 3950X Review

#154

Post by davidy123 »

bluegreyred wrote: post #69
davidy123 wrote: post #44

Slight meta, and part of a larger 'rant,' but when are we going to get away from reviews that might as well have been printed on glossy pages in PC Magazine in 1986? Anandtech has been at it since the 90s and hasn't changed their format at all. This is a serious decades-long stagnation of the web. The graphs should be dynamic (able to choose what scenarios and components to compare, and able to search within them), a…

the german site computerbase.de does this. they also do crowdsourced benchmarks for various applications and video game titles. I find that their articles are some of the best on the web, though you'll need machine translation to read them. case in point: https://www.computerbase.de/2019-11/amd-ryzen-3950x-test/3/#... you can press the button on the top right of the graph to show/hide other, less relevant entries. cl…

It's funny how the German sites tend to be so much better at this. I don't follow computerbase, but notebookcheck has some of the most thorough and consistent reviews, and they allow cross-device comparison. They cover news now, too, though I consider that a drawback since it's less focused.

There are a very large number of review sites, Anandtech being one of the first, presumably all following the same sustainability formula, but despite all that effort, consistency, thoroughness, built in tools, and building on aggregated output are the exception.

Re: AMD Ryzen 9 3950X Review

#155

Post by bootloop »

fock wrote: post #96

Can anyone comment on the IOMMU-grouping on typical boards? From what I've just googled, the CPU "lanes" seem to support ACS so it could indeed be working to replace my IGP+GPU-for-the-gaming-VN-Haswell system with a dual-GPU-Ryzen one (contrary to what I believed previously). Have been eyeing TR for that reason, but as I don't really need this amount of I/O (and cores), I might be well served by AM4.

I am finally running a Ryzen 1700 + dual GPU system so I could get rid of dual-booting:

CPU: AMD Ryzen 7 1700

Mainboard: GA-AX370-Gaming K5 (Bios: F25)

GPU (Host): Gigabyte GeForce GTX 970

GPU (Guest): Gigabyte GeForce RTX 2060

OS (Host): Arch Linux

OS (Guest): Win 10

As said it appears to me IOMMU grouping depends a lot on the board used. For mine it looks like all the chipset external PCIes are in one group and the CPU PCIes (basically the 16x for GPUs) are in two seperat groups. So I run two GPUs on 8x on each slot now. I was also able to pass through an USB 3.0 on-board controller as it was also in it's own group.

But there is the ACS patch so you are able to pass through single devices which are in a physical IOMMU group (not only cpu lanes). I don't have personal exp with it tho.

Warning: There are issues with the latest Ryzen gen and bios updates and they break vifo.

Re: AMD Ryzen 9 3950X Review

#156

Post by orclev »

zamubafoo wrote: post #41
intarga wrote: post #31

Earlier quoted context omitted.

>While the gaming market is (usually) seeking the highest top single core clock speed with respect to CPUs, it also relies on other expensive hardware. Meanwhile the dev/content creator/workstation market is much better served by these multi-core behemoths. It's not really that cut and dried, plenty of dev workloads are better served by high top single core, while plenty of gaming workloads are better served by multi…

You are absolutely correct. I might be wrong but I've always heard that the general rule of thumb for gaming CPU purchases is get best single core speed if you aren't buying for a specific game. Also, I based my statement on my workloads which involves a lot of virtualization.

The gaming market is in some ways on the trailing edge of technology. While they're often one of the first to embrace new hardware, they're very slow to adopt changes in software or architecture because the engines that drive most of that are few and tend to evolve slowly. Until Unreal Engine, Unity, or one of the other major game engines fully embrace an architecture that allows developers to easily saturate multiple cores, most games are still going to be bottlenecked by single core performance. On the other hand, it should be blatantly obvious to anyone even remotely connected with development that full multi-core utilization is the only way to keep improving performance going forward, so sooner or later the game engine developers are going to need to fully embrace that.

Re: AMD Ryzen 9 3950X Review

#157

Post by mmmrk »

davidy123 wrote: post #133
xondono wrote: post #129

Earlier quoted context omitted.

User contributed is not a good idea. One of the sad realities is that even in the enthusiast market a lot of people don’t know/don’t care to tune and test their system, and there’s a lot of traps that would make the results useless. There’s some of it that you can check and verify, but things like watercooling or chosing the appropriate RAM configuration and timing will have a big impact.

I think this is solve-able through user reputation.

Reputation is not an objective measure.

Re: AMD Ryzen 9 3950X Review

#159

Post by redisman »

overcast wrote: post #102
kevin_thibedeau wrote: post #100

Earlier quoted context omitted.

As long as you want unnecessary lighting and gaping, EMI releasing, windowed cases.

All easily avoidable.

I just build a mid-range gaming PC. CPUs don't have RGB, memory modules don't have RGB, some overpriced motherboards have RGB, HDD/SSD don't have RGB, some overpriced GPUs had them but I was looking for value for money.

Re: AMD Ryzen 9 3950X Review

#160

Post by tempguy9999 »

NicoJuicy wrote: post #151
mrandish wrote: post #23

I noticed yesterday's articles about Intel's upcoming unusually large release of security mitigations. Under serious competitive threat for the first time in a while, I'm curious if Intel may have slowed the release of some mitigations to land after this round of comparison benchmarks.

Intel has been cheating their single core performance against AMD for many years, thanks to predictive and insecure algorithms. Reality is just catching up, AMD is already number one for a long time. Intell's cheating just didn't make it clear. I've waited for a year to express my opinion, before I could see any evidence. The time has come that AMD is even beating Intell's fake single core performance, so slowly we w…

Well, maybe. Probably so, in part, but intel makes large dies rather than AMD's style of multiple smaller dies, which means signals have less far to go (and possibly don't have to be boosted to get between mini-dies).

Perhaps someone who knows about this stuff can comment.

Edit: the mini-dies are chiplets/CCDs. Couldn't remember the name.

Post reply on HN