The impact of your CPU on gaming, Intel's 6700K verus 6950X

Subject: Processors | June 27, 2016 - 02:40 PM |
Tagged: dx12, 6700k, Intel, i7-6950X

[H]ard|OCP has been conducting tests using a variety of CPUs to see how well DX12 distributes load between cores as compared to DX11.  Their final article which covers the 6700K and 6950X was done a little differently and so cannot be directly compared to the previously tested CPUs.  That does not lower the value of the testing, scaling is still very obvious and the new tests were designed to highlight more common usage scenarios for gamers.  Read on to see how well, or how poorly, Ashes of the Singularity scales when using DX12.

View Full Size

"This is our fourth and last installment of looking at the new DX12 API and how it works with a game such as Ashes of the Singularity. We have looked at how DX12 is better at distributing workloads across multiple CPU cores than DX11 in AotS when not GPU bound. This time we compare the latest Intel processors in GPU bound workloads."

Here are some more Processor articles from around the web:

Processors

Source: [H]ard|OCP

June 27, 2016 | 04:05 PM - Posted by aparsh335i (not verified)

For those of you that don't want to waste time reading through all this BS - the summary is that an i7-6700k in most cases is close to as good or slightly better than the 6x more expensive i7-6950x. The Fury X is slower than the GTX 1080 by about 9% or something, nothing drastic. Translation...save your money, get an i5 unlocked and put as much into better video cards and better monitor.

June 27, 2016 | 04:12 PM - Posted by Anonymous (not verified)

Thank you jeebush, too late for me tho.

June 28, 2016 | 09:38 AM - Posted by Stefem (not verified)

Except they tested the GTX 980 Ti against the Fury X... there is no GTX 1080 in the analysis, mind show you what you would want to see? ;)

June 27, 2016 | 04:10 PM - Posted by Anonymous (not verified)

Godd-fucking-dammit JERMEY! Why couldnt you at least summarize that stupid fucking article for those of us who rather not venture over to the highly cancerous interwebsite that is HARdOcock or whatever it is fucking called. FUCK MAN. It took me at least 10 minutes to sift through that bullshit.

June 27, 2016 | 07:16 PM - Posted by Anonymous (not verified)

http://i.imgur.com/0cJmFJU.gif

June 27, 2016 | 07:54 PM - Posted by Anonymous (not verified)

god bless

June 28, 2016 | 09:58 AM - Posted by Daniel Masterson (not verified)

Ya but you are here making THIS site cancerous. Hardocp is a great site.

June 27, 2016 | 05:09 PM - Posted by brucek2

How much longer is the tech community going to allow the DX12 / Ashes of Singularity honeymoon to continue?

It was exciting at first. But it's been a few months now, and it's still the only title I hear about for this (and that only in hardware reviewing circles; never in game playing circles.)

At some point, either DX12 scaling benefits show up in mainstream games that we're actually playing, or it might as well not exist at all.

June 27, 2016 | 06:36 PM - Posted by Anonymous (not verified)

I agree that this one game as a benchmark is pointless as it doesn't translate to anything. We've no idea if it's an outlier and terrible, really good, or somewhere in between.

We're going to need several DX12 and Vulcan games to really compare and see how good the hardware and software of these respective items really are.

As it's stands it's little more than a tech demo.

June 27, 2016 | 05:13 PM - Posted by Anonymous (not verified)

Both DX12 and Vulkan should be better at moving more gaming compute off of the CPU and onto the GPU for even less dependency on Intel's overpriced x86 cores! That is if you have a GPU with Asynchronous-Compute fully implemented in the GPU's hardware. So for AMD's GCN based GPUs, even more so with the GCN Polaris Micro-Architectural improvements a quad core i5/i7, or Bristol Ridge(Excavator), CPU should do fine for most gaming once the games/gaming ecosystem/software begins to utilize the full graphics API features in Vulkan/DX12. Those older games may still need 6/8 core i7's and the New Zen 8 core SKUs, with more CPU bound games/gaming engine designs.

I always get quad core i7's and will be looking at the 8 core Zen SKUs for a little more processing power on the CPU side to take more of the OS bloat pressure off of games, but that AMD GCN Asynchronous-Compute fully implemented in the GPU's hardware is going to become very nice once the gaming/gaming engine ecosystems begins to accelerate more of the non graphics gaming compute on the GPU along with the graphics compute. This is where having some DX12/Vulkan explicit multi-adaptor will be best utilized, with more gaming compute done on any multi-GPU systems where the games/gaming engine makers can push more non graphics gaming compute onto the GPUs.

For example say a person can have a system with 3 RX480s, well maybe the game's graphics may not scale well across more than 2 RX480s, but with explicit multi-adaptor the games maker can run graphics on two of the GPUs and run non graphics gaming/physics acceleration on the third GPU. It all depends on how the game is made to scale across multi-GPU configurations using explicit multi-adaptor.

June 27, 2016 | 07:44 PM - Posted by Anonymous (not verified)

Really PcPer? After all the garbage they posted in the last few weeks you are still posting a link to them?

Not clicking to that site,

June 28, 2016 | 03:33 AM - Posted by Anonymous (not verified)

VERUS????

June 28, 2016 | 05:15 AM - Posted by Orthello (not verified)

Certainly has been some garbage over there at H lately , so much so i'll no longer read that site, its full of venom. They often have an agenda and they try to push that onto everything.

Its just not a impartial site anymore which is a pity.

June 28, 2016 | 08:15 AM - Posted by Anonymous (not verified)

AMD got butthurt by HardOCP's criticism which they do to all vendors. That's why AMD decided not to give them review samples anymore. That is how much AMD is scared of these review sites that tell it like it is now that their internet propaganda machine and legions of shills polluting all the forms and comment sections to bash the competition at every opportunity and polish the turds that come out of AMD's ass.

June 28, 2016 | 09:26 AM - Posted by Anonymous (not verified)

The whole review sample process is not fair by design, so both AMD's and Nvidia's review sample processes need to be regulated, and an independent third party contracted with to randomly select the review sites that get the samples!

HardOCP is just another review site that gets ad revenues for the makers of the products that it reviews, so that conflict of interest exists internet wide. The whole internet based review sample as well as internet based reviewer/website objectivity is suspect.

HardOCP is just kowtowing to their revenue stream, it has always been thus!

June 28, 2016 | 10:28 AM - Posted by Stefem (not verified)

That's true but I don't think samples have to be sent to a randomly selected reviewer, if you think it's the same in every sector, ranging from car to video games each maker send samples to reviewer and of course they place ad where interested people could see it (would be very stupid to place a CPU ad on a cooking magazine).
It's a problem about ethics, look at the plethora of improvised reviewer on youtube for example, they don't relay on ad and they don't get samples from makers but still review are not unbiased nor "properly done".
On the other hand there are few blog run by really competent and smart people who made exceptional analysis and some of them work or worked for rival companies!
Sadly this kind of people are very rare but we should at least stop feeding monster of incompetence like wccftech and similar.

June 28, 2016 | 01:29 PM - Posted by Anonymous (not verified)

Say what you will about Wccftech's "journalism" and it's true that they are clickbait by design! But at least Wccftech is running ads for non PC/Laptop/phone/technology related products! So I'm just as concerned about any review websites that have 100% of the display ads for only the kind of products that that website regularly reviews, and no ads for laundry soap, or cars, or vacation packages, etc.

When I visit a website to read the reviews and see only ads from the very same industry, and vary same producers, of the products that are reviewed by that website, in a ratio of 0% of the ads unrelated to the reviewed products to 100% of the ads from the very same types of products that the website reviews, then my trust of that website is also very low.

That review site sample awarding needs to be fully randomized, and under control of a third party contractor that is certifiably independent of the product maker's sample awarding influence! Just like the state lotteries have to use an independent lottery number generating provider to choose the winning numbers, the awarding of review samples needs to be randomized, to remove any producer influence/review bias on the reviewers/review process.

June 28, 2016 | 02:36 PM - Posted by Stefem (not verified)

Well I'm on HardOCP.com now and I see ads from ASUS, Jeep, Seasonic and pchound.com, also I'm sure you know that we get personalized ads based on our "thought interests" so you may get different ad that me.
Sadly wccftech has much more traffic than a respectable site like PCPer or the Tech Report, nowadays ads are targeted "per person" and "per audience" but more traffic means as always more interest by the advertising agency.

I found the lottery comparison a bit forced, while in the lottery case this solution prevent frauds the same method does not prevent biased reviews.

June 28, 2016 | 03:39 PM - Posted by Lance Ripplinger (not verified)

Comparing current gen CPU's might be more useful if it were to span back to the first generation Intel Core chips like my i7-860. I would rather see how much performance difference I would gain with each generation over my "Lynnfield" architecture chip.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd> <blockquote><p><br>
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.