In your case it may very well be CPU dragging GPU down, especially when non K processor meaning no breathing room provided by OC. i5 2500 and 980Ti are hardware genrations apart.
-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
Skylake review thread
- Thread starter LordOfChaos
- Start date
In your case it may very well be CPU dragging GPU down, especially when non K processor meaning no breathing room provided by OC. i5 2500 and 980Ti are hardware genrations apart.
Coulomb_Barrier
Member
If bench site isn't in English language it should be ignored as obscure and untrustworthy as they dare type in different language than your native lang? Oh wow. Language of site shouldn't matter if benches are solid work and how they got to results is laid out well.
Edit: Sorry for double!
No, having to rely on a bad Google translation to try and understand the methodology and conclusion is far from ideal.
Anandtech, BitTech and KitGuru are very reputable, and all came to the same conclusion. It's silly to try and say they got things wrong or question their methods. There's no need to try and scour the net for more positive conclusions on Skylake. It delivers incremental improvements, if any, over Haswell. That's all we need to know.
Prophet Steve
Member
Are you running MSAA? Then it may still be a GPU bottleneck. Otherwise something seems wrong.
The 2500 might just not cut it yes, but resolution should not have a big, if any impact on the CPU performance.
cripterion
Member
Doesn't seem to be a big upgrade over 4790K... whelp, I'll stick to what I have.
Good thing then that e.g. Eurogamer does report in English language and explain their testing methods through which they got their gaming numbers. Those numbers also look very similar to ones reported by few "non English language" sites.
Anand is one of few bench sites that used slowest possible DDR4, 2133. So yeah I'm calling them out on that, especially when there is benches out there showing that Skylake scales well with memory performance. They even mention not running DDR3 in Haswell spec because out of spec DDR3 is so common yet run Skylake in mem spec because..?
Also liking that "scouring the net" jab there, but I didn't need scour anything. Google did that for me :b
OriginofHysteria
Banned
How are the thermals on skylake compared to haswell?
Prophet Steve
Member
Anandtech has already been mentioned, but looking at the KitGuru review they mention there is a significant difference with higher speeds, although with bandwidth. Their game benchmarks are all GPU bottlenecked.
I think I have most things on max. Maybe I should turn off all AA as I have a gsync monitor?
Very appreciable differences between stock 2500k and 6700k with fast RAM. While I will not be upgrading to a 6700k, I don't feel it's the need that the new CPU is not an upgrade. My usage patterns and upgrade costs inhibit my ability to do so at this time.
With VR releasing later this year/early next year, drive for high-frame rates, and CPU intensive games, Haswell and Skylake will possibly have the 2x00k crowd look for the need to upgrade. While my 2500k will live long at 1080p at 60fps, I do want to upgrade to 120hz monitors, where Skylake will make more of an impact in regards to performance.
With VR releasing later this year/early next year, drive for high-frame rates, and CPU intensive games, Haswell and Skylake will possibly have the 2x00k crowd look for the need to upgrade. While my 2500k will live long at 1080p at 60fps, I do want to upgrade to 120hz monitors, where Skylake will make more of an impact in regards to performance.
Prophet Steve
Member
G-Sync doesn't have anything to do with AA. But if MSAA is on, yeah, that requires a lot of performance and only costs more when you go up in resolution.
Coulomb_Barrier
Member
That's the problem. Most games are GPU bottlenecked. KitGuru are benching for the benefit of a real-world scenario (people with good dGPUs), and in these conditions, a 6700K shows zero gains over a 4790K. I mean this is nothing damning, as this is how it's been for the last few CPU gens. And we've got people here trying to prove that with a 6700K paired with a high end dGPU you're going to see 10-25+ fps difference in games over the same system with a 4790K. Think about that for a second. It ain't happening and something is up with that set of rogue benches, clearly.
Quote from http://www.eurogamer.net/articles/digitalfoundry-2015-intel-skylake-core-i7-6700k-review
But our focus is all about gaming performance, and measuring that in a meaningful way that actually brings CPU capabilities to the forefront is not exactly easy. A typical benchmark concentrates on one task and hammers away at it repeatedly, making for easy to track, comparable results. Gameplay stresses the CPU in different ways all the time, different games utilise the processor to varying degrees, and some do not even utilise all of the threads available on an i7. On top of that, the benchmarks included with games generally concentrate on graphics performance. It's for these reasons - and more - that most of the Skylake reviews we've seen so far present gaming results that show little or no difference between any Intel quad. And yet, play the Welcome to the Jungle level in Crysis 3 using Sandy Bridge and then with Skylake and it's immediately obvious that the newer tech provides a tangible, worthwhile boost.
We've attempted to do something about this by adopting two measures - firstly, in order to isolate CPU performance as much as possible, we've paired the processor with an overclocked Titan X running at 1080p resolution. The idea here is that the graphics hardware is so fast it can handle ultra settings or equivalents, making the processor the bottleneck - a scenario that works on most titles, but falls short on others. On top of that, all but one of our benchmark clips come from actual, repeatable gameplay scenarios - a wise move, based on the virtual non-results we get from Shadow of Mordor's in-built benchmark.
We've got a bunch of different comparisons lined up, but we'll kick off with the i7 6700K in stock and overclocked configurations, compared against three prior Intel generations. We strongly recommend watching the videos to get an idea of how CPU performance actually works in practise: where the processor workload comes to the forefront, you'll see the differential. Where it's less of an issue, GPU takes precedence and you'll see performance converge. In part this explains why the bar charts found in many PC reviews don't really cut it when it comes to comparing what the CPU is actually capable of: the differences are averaged out when in areas of the benchmark run where it's actually the graphics card that is the limiting factor. Our approach does throw up some interesting results though, and watching the video certainly puts them into context.
They are benching CPU like CPU benchmark should, benching against dGPU bottleneck doesn't tell you anything about CPU performance and improvements. Maybe I'm weird, but I actually like to see how CPU performs in gaming situations where performance is dictated by CPU. Also some people seem to want it have both ways, claim that Skylake isn't upgrade while saying that 2500K is fine at 1080p when running 980Ti. Think about that for a second.
Coulomb_Barrier
Member
Yeah, out of interest, it serves a purpose. But other than that, their tests are largely pointless. Unless you play those games where performance is dictated purely by the CPU. Oh hold on a minute, they don't exist. Most people have a dGPU, you can't expect AnandTech and the rest to review Skylake in totally unrealistic scenarios where you remove all influence of a dGPU or even its own iGPU in gaming benchmarks and find it does have performance gains, and slap a 'Recommended' tag on the end of it as it blasts older gen when you jump through several hoops to find a performance advantage. Their reviews are not studies on the isolated effect of a Skylake CPU on gaming performance. They are written to inform the reader how viable these new processors are as a purchase, and as an upgrade for your PC.
Prophet Steve
Member
The only thing benchmarks like that are saying is that the CPU is not performing bad enough to get below the GPU bottleneck. I get no information or comparison data from them otherwise. It is moronic.
If I know how much better this CPU performs in games, even if they are on settings where you normally would be GPU bottlenecked, I have a much better idea of how much the CPU is going to last. Or when I want to game on 120 FPS, I'll walk into plenty of bottlenecks. Showing a GPU bottleneck is useless.
This way the information you are presenting is that the CPUs are just as good as a few generations back. You get one and then it turns out a few years later your CPU is bottlenecking which you wouldn't have had if you went for a newer one.
OriginofHysteria
Banned
I have the same but seeing that z170 Asus MB in White and black is seriously making me consider changing as it would go perfectly with my rig.
Yh I really like the colours board manufactures of gone with the chipset. The one thing I want right now is a smaller case.
Yet you argue that benches where they actually test CPU in gaming instead of GPU should be ignored because they don't run CPU benchmark while hitting GPU bottleneck? That is just what I don't seem to understand here, in theory you apparently look for CPU benches yet when it comes to gaming you are ok it being GPU bench? How benching GPU performance has anything to do with CPU?
Also aren't all benchmarks about any PC HW serving purpose out of interest?
Winterblink
Member
Supposedly better, the voltage regulators are moved out of the chip.
I'm unfortunately in the same boat as you. Gonna need to save up a bit more since I just blew my load on an OSW but thankfully all I play these days are emulated games, Resident Evil 5 and Elsword ;p
Unknown Soldier
Banned
Intel is still using the awful TIM between die and IHS. The die is smaller than ever due to 14nm process, the TIM sucks, and well you can guess the rest.
Considering these are supposed to be the overclocking CPUs it's disheartening to realize that you need to delid and void your warranty just to be able to get optimal thermal performance out of the chip. The temperature difference is insane too if you delid and add your own decent thermal paste and slap the cooler on naked or replace the IHS, we're talking 16-20C difference.
I swear to God it would cost Intel 1 dollar more per CPU to put decent TIM on it, I would pay the extra goddamn dollar to get better thermal performance.
Pay more for a chip with the same shitty TIM/gap and no heatsink
New CPU and DDR4 work really well for HandBrake, one of the few things that benefits a lot.
LOL, I still have the first i5 that came out, the 750 I think. Still doing fine, although it's been overclocked by a gig for like 6 years, prolly gonna go out soon
There is more of us i5 750 bros than I thought
Running mine @ 3,8GHz and going strong, but it was still time to move on and get more power under the hood./highfive
IINM they have decent TIM but some SKUs might have had it applied unevenly. If Intel cares enough they'll just solder.