Intel Skylake / Kaby Lake

Page 583 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Darkhelmutt

Junior Member
Apr 22, 2017
17
2
16
About what was to be expected after comparing the graphics scores. ~4-5% better in Firestrike FPS and less than 1 FPS difference in Timespy.

Have you run Superposition? Lets try it in 1080p Extreme and 1080p High...anything higher and we will probably be GPU limited.

Superposition, little warm today card throttling. I could bump card up a couple of bins and probably maintain 2025. Or could run bench again in morning when cooler.

Any whoo @ 1987 to 2012.
FKFFnyu.png


vLrxEst.png
 

Charlie22911

Senior member
Mar 19, 2005
614
228
116
Superposition, little warm today card throttling. I could bump card up a couple of bins and probably maintain 2025. Or could run bench again in morning when cooler.

Any whoo @ 1987 to 2012.

Is there any real merit to the claims of performance regression vs Broadwell-E? Comparing superposition vs 7700k doesn't seem to be a very good CPU test even at 1080p...

FYI Superposition doesn't use SLI.




EDIT: Clarification
 

DrMrLordX

Lifer
Apr 27, 2000
21,633
10,845
136
They are here to diminish the perception of sky-x ("poor", "worse", "not the best"), not to hype coffee lake, there's a difference.

Honestly. Just view some publicly-available benchmarks of Skylake-X and compare its performance in games to Kabylake and Broadwell-E. Skylake-X doesn't come out all that well in those reviews. It's rational to conclude that 6c/12t Coffeelake-S will be the top-dog gaming CPU, especially given the clockspeeds and turbo bins available for the chip.

Maybe those reviews are too early, or there's a UEFI/firmware problem, or they should have OCed mesh to 3.2 GHz, or . . . whatever. I don't know for sure.

I do know that many people here are still interested in spreading the truth rather than trying to "control the narrative" or whatever it is you think people are doing here. We HAVE had posters like that, and fortunately many of them have become eerily silent as of late.
 
  • Like
Reactions: french toast

TahoeDust

Senior member
Nov 29, 2011
557
404
136
Superposition, little warm today card throttling. I could bump card up a couple of bins and probably maintain 2025. Or could run bench again in morning when cooler.

Any whoo @ 1987 to 2012.
FKFFnyu.png


vLrxEst.png

Is there any real merit to the claims of performance regression vs Broadwell-E? Comparing superposition vs 7700k doesn't seem to be a very good CPU test even at 1080p...

FYI Superposition doesn't use SLI.




EDIT: Clarification

These were at 2012MHz...pretty damn similar. Looks like the 7700k has ~1% advantage...and the 6900k is a wash.

hXMMDD1.png


Nn5qWaK.png



Thanks for posting those @Charlie22911. I was curious about the 6900k.
 
Last edited:
  • Like
Reactions: Charlie22911

coffeeblues

Member
Jun 23, 2017
49
18
36
Honestly. Just view some publicly-available benchmarks of Skylake-X and compare its performance in games to Kabylake and Broadwell-E. Skylake-X doesn't come out all that well in those reviews. It's rational to conclude that 6c/12t Coffeelake-S will be the top-dog gaming CPU, especially given the clockspeeds and turbo bins available for the chip.

Maybe those reviews are too early, or there's a UEFI/firmware problem, or they should have OCed mesh to 3.2 GHz, or . . . whatever. I don't know for sure.

I do know that many people here are still interested in spreading the truth rather than trying to "control the narrative" or whatever it is you think people are doing here. We HAVE had posters like that, and fortunately many of them have become eerily silent as of late.

You seem to have missed the benchmark comparisons that have started recently in this thread. The initial reviews are being used as an argument that Skylake-X is unusable, so you'll understand my skepticism that the reviewers honestly did their best to evaluate the product.

Now you want to spread the truth and you want Coffee Lake to be the best cpu, but learn the difference between assumption and conclusion and don't claim to be rational when mistaking one for the other. We want new products and higher performance, but we don't have information how game engines scale on 8700k vs 7700k (and in some cases it's the drivers that aren't scaling). This thread is not an exception when it comes to flawed logic and arguments.
 

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
  • Like
Reactions: Drazick

DrMrLordX

Lifer
Apr 27, 2000
21,633
10,845
136
The initial reviews are being used as an argument that Skylake-X is unusable

No, they are being used to support the argument that Skylake-X represents a performance regression in games, not that it's "unusable".

so you'll understand my skepticism that the reviewers honestly did their best to evaluate the product.

My understanding is that you're being awfully testy for some reason. Angry that a mainstream Intel product is going to upstage an HEDT Intel product?

Now you want to spread the truth

I always want to spread the truth . . . not just now.

and you want Coffee Lake to be the best cpu

Not really. I just have to accept it as a highly-probable fact, since right now, the best gaming CPU in a lot of games is . . . Kabylake. So take Kabylake, keep clocks the same, give it two more cores and some more L3. BAM! Coffeelake. Which will be the best gaming CPU.

The whole crux of your argument is that you think there's a smear campaign going on in this thread intended to run down Skylake-X. There isn't, or at least there isn't one of which I am aware.

but learn the difference between assumption and conclusion and don't claim to be rational when mistaking one for the other.

You do realize that deductive reasoning actually has meritorious applications, don't you?

We want new products and higher performance, but we don't have information how game engines scale on 8700k vs 7700k (and in some cases it's the drivers that aren't scaling). This thread is not an exception when it comes to flawed logic and arguments.

So you honestly think the 8700k will be SLOWER than the 7700k? Because right now, the 7700k is still the top-dog gaming CPU. Not the 7900X or 7920X or anything else . . . 7700k. You can get a few games faster on Broadwell-E vs Kabylake at stock because of the massive L3, but the 8700k will go a long way towards mitigating that effect.
 

Charlie22911

Senior member
Mar 19, 2005
614
228
116
Took some toying to figure out why my aspect ratio was broken in 16:9 modes... FYI set your desktop first if you have an ultrawide. Game settings set to default first and then changed to above, SLI off.

6900k 4.2GHz, uncore 3.2GHz, and RAM 3.2GHz cas14. These are my best stable settings (HCI, Prime, etc).

DX11:
1080p
ROTR_1080_dx11.png


1440p
ROTR_1440_dx11.png



DX12:
1080p
ROTR_1080_dx12.png


1440p
ROTR_1440_dx12.png


EDIT: Apparently forum resizes photos automatically, I'm stuck in the early 2000s...
 
Last edited:

TahoeDust

Senior member
Nov 29, 2011
557
404
136
Took some toying to figure out why my aspect ratio was broken in 16:9 modes... FYI set your desktop first if you have an ultrawide. Game settings set to default first and then changed to above, SLI off.

6900k 4.2GHz, uncore 3.2GHz, and RAM 3.2GHz cas14. These are my best stable settings (HCI, Prime, etc).
DX11:


DX12:


EDIT: Clickable thumbnails

I just left my aspect ratio stretched...it does not effect performance. The only place I see any even measurable difference is at DX12 1440p, and that is still within the margin of error. It will be interesting to see what the 7700k does.

DX11:
1080p
KvkhW2l.jpg


1440p
ddG4irD.jpg


DX12:
1080p
WQLG2xk.jpg


1440p
77HOdfR.jpg