• Guest, The rules for the P & N subforum have been updated to prohibit "ad hominem" or personal attacks against other posters. See the full details in the post "Politics and News Rules & Guidelines."

Intel Skylake / Kaby Lake

Page 603 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

SAAA

Senior member
May 14, 2014
510
106
116
Yes and it still baffles me how they can state 11% single thread gain for 8700K vs 7700K with a clock speed advantage and then 17-16% for i3s with a speed deficit. The i5s too are weird when you normalize clock/clock. On average the highest end CPU has the smallest gain...

Moving onto the why that gain happens, does L3 affect any benchmark that much? Also with different results because i3s with 50% more L3 gain much more than i7s with 50% too. Then you have some early leaks from other sites... I'm pretty sure there will be some kind of IPC increase, maybe a token 5%, that's still needed after 2 years of skylake rehashes.
 
Aug 11, 2008
10,451
641
126
Yes and it still baffles me how they can state 11% single thread gain for 8700K vs 7700K with a clock speed advantage and then 17-16% for i3s with a speed deficit. The i5s too are weird when you normalize clock/clock. On average the highest end CPU has the smallest gain...

Moving onto the why that gain happens, does L3 affect any benchmark that much? Also with different results because i3s with 50% more L3 gain much more than i7s with 50% too. Then you have some early leaks from other sites... I'm pretty sure there will be some kind of IPC increase, maybe a token 5%, that's still needed after 2 years of skylake rehashes.
Well,
Yes and it still baffles me how they can state 11% single thread gain for 8700K vs 7700K with a clock speed advantage and then 17-16% for i3s with a speed deficit. The i5s too are weird when you normalize clock/clock. On average the highest end CPU has the smallest gain...

Moving onto the why that gain happens, does L3 affect any benchmark that much? Also with different results because i3s with 50% more L3 gain much more than i7s with 50% too. Then you have some early leaks from other sites... I'm pretty sure there will be some kind of IPC increase, maybe a token 5%, that's still needed after 2 years of skylake rehashes.
The CL i3 does not have hyperthreading, and maybe hyperthreading has negative scaling on ST performance in this particular benchmark? Or maybe it is related to the cache. They both have a 50% increase, but maybe the improvement is more significant at smaller levels. Overall though, obviously the line-up is much improved at every point for MT performance. The improvement in ST is surprising in some cases though. Obviously, we have to take these benchmarks with a large grain of salt until we see a much wider range of benches from independent sources.
 

pantsaregood

Senior member
Feb 13, 2011
993
37
91
Came to say 11% increase on single-threaded performance on the i7-8700K compared to the i7-7700K isn't going to happen, but it actually looks feasible if we assume linear scaling across the board.

The i7-8700K will turbo 4.44% higher than the i7-7700K, and its IMC can officially run RAM at an 11.08% higher clock speed. If we assume RAM bottlenecks performance enough to scale linearly, then theoretically single-thread performance on the i7-8700K could outperform the i7-7700K by around 16%. The increased L3 cache should affect performance insignificantly but measurably, given that the additional 2 MB L3 on the i7-7700K does little to benefit single-thread performance over the i5-7600K.

So, that leaves us with an idealized case of ~16% gain in single-thread performance in bandwidth intensive loads, but only a 4.44% increase in performance when not bandwidth-limited. None of this requires an IPC increase in any capacity, so unless the i7-8700K can overclock further than the i7-7700K, they're going to cap out at around the same single-thread performance.
 

LTC8K6

Lifer
Mar 10, 2004
28,520
1,573
126
The i5 and i3 do not have a memory speed advantage though, and they are still shown with large gains.
 

JoeRambo

Golden Member
Jun 13, 2013
1,144
1,007
136
Then perhaps you would like to explain why there's a Core i7-8700K running with 4.7 GHz Turbo inside a Dell PC according to SiSoftware.
12MB L3 is probably bogus as well if we are to believe some earlier claims from some guys in this thread :) 4.7Ghz ST turbo and 12MB of L3 driven by 4.4Ghz uncore combined with fast memory are going to surprise some guys that's for sure.
 

eddman

Senior member
Dec 28, 2010
239
87
101
This is getting confusing.

S, SP (65-95W): Coffee Lake
S, LP (35W): ? (CFL?)
H, SP (45W): ? (I suppose intel would go for 6 cores here, so CFL?)
U, SP (28W): ? (Perhaps quad-core CFL's better transistor performance would make it more suitable here?)
U, LP (15W): Kaby Lake Refresh (but there were rumors of U CNL parts too. Maybe those will be dual-core?)
Y (4.5W): Cannon Lake
 
Last edited:

Qwertilot

Golden Member
Nov 28, 2013
1,586
243
106
Very confusing :)

The S, LP is presumably going to be called T?! Interesting to see if they do stick with 6 cores in that power bracket or not. Maybe if they do they'll push it back 'up' to 45w where iirc it was a generation or two back?
 

nvgpu

Senior member
Sep 12, 2014
629
201
81
https://www.intel.com/content/dam/www/public/us/en/documents/product-briefs/dual-band-wireless-ac-9560-brief.pdf

The Intel Wireless-AC 9560 adapter is a CRF (companion RF module) supporting
the 1st generation integrated Intel wireless 802.11ac solution comprised of CNVi
and a CRF. The solution provides Bluetooth 5 and 2x2 802.11ac Wi-Fi including
wave 2 features such as 160MHz channels delivering up to 1.73Gbps
and downlink MU-MIMO.

CRF: Companion RF module in M.2 form factor supporting integrated solution.
CNVi; Refers to the integrated wireless IP portion residing in the SOC/PCH.
 
Last edited:
  • Like
Reactions: Ajay

TahoeDust

Senior member
Nov 29, 2011
557
404
136
Last edited:

ASK THE COMMUNITY