My i7 7700k Has Arrived - Insights, Benchies & Overclocks Inside (Now With Delid!)

Page 11 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

LTC8K6

Lifer
Mar 10, 2004
28,520
1,575
126
Just because it was also on 14nm didn't mean it had to be the exact same core.
It was always going to be the same core, though. This was noted in the intro announcements.

Intel hasn’t gone into much detail regarding the new 14nm+ process itself in terms of specifics, but has listed a number of performance gains that come out of the new CPU. The fundamental microarchitecture between Skylake and the new Kaby Lake parts is practically unchanged (DMI 3.0 now allows PCIe 3.0 x4 NVMe drives from the integrated PCH), but the updated fin profile and reduced ‘strain’ by the larger fin pitch is being quoted as giving a 12% performance increase due to process alone, typically through additional frequency for the same power. The main benefits to KBL will be in that frequency due to the 14nm+ process as well as the new media capabilities.

Practically unchanged.

Same core with higher clocks. New media capability.

Exactly what we got.
 

DerpEinstein

Junior Member
Dec 24, 2016
21
6
6
Kaby lake has the same architecture as skylake. Kaby lake just has a 14nm+ refresh and a encoder for video.
I am aware of that, and what about coffee lake? Is there any room at all to improve skylake further?! I kind of doubt it.

That means intel will have 3 years without any improvement.

The door is wide open for AMD to catch up.
 

jpiniero

Lifer
Oct 1, 2010
14,510
5,159
136
I am aware of that, and what about coffee lake? Is there any room at all to improve skylake further?! I kind of doubt it.

No IPC improvement, not even any GPU improvement. Just the additional cores and maybe the return of the FIVR.

We'll have to see about Skylake-X.
 

DerpEinstein

Junior Member
Dec 24, 2016
21
6
6
No IPC improvement, not even any GPU improvement. Just the additional cores and maybe the return of the FIVR.

We'll have to see about Skylake-X.
Skylake-x is basically kabylake-x, which should be a good improvement over broadwell-e, but the boards are quite expensive.
 

RichUK

Lifer
Feb 14, 2005
10,320
672
126
Looks like your IMC is having trouble with memory overclocking since it was working with the old CPU.

I had the same memory overclocks / limitations on the old CPU, and considering the memory is rated at 3000, I have to assume it is the cheap board holding me back.
 
  • Like
Reactions: Drazick

BonzaiDuck

Lifer
Jun 30, 2004
15,699
1,448
126
It's too early to draw conclusions, but my initial observations are:

  • The 7700k has the potential to be stable at 5Ghz @ ~1.4v, with exceptional samples reaching 5.1 / 5.2 perhaps.
  • The power draw when overclocked 'seems' high, evidenced by the vdroop of 0.08v when under IBT load! I have no other means of measuring power draw - Let's hope the AT review focuses in on this.
  • I am unclear at this point whether temps are being reported correctly, but i believe not given the general expectations is for higher temps under poor air cooling (plus I witnessed a discrepancy of ~15 degrees between the BIOS and windows) .
  • Could an AIO cooler be standard in order to run overclocks north of 4.8ghz?
  • If you're going to overclock, get a board that has strong power management to sustain voltage with limited vdroop! I'm not sure whether Z270 motherboards will offer better VRMS etc in anticipation for the 7700k.
This is all very interesting to me. First -- can you link me to a reseller for the "de-lidding tool?"

I ordered my 6700K binned @ 4.8Ghz, $50 for the CLU-re-lidding. Silicon Lottery had pronounced, and I had verified, that it cut 12C off the temperatures for my particular chip.

Instead of an AiO or custom-AiO as I had been considering, I picked a ThermalRight LG Macho for being bested by an EKWB Predator 240 by 5C degrees. So with the CLU-relid, my 6700K performs better than the Predator by 7C.

It would be tempting to see what Silicon Lottery offers in an i7-7700K binned at 5 Ghz (or more? Perhaps?). Or I could face the real lottery and delid it myself. I'm just guessing that the VRMs on my Saber Z170 S board are as good as they seem to be now -- for what is basically a 12-phase or 8+4 power design. And as I understand it, the +4 is for the iGPU, which I don't use.

Someone else mentioned this "wasted silicon" of the iGPU. For a quad core chip, it's a peripheral advantage prior to ordering a dGPU, and of possible unknown advantage thereafter. Think of it as the spare tire for a Duesenberg roadster.

With the Sabertooth, some reviews suggest I lose 100Mhz in OC potential for second-tier choice over the 1st-tier 16-phase models like the "Maximi" models, the Deluxe, maybe one other ASUS board. But the benchmarks put the Saber among those top-four ASUS boards. The two "VCORE" VRMs never exceed 44C under severe stress, and never break out of the 30's with normal gaming usage.

But do I really want to spend that money just to see what 5Ghz is like "on air?" I can't see falling short in comparison to your AIO cooler, unless 5C is going to make a lot of difference. Whether or not it does, I wouldn't want to run my system with peak package C exceeding 75 or 76C. For now, I'll clutch the money and stick with the Skylake.
 

RichUK

Lifer
Feb 14, 2005
10,320
672
126
This is all very interesting to me. First -- can you link me to a reseller for the "de-lidding tool?"

I ordered my 6700K binned @ 4.8Ghz, $50 for the CLU-re-lidding. Silicon Lottery had pronounced, and I had verified, that it cut 12C off the temperatures for my particular chip.

Instead of an AiO or custom-AiO as I had been considering, I picked a ThermalRight LG Macho for being bested by an EKWB Predator 240 by 5C degrees. So with the CLU-relid, my 6700K performs better than the Predator by 7C.

It would be tempting to see what Silicon Lottery offers in an i7-7700K binned at 5 Ghz (or more? Perhaps?). Or I could face the real lottery and delid it myself. I'm just guessing that the VRMs on my Saber Z170 S board are as good as they seem to be now -- for what is basically a 12-phase or 8+4 power design. And as I understand it, the +4 is for the iGPU, which I don't use.

Someone else mentioned this "wasted silicon" of the iGPU. For a quad core chip, it's a peripheral advantage prior to ordering a dGPU, and of possible unknown advantage thereafter. Think of it as the spare tire for a Duesenberg roadster.

With the Sabertooth, some reviews suggest I lose 100Mhz in OC potential for second-tier choice over the 1st-tier 16-phase models like the "Maximi" models, the Deluxe, maybe one other ASUS board. But the benchmarks put the Saber among those top-four ASUS boards. The two "VCORE" VRMs never exceed 44C under severe stress, and never break out of the 30's with normal gaming usage.

But do I really want to spend that money just to see what 5Ghz is like "on air?" I can't see falling short in comparison to your AIO cooler, unless 5C is going to make a lot of difference. Whether or not it does, I wouldn't want to run my system with peak package C exceeding 75 or 76C. For now, I'll clutch the money and stick with the Skylake.

http://rockitcool.myshopify.com/
 
  • Like
Reactions: Drazick and Burpo

Dave2150

Senior member
Jan 20, 2015
639
178
116
Why were you expecting ipc gains from the same core as Skylake?

Some people don't spend every day reading news of future CPU's yet to be released, so it's natural that they express their disappointment on various forums after they discover this fact.
 

Ajay

Lifer
Jan 8, 2001
15,332
7,792
136
Hmm, I sure hope that KBL-X has Better thermal properties than the normal 'S' versions.
The difference between standard and delidded temps is dang high. Seems like KBL-X would be otherwise pointless.
 

dahorns

Senior member
Sep 13, 2013
550
83
91
Some people don't spend every day reading news of future CPU's yet to be released, so it's natural that they express their disappointment on various forums after they discover this fact.
And those people have no clue what ipc means . . .
 

BonzaiDuck

Lifer
Jun 30, 2004
15,699
1,448
126

No hurry, then, to e-mail Silicon Lottery for likely confirmation they will begin binning and delidding 7700K CPUs. It's a toss-up for me: take the time to do the crime myself, or pay $50 on top of the binned speed.

Point is -- I think you got to 5.0 Ghz or beyond. Today, I finally chose to tune in a good voltage for 4.7Ghz on this Skylake -- delidded, air-cooled. I only needed to re-enter BIOS a second time. The monitored VCore for 4.7 LinX and "Maximum" IntelBurnTest goes to a minimum under load of 1.408V. Since a lot of boards are shipped with default to 1.4V with "Auto," the board makers must have been informed by Intel. The unloaded peak Turbo voltage goes to 1.424V.

And since I can drop a Kaby Lake K into this Z170 board after a BIOS update, I won't worry so much about the 6700K's longevity.

Maximum Package temperature with LinX -- 77C, 25 iterations Max Problem-size and Memory. Average taken only during the test run: mid-60's. Maximum Package temperature with IBT "Maximum" -- about 70C.

The chip is binned at 4.8 Ghz. I wonder if I should try? Maybe next month . . . Too wild on the voltage side, though . . . .
 

RichUK

Lifer
Feb 14, 2005
10,320
672
126
Interesting, so the new BIOS actually allows a negative multiplier offset for AVX instructions which has allowed me to overclock to 5.1Ghz and auto drop down to 5Ghz when avx instructions are executed. Not sure if this is a new feature for Kaby Lake processors?

I have left the voltage configuration the same as the 5Ghz overclock that was 24 hour Prime95 v27.9 stable - 230mv offset with LLC level 2 - and now have prime95 v26.6 stability at 5.1Ghz for over 2 hours using the 1344 small FFT custom test (guide here: http://overclocking.guide/stability-testing-with-prime-95/). The voltage stabilises at 1.376v (instead of 1.36v when using v27.9 @ 5Ghz) as v26.6 isn't putting as much load on the processor.

When I execute Prime95 v27.9 or v28.10 the multiple automatically drops to x50 / 5Ghz as expected, so I have essentially found another 100Mhz using the AVX offset.

avx_offset.jpg


5_1ghz.jpg


Some 5.3Ghz Cinibench R15, why not:

CBR15_5_3_GHZ.jpg
 
  • Like
Reactions: Drazick and .vodka

.vodka

Golden Member
Dec 5, 2014
1,203
1,537
136
When I execute Prime95 v27.9 or v28.10 the multiple automatically drops to x50 / 5Ghz as expected, so I have essentially found another 100Mhz using the AVX offset.

Oh man, this is a great feature.

What voltage can you do 5.3Ghz non-AVX at?
 

RichUK

Lifer
Feb 14, 2005
10,320
672
126
Oh man, this is a great feature.

What voltage can you do 5.3Ghz non-AVX at?

I couldn't get 5.3ghz stable even with 1.55v, it would fail (blue screen) using P95 v26.6 in about 30 seconds.

Will be trying 5.2 a bit later - suspect I'll need around 1.45v at least, assuming it doesn't fail.
 
  • Like
Reactions: Drazick and .vodka
Dec 22, 2016
37
5
16
would upgrading from a 6600k to a 7700k a worth it upgrade from your point of view, im already at 4.6 ghz without delid on my 6600k i wanna see how far i can push it with a delid
 

BIPOLARCAT

Junior Member
Jan 6, 2017
3
0
6
So sadly, my board (Gigabyte Z170N-Gaming 5) does not support multiplier overclocking with the 7700K, even with the new BIOS that supports Kaby Lake.

The good news is that this is in a SFF system, so I'm more interested in seeing how low I can undervolt this thing to minimize load temps. Fortunately voltage adjustment works...

I do want the XMP support to get fixed though, I'm using fast DDR4-3200, but it's limited to DDR4-2133 :(

Really?! Where did you find this out??? I have a Gigabyte Z170XP-SLI and I was about to get a 7700k. Do you know if this is the case with all Z170 Gigabyte motherboards? If I can't overclock a 7700k, I think getting a 6700k might be a better option.