Intel Skylake / Kaby Lake

[firewolfsm]

It's a good point, going off the trend from 90 -> 60 -> 45 nm I would have expected 14nm chips to run at or under 1 v.

 

[Ken g6]

no pentium with hyperthreading, i could swear i saw that in the skylake lineup, probably mobile. Me likey the idea ^_^

Yeah, I thought I saw that, and that all the mobile i5's were quad-core. What happened?

 

[jpiniero]

The launched mobile Skylake Pentiums do have hyperthreading. The tray price is also $161.

 

[DrMrLordX]

It's a good point, going off the trend from 90 -> 60 -> 45 nm I would have expected 14nm chips to run at or under 1 v.

What seems to be happening is that, instead of lower operating voltages, the newer chips have lower operating currents. Broadwell and Skylake seem capable of handling vcore up to/through 1.4v (depending on whom you ask) which is very interesting.

Granted, that's just the reported vcore . . . apparently both chips do interesting things with that power plane on the CPU that means the "actual" voltage fed to the cores is lower.

 

[t0mt0m]

Any update on the expected timings of the next few series of Skylake CPUs? Eg Xeon E3 v5's? Early October for release of some desktop CPUs?

 

[Fjodor2001]

Any update on the expected timings of the next few series of Skylake CPUs? Eg Xeon E3 v5's? Early October for release of some desktop CPUs?

I think September 27 is the date for most Skylake SKUs. Xeon E3 v5 will be October 27, some other SKUs in November and 2016Q1.

http://www.cpu-world.com/news_2015/...edule_of_Intel_mobile_Skylake_processors.html

 

[utahraptor]

Is there some benefit of using a new Xeon v5 in z170? Can it be overclocked?

 

[dark zero]

Yeah, I thought I saw that, and that all the mobile i5's were quad-core. What happened?

Not all the i5 mobile are Quad Core... only the H Tier was.
And the Pentium U has Hyper thread. the rest got screwed hard.

 

[Bearmann]

So far it's looking better than Sandy bridge for me. I've gone from a 2600k to a 6700k with very similar coolers and I'm getting idle temps around 28c on the 6700k vs around 36c on the 2600k both at stock speeds. The CPU temps are not much higher than ambient in the case which is awesome. Haven't had time to test load temps yet.

Also my power usage numbers look like this:

2600k+z77 mobo: 80w idle, 300w under load (witcher 3 gameplay)

6700k+z170 mobo: 55w idle, 240w under load

All other components besides CPU, ram and mobo are unchanged.

That's for the entire computer, or just the i7 6700k CPU? I was going to get a 550 watt gold PSU, but grabbed a 650 watt platinum EVGA this AM since it was on sale. I knew it was more than I needed, but the price and warranty was too good to pass up. Maybe I should have, though. I'll need to add a neon sign or something to it to suck up some more power.

 

[Deders]

Talking of power usage, check out these efficiency charts. Basically Sandra applies increasing loads from 1% to 100% and measures the average clockspeed the processors are running at in order to process these loads.

Bear in mind that my CPU is overclocked to 4.6GHz, hence why it peaks higher than the others at the end.

 

[phillyman36]

For those still looking Newegg has them (i76700k) in stock right now but only with a combo deal

9/22/15 @5am eastern time

 

[Head1985]

Another part of my test is HT scalling at 4.6Ghz.As you see slow GTX970 causes not so great HT scalling.I already tested it with cpu at 2.5ghz to neglet slow GTX970 and results are very interesting :twisted:

Test system:
i7 6700k 4.6Ghz
3000Mhz DDR4 CL15 1T
GTX970 1500/8000
win 10
Same test method like in previous test http://forums.anandtech.com/showpost.php?p=37684080&postcount=4726

CRYSIS3 HT OFF
Screenshot
http://abload.de/img/crysis3_2015_08_27_21xeja3.png
Fraps FPS Avg: 62.620 - Min: 51 - Max: 79
Frametimes and comparable fps
http://abload.de/img/crysis3fdkzc.jpg
CRYSIS3 HT ON
screenshot
http://abload.de/img/crysis3_2015_08_27_17t9krr.png
Fraps FPS Avg: 75.393 - Min: 71 - Max: 80
Frametimes and comparable fps
http://abload.de/img/crysis3wdkb9.jpg

Watch dogs HT OFF
Fraps FPS Avg: 77.327 - Min: 59 - Max: 93
Frametimes and comparable fps
http://abload.de/img/watchdogswojtv.jpg
Watch dogs HT ON
Fraps FPS Avg: 84.063 - Min: 73 - Max: 93
Frametimes and comparable fps
http://abload.de/img/watchdogspfk0v.jpg

Witcher3 HT OFF
Fraps FPS Avg: 68.359 - Min: 59 - Max: 77
Frametimes and comparablei fps
http://abload.de/img/witcher34wkjd.jpg
Witcher3 HT ON
Fraps FPS Avg: 68.776 - Min: 59 - Max: 79
Frametimes and comparable fps
http://abload.de/img/witcher3jnks4.jpg

Skyrim HT OFF
screenshot
http://abload.de/img/tesv_2015_08_27_22_13zukbx.png
Skyrim HT ON
screenshot
http://abload.de/img/tesv_2015_08_28_03_44krkc8.png

GTA5 HT OFF
Fraps FPS Avg: 65.847 - Min: 51 - Max: 75
Frametimes and comparable fps
http://abload.de/img/gta5rvk2q.jpg
GTA5 HT ON
Fraps FPS Avg: 65.425 - Min: 53 - Max: 75
Frametimes and comparable fps
http://abload.de/img/gta5zsk3p.jpg

Far cry4 HT OFF
Fraps FPS Avg: 89.517 - Min: 74 - Max: 109
Frametimes and comparable fps
http://abload.de/img/farcry4x5klr.jpg
Far cry4 HT ON
Fraps FPS Avg: 89.640 - Min: 77 - Max: 107
Frametimes and comparablei fps
http://abload.de/img/farcry4fnksw.jpg

Dying Ligt HT OFF
Fraps FPS Avg: 74.300 - Min: 57 - Max: 93
Frametimes and comparable fps
http://abload.de/img/dyinglightgvjzg.jpg
Dying Light HT ON
Fraps FPS Avg: 74.146 - Min: 56 - Max: 92
Frametimes and comparable fps
http://abload.de/img/dyinglighth9jqs.jpg

WOT HT OFF
Fraps FPS Avg: 110.366 - Min: 83 - Max: 124
Frametimes and comparable fps
http://abload.de/img/beznzvultj97.jpg
WOT HT ON
Fraps FPS Avg: 110.002 - Min: 84 - Max: 123
Frametimes and comparable fps
http://abload.de/img/wotvzjdn.jpg

 

[Sweepr]

Skylake-H Core i7 6700HQ and Core i7 6820HQ Tested (NotebookCheck)

Some results here: www.notebookcheck.com/Skylake-fuer-Notebooks-Core-i7-6700HQ-und-i7-6820HK-im-Test.150722.0.html

Power went down quite a bit compared to Haswell.

 

[psolord]

Another part of my test is HT scalling at 4.6Ghz.As you see slow GTX970 causes not so great HT scalling.I already tested it with cpu at 2.5ghz to neglet slow GTX970 and results are very interesting :twisted:

Test system:
i7 6700k 4.6Ghz
3000Mhz DDR4 CL15 1T
GTX970 1500/8000
win 10
Same test method like in previous test http://forums.anandtech.com/showpost.php?p=37684080&postcount=4726

CRYSIS3 HT OFF
Screenshot
http://abload.de/img/crysis3_2015_08_27_21xeja3.png
Fraps FPS Avg: 62.620 - Min: 51 - Max: 79
Frametimes and comparable fps
http://abload.de/img/crysis3fdkzc.jpg
CRYSIS3 HT ON
screenshot
http://abload.de/img/crysis3_2015_08_27_17t9krr.png
Fraps FPS Avg: 75.393 - Min: 71 - Max: 80
Frametimes and comparable fps
http://abload.de/img/crysis3wdkb9.jpg

Watch dogs HT OFF
Fraps FPS Avg: 77.327 - Min: 59 - Max: 93
Frametimes and comparable fps
http://abload.de/img/watchdogswojtv.jpg
Watch dogs HT ON
Fraps FPS Avg: 84.063 - Min: 73 - Max: 93
Frametimes and comparable fps
http://abload.de/img/watchdogspfk0v.jpg

Witcher3 HT OFF
Fraps FPS Avg: 68.359 - Min: 59 - Max: 77
Frametimes and comparablei fps
http://abload.de/img/witcher34wkjd.jpg
Witcher3 HT ON
Fraps FPS Avg: 68.776 - Min: 59 - Max: 79
Frametimes and comparable fps
http://abload.de/img/witcher3jnks4.jpg

Skyrim HT OFF
screenshot
http://abload.de/img/tesv_2015_08_27_22_13zukbx.png
Skyrim HT ON
screenshot
http://abload.de/img/tesv_2015_08_28_03_44krkc8.png

GTA5 HT OFF
Fraps FPS Avg: 65.847 - Min: 51 - Max: 75
Frametimes and comparable fps
http://abload.de/img/gta5rvk2q.jpg
GTA5 HT ON
Fraps FPS Avg: 65.425 - Min: 53 - Max: 75
Frametimes and comparable fps
http://abload.de/img/gta5zsk3p.jpg

Far cry4 HT OFF
Fraps FPS Avg: 89.517 - Min: 74 - Max: 109
Frametimes and comparable fps
http://abload.de/img/farcry4x5klr.jpg
Far cry4 HT ON
Fraps FPS Avg: 89.640 - Min: 77 - Max: 107
Frametimes and comparablei fps
http://abload.de/img/farcry4fnksw.jpg

Dying Ligt HT OFF
Fraps FPS Avg: 74.300 - Min: 57 - Max: 93
Frametimes and comparable fps
http://abload.de/img/dyinglightgvjzg.jpg
Dying Light HT ON
Fraps FPS Avg: 74.146 - Min: 56 - Max: 92
Frametimes and comparable fps
http://abload.de/img/dyinglighth9jqs.jpg

WOT HT OFF
Fraps FPS Avg: 110.366 - Min: 83 - Max: 124
Frametimes and comparable fps
http://abload.de/img/beznzvultj97.jpg
WOT HT ON
Fraps FPS Avg: 110.002 - Min: 84 - Max: 123
Frametimes and comparable fps
http://abload.de/img/wotvzjdn.jpg

thanks

would you be interested in partaking in the Infiltrator thread?
http://forums.anandtech.com/showthread.php?t=2448853

I would be interested in the whole run, but specifically how your skylake handles the drone scene.

Also where can I get that FLA Calculator?

 

[djsolidsnake86]

the story that major ram frequency gain fps in game is true? what frequency do you suggest to use for a 6600k?
do you think that a processor like 6600k would be good also for next generation gaming (ps5 and xbox two) ?
thank you guys

 

[Arachnotronic]

Skylake-H Core i7 6700HQ and Core i7 6820HQ Tested (NotebookCheck)

Some results here: www.notebookcheck.com/Skylake-fuer-Notebooks-Core-i7-6700HQ-und-i7-6820HK-im-Test.150722.0.html

Power went down quite a bit compared to Haswell.

Good results from Skylake!

 

[RussianSensation]

Good results from Skylake!

I disagree. Those results are awful from a new architecture and a 14nm node shrink! My soon to be 3-year-old 22nm i7 3635QM IVB scores 6.38 in Cinebench vs. 7.78 on the i7 6820HQ/HK ==> That means just a 22% increase in performance over nearly 3 years. ~7.8 is exactly what I predicted for these i7 6800 series mobile chips. Yawn, so it saves 5-7W of power under load. Who cares. If we wanted to save a lot more power, we'd move down to a MacBook Air, MacBook, iPad Pro/Surface 4 or even get a new smartphone style mobile device to reduce the power usage footprint. When I am looking for a possible upgrade from my IVB i7 in the 45-47W TDP, I see nothing attractive about the locked mobile Skylake i7s.

For i7 4700/4800/4900/5700 users, mobile Skylake is basically a failure of an upgrade unless they buy a laptop with the HK series where they can overclock to 4.0Ghz (but such a laptop is unlikely to be a proper light laptop -> aka 4.5-5.5 lbs 15.6", but more likely one of those bulky briefcase 7.5-11 lbs gaming laptops without Optimus with horrible battery life and basically worthless for productivity/mobility in the real world). The main reasons to be excited about new powerful laptops are GPU performance (desktop GTX980, and of course 16nm HBM2 GPUs next year0, 1.5-2GB/sec PCIe SSDs, superior screens with GSync, etc. The 1st generation i7 mobile Skylake SKUs are total duds imho as a viable upgrade paths from any modern i7 such as mine. Hopefully in 3 years Intel will actually improve single-threaded performance by 50% and possibly gives us a 6-core i7 in the 45W-47W TDP laptop segment so it'll finally be worthwhile to actually upgrade from IVB/SB laptops.

 

[mh272]

Guys, Is this BS? http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=9788452&CatId=11862

I want the non k version, but I thought it wasn't to be released until the 27th?

And yes, I know its tigerdirect.

 

[frozentundra123456]

I disagree. Those results are awful from a new architecture and a 14nm node shrink! My soon to be 3-year-old 22nm i7 3635QM IVB scores 6.38 in Cinebench vs. 7.78 on the i7 6820HQ/HK ==> That means just a 22% increase in performance over nearly 3 years. ~7.8 is exactly what I predicted for these i7 6800 series mobile chips. Yawn, so it saves 5-7W of power under load. Who cares. If we wanted to save a lot more power, we'd move down to a MacBook Air, MacBook, iPad Pro/Surface 4 or even get a new smartphone style mobile device to reduce the power usage footprint. When I am looking for a possible upgrade from my IVB i7 in the 45-47W TDP, I see nothing attractive about the locked mobile Skylake i7s.

For i7 4700/4800/4900/5700 users, mobile Skylake is basically a failure of an upgrade unless they buy a laptop with the HK series where they can overclock to 4.0Ghz (but such a laptop is unlikely to be a proper light laptop -> aka 4.5-5.5 lbs 15.6", but more likely one of those bulky briefcase 7.5-11 lbs gaming laptops without Optimus with horrible battery life and basically worthless for productivity/mobility in the real world). The main reasons to be excited about new powerful laptops are GPU performance (desktop GTX980, and of course 16nm HBM2 GPUs next year0, 1.5-2GB/sec PCIe SSDs, superior screens with GSync, etc. The 1st generation i7 mobile Skylake SKUs are total duds imho as a viable upgrade paths from any modern i7 such as mine. Hopefully in 3 years Intel will actually improve single-threaded performance by 50% and possibly gives us a 6-core i7 in the 45W-47W TDP laptop segment so it'll finally be worthwhile to actually upgrade from IVB/SB laptops.

50% increase in 3years and a hex core laptop? Don't ask for much do you? CPUs are a mature product now. The days of 20 to 40 percent gain and a worthwhile upgrade with each generation are gone. You may as well get used to it.

 

[iSkylaker]

Guys, Is this BS? http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=9788452&CatId=11862

I want the non k version, but I thought it wasn't to be released until the 27th?

And yes, I know its tigerdirect.

Yes, to your first question
http://us.hardware.info/productinfo/315133/intel-core-i7-6700

Hover over the red status icon.

 

[iSkylaker]

Another part of my test is HT scalling at 4.6Ghz.As you see slow GTX970 causes not so great HT scalling.I already tested it with cpu at 2.5ghz to neglet slow GTX970 and results are very interesting :twisted:

Do you have performance numbers at that clock speed?

 

[mh272]

Yes, to your first question
http://us.hardware.info/productinfo/315133/intel-core-i7-6700

Hover over the red status icon.

I had a feeling it was BS. Thanks!

 

[LTC8K6]

The different 6700 box indicates that a cooler is included. I wonder if it is the TS15A?

EDIT: Wouldn't make any sense with a 65W TDP, so probably the Haswell standard cooler.

 

[Sweepr]

A bit surprised by the difference here:

www.anandtech.com/show/9626/msi-z170a-gaming-m7-review

To be clear there was a minor difference in most tests, but every % matters in per clock performance comparisons. So far there's the RAM scaling factor, the FCLK issue/mess and now motherboard selection. Skylake is a tricky chip.

 

[Deders]

A bit surprised by the difference here:

www.anandtech.com/show/9626/msi-z170a-gaming-m7-review

To be clear there was a minor difference in most tests, but every % matters in per clock performance comparisons. So far there's the RAM scaling factor, the FCLK issue/mess and now motherboard selection. Skylake is a tricky chip.

"Where the MSI Z170A Gaming M7 wins comes down to our BIOS implementing MultiCore Enhancement by default, resulting in a better stock performance. As a result, all our base processor tests but the MSI ahead."