• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Prepare to Skylake OGL Cinebench WRs

Joined
Jan 3, 2007
Messages
454 (0.07/day)
System Name celer
Processor Pentium 4 650 3.4GHz 2MB L2
Motherboard MSI PM8M3-V
Cooling Thermalright SI-128 SE
Memory 2048MB OCZ 2-3-2-5 2T at 200Mhz
Video Card(s) Sapphire Radeon R9100
Storage 250G Samsung 850 PRO (MZ-7KE256BW) - 1024G WD Black (WD1003FZEX)
Display(s) 19' iiyama ProLite E1980SD 1280x1024 75Hz DVI
Case Eurocase moded
Audio Device(s) onboard
Power Supply Enermax 620W Liberty
Mouse Logitech MX510 red
Keyboard eTech PS/2 keyboard
Software Win XP SP3
Benchmark Scores http://hwbot.org/submission/2455634_
China based web TechBang showed (and then quicky removed, but what was once on the net... you know :D ) some Skylake comparsions to Haswell. Let's start with CPU-Z and GPU-Z screens:



The Vcore looks too low to me to be true, but whatever. The tests showed that on average, the Skylake is only about 6.1% faster that Haswell. In some tests is even slower. But there are some instances, where it is 29.1% faster, like the Cinebench 15 OpegGL test:



So to put long story short, expect some Cinebench OGL world records on HWbot pretty soon :D


BTW, as for APU performance, Intel lose big time. AMD Kaveri A10-7870K is getting in the Sky Diver test (witch is very closely tied to the games real world gaming framerates) 6400 points, however Skylake reach only 4650 points. Without eDram's Kaveri stayed unbeated.



http://diit.cz/clanek/skylake-az-o-29-procent-rychlejsi-nez-haswell
http://www.computerbase.de/2015-07/...-vergleichen-core-i7-6700k-und-core-i7-4790k/
http://www.techbang.com/posts/24629...aymore-motherboard-measured-experience?page=2 (removed)
 
Joined
Oct 22, 2014
Messages
13,210 (3.84/day)
Location
Sunshine Coast
System Name Black Box
Processor Intel Xeon E3-1260L v5
Motherboard MSI E3 KRAIT Gaming v5
Cooling Tt tower + 120mm Tt fan
Memory G.Skill 16GB 3600 C18
Video Card(s) Asus GTX 970 Mini
Storage Kingston A2000 512Gb NVME
Display(s) AOC 24" Freesync 1m.s. 75Hz
Case Corsair 450D High Air Flow.
Audio Device(s) No need.
Power Supply FSP Aurum 650W
Mouse Yes
Keyboard Of course
Software W10 Pro 64 bit
If the K version is typical and can be overclocked, then it will have a huge advantage over the non K version.
i7-6700 @ 3.4 Ghz compared to i7-6700K @ 4.0 Ghz with TDP of 65 and 95 respectively.
 
Joined
Jan 3, 2007
Messages
454 (0.07/day)
System Name celer
Processor Pentium 4 650 3.4GHz 2MB L2
Motherboard MSI PM8M3-V
Cooling Thermalright SI-128 SE
Memory 2048MB OCZ 2-3-2-5 2T at 200Mhz
Video Card(s) Sapphire Radeon R9100
Storage 250G Samsung 850 PRO (MZ-7KE256BW) - 1024G WD Black (WD1003FZEX)
Display(s) 19' iiyama ProLite E1980SD 1280x1024 75Hz DVI
Case Eurocase moded
Audio Device(s) onboard
Power Supply Enermax 620W Liberty
Mouse Logitech MX510 red
Keyboard eTech PS/2 keyboard
Software Win XP SP3
Benchmark Scores http://hwbot.org/submission/2455634_
Maybe the advantage will not be too big, because Intel Z170 chipset should provide some overclocking options even for locked Skylake CPU's, according to this: http://benchlife.info/non-k-cpu-can...lake-platform-will-better-than-last-07272015/

BCLK can be changed as well, as clock of the IGP and ram speed for the IGP. At least something... :) 5.2GHz on air and 6.5GHz with LN2 should be taken with grain of salt, tough.
 
Joined
Nov 9, 2010
Messages
5,640 (1.16/day)
System Name Space Station
Processor Intel 13700K
Motherboard ASRock Z790 PG Riptide
Cooling Arctic Liquid Freezer II 420
Memory Corsair Vengeance 6400 2x16GB @ CL34
Video Card(s) PNY RTX 4080
Storage SSDs - Nextorage 4TB, Samsung EVO 970 500GB, Plextor M5Pro 128GB, HDDs - WD Black 6TB, 2x 1TB
Display(s) LG C3 OLED 42"
Case Corsair 7000D Airflow
Audio Device(s) Yamaha RX-V371
Power Supply SeaSonic Vertex 1200w Gold
Mouse Razer Basilisk V3
Keyboard Bloody B840-LK
Software Windows 11 Pro 23H2
The tests showed that on average, the Skylake is only about 6.1% faster that Haswell.

Is that what THEY said, or what you calculated? After doing the math, I came up with an average of 7.63% faster.

Not that it matters much because this is one random leaked test. When the chips actually launch and we see some real world tests, then will know how gaming performance compares.

Keep in mind though that most making the jump to Skylake will be doing so as much if not more for DDR4 support and a more future ready platform for GPUs like Pascal.
 

cadaveca

My name is Dave
Joined
Apr 10, 2006
Messages
17,232 (2.63/day)
5.2GHz on air and 6.5GHz with LN2 should be taken with grain of salt, tough.


Screenshot was highest "suicide" possible (ie, not stable). Anyone thinking differently is misguided. That poster said as much too.
 
Joined
Jan 3, 2007
Messages
454 (0.07/day)
System Name celer
Processor Pentium 4 650 3.4GHz 2MB L2
Motherboard MSI PM8M3-V
Cooling Thermalright SI-128 SE
Memory 2048MB OCZ 2-3-2-5 2T at 200Mhz
Video Card(s) Sapphire Radeon R9100
Storage 250G Samsung 850 PRO (MZ-7KE256BW) - 1024G WD Black (WD1003FZEX)
Display(s) 19' iiyama ProLite E1980SD 1280x1024 75Hz DVI
Case Eurocase moded
Audio Device(s) onboard
Power Supply Enermax 620W Liberty
Mouse Logitech MX510 red
Keyboard eTech PS/2 keyboard
Software Win XP SP3
Benchmark Scores http://hwbot.org/submission/2455634_
The tests showed that on average, the Skylake is only about 6.1% faster that Haswell.

Frag Maniac -
Is that what THEY said, or what you calculated? After doing the math, I came up with an average of 7.63% faster.

But you calculated it WRONG. You included the graphic tests (3DMark Sky Diver, 3DMark Cloud Gate), that have no place to meddle with the score. When you exclude them, you get 6.1%, witch is IMHO pathetic.

And your math is WRONG even if you include these graphic tests, as then it is 6.9% faster:

+6.7%
+2.8%
–1.1%
–1.1%
–1.9%
+29.1%
+3.2%
–0.6%
+14.7%
+8.7%
+6.7%
–4.7%
+21.5%
+9.3%
+7.5%
+2.5%
+11.4%
+9.7%

18 values. Counted together (beware of the slower results (!)) it is 124.4. Divided by 18 it is 6.9111% faster.

If you take off these 3D Mark Sky Diver and Cloud Gate tests, you get total 101. Divided by 16 it is 6.3125% faster.
(seems that there is slight mistake in the original Diit.cz calculation at this point...)

However it would be fair to remove the other 3D Mark tests too (Fire Strike and Fire Strike Extreme) as well, as the Cinebench R15 OpenGL test, as these have nothing to do with CPU power!

So, futher subtract/add from 101 - these scores and you get 74.9. Divide that now by 13 and you get reall Skylake performance increase over Haswell:

5,7615%

That is is. Game over, dude.


...


+6.7%
+2.8%
-1.1%
+3.2%
-0.6%
+6.7%
-4.7%
+21.5%
+9.3%
+7.5%
+2.5%
+11.4%
+9.7%

13 values, total 74.9. 5.7% total.

...


Now if you want to be really thorough, then take a look as Sandra 2015 Multimedia 1 and Multimedia 2 tests. Yes, these +21.5 and +9.3% increases are serious and pushing the total Skylake performance increase up. But are they really about CPU...?

Just ask yourself...



Screenshot was highest "suicide" possible (ie, not stable). Anyone thinking differently is misguided. That poster said as much too.

Yep, you are right. CPU-Z score is not anything. Well, CPU-Z score paired with SuperPi 32M result... that is something else :) Or at least 1M:
cadaveca -
 
Joined
Nov 9, 2010
Messages
5,640 (1.16/day)
System Name Space Station
Processor Intel 13700K
Motherboard ASRock Z790 PG Riptide
Cooling Arctic Liquid Freezer II 420
Memory Corsair Vengeance 6400 2x16GB @ CL34
Video Card(s) PNY RTX 4080
Storage SSDs - Nextorage 4TB, Samsung EVO 970 500GB, Plextor M5Pro 128GB, HDDs - WD Black 6TB, 2x 1TB
Display(s) LG C3 OLED 42"
Case Corsair 7000D Airflow
Audio Device(s) Yamaha RX-V371
Power Supply SeaSonic Vertex 1200w Gold
Mouse Razer Basilisk V3
Keyboard Bloody B840-LK
Software Windows 11 Pro 23H2
Yeah I wasn't acknowledging that some were GPU vs CPU based, and I may have used the wrong divisible too.

I can't stress enough though that my saying it doesn't really matter is true though, because this is one random test, that has yet to be even confirmed accurate by anyone (which may be WHY it was posted then taken down), and it has nothing to do with real world game performance either.
 
Joined
Jan 3, 2007
Messages
454 (0.07/day)
System Name celer
Processor Pentium 4 650 3.4GHz 2MB L2
Motherboard MSI PM8M3-V
Cooling Thermalright SI-128 SE
Memory 2048MB OCZ 2-3-2-5 2T at 200Mhz
Video Card(s) Sapphire Radeon R9100
Storage 250G Samsung 850 PRO (MZ-7KE256BW) - 1024G WD Black (WD1003FZEX)
Display(s) 19' iiyama ProLite E1980SD 1280x1024 75Hz DVI
Case Eurocase moded
Audio Device(s) onboard
Power Supply Enermax 620W Liberty
Mouse Logitech MX510 red
Keyboard eTech PS/2 keyboard
Software Win XP SP3
Benchmark Scores http://hwbot.org/submission/2455634_
Glad you agree on the numbers now, but it was pain to re-check the numbers. I made few errors too and discovered even error in the original article... nevermind.

However when information is taken down, it is mostly because it is true. And after all, this information is consistent with the history.
What performance increases Intel did manage in past years? 2 to 5% per new CPU line? Same there.

What is IMHO much worser is, that the old 22nm process that made Haswell have lower TDP 88W. Now since Skylake made with 14nm process, yet have a TDP 95W. Higher TDP! No wonder Intel need new cooler:
http://www.tweaktown.com/news/46170...k-series-ship-without-stock-cooler/index.html

So it is possible, that the 14nm process have a huge leakage. This might be perfect for LN2 guys and some Earth-shaking WR to be made, but for average Joe it is bad news. I hoped for being able to quietly cool down serious power, but once again this is not going to be the case... :(

And no, integrated GPU is not counted into the TDP number. No. You cannot explain it this way away. That would be wishfull thinking only.

(hopefully I'm wrong there)
 
Top