Intel Skylake / Kaby Lake

[Sweepr]

Do a reality check. Broadwell-U is already faster. And Carrizo has a big lead only in 3dmark, real games are looking much worse as expected. Maybe you are blinded from AMDs marketing slides which are usually based on 3dmark.

Once their graphics performance lead is gone I wonder what's left?
Poor Carrizo will get smacked in ST CPU performance metrics and still won't match the slowest Skylake-U Core i5 in MT CPU stuff. No wonder even AMD doesn't have high expectations.

 

[ClockHound]

Skylake is not in the AIO, it's in the pencil. Better now?

Brilliant! Sure, Intel missed the i-smartphone market by years, but they're ready to takeover the SmartPencil niche with Skylake-HB. ;-)

 

[mikk]

You havent even provided a single link to prove that, but you keep preaching. On the other hand i have provided two links that clearly demonstrates Kaveri R6 (19W TDP) is faster than Broadwell-U HD5500 even with very old Drivers (Catalyst 13.)

You didn't provide anything from ULV Kaveri, I did. Providing links from Carrizo to make a proof about Kaveris superior performance is another laughable joke from some fanatics. In the end you are claiming that ULV Kaveri is as fast as Carrizo. Not long ago AMD itself and fanatics claimed that Carrizo is a big improvement in a low power environment over Kaveri.


Anno 2070
Kaveri 19W= 31 fps
Broadwell 15W= 41.6 fps

Tomb Raider
Kaveri 19W= 51.5 fps
Broadwell 15W= 60.6 fps

Bioshock Infinite
Kaveri 19W= 46 fps
Broadwell 15W= 43.13 fps

Metro Last Light
Kaveri 19W= 20.5 fps
Broadwell 15W= 22.5 fps

Thief
Kaveri 19W= 12.5
Broadwell 15W= 20.6
http://www.notebookcheck.net/HP-EliteBook-755-G2-J0X38AW-Notebook-Review.126452.0.html
http://www.notebookcheck.net/Acer-Aspire-E5-551-T8X3-Kaveri-A10-7300-Notebook-Review.122063.0.html
http://www.notebookcheck.net/Acer-Aspire-V3-371-Notebook-Review.135831.0.html


No benchmarks from Diablo3, Dota 2, Rome II for this Broadwell device but looking to other devices it would lose in this games as well. I shall remind you of your claim:

Even Kaveri 20W TDP will be faster than Skylake-U GT2 in real Games

Maybe you are expecting Intel going backwards with Skylake-U graphics.

I told you "19W Kaveri hardly beat 15W Haswell in gaming" and you asked me if I'm joking. Do a reality check next time.

Once their graphics performance lead is gone I wonder what's left?
Poor Carrizo will get smacked in ST CPU performance metrics and still won't match the slowest Skylake-U Core i5 in MT CPU stuff. No wonder even AMD doesn't have high expectations.

We have seen in the past they try to ignore the reality, they love to tell lies or just deny unwanted facts with things like the tester was paid by Intel.

 

[AtenRa]

You didn't provide anything from ULV Kaveri, I did. Providing links from Carrizzo to make a proof about Kaveris superior performance is another laughable joke from some fanatics. In the end you are claiming that ULV Kaveri is as fast as Carrizo. Not long ago AMD itself and fanatics claimed that Carrizo is a big improvement in a low power environment over Kaveri.


Anno 2070
Kaveri 19W= 31 fps
Broadwell 15W= 41.6 fps

Tomb Raider
Kaveri 19W= 51.5 fps
Broadwell 15W= 60.6 fps

Bioshock Infinite
Kaveri 19W= 46 fps
Broadwell 15W= 43.13 fps

Metro Last Light
Kaveri 19W= 20.5 fps
Broadwell 15W= 22.5 fps

Thief
Kaveri 19W= 12.5
Broadwell 15W= 20.6
http://www.notebookcheck.net/HP-EliteBook-755-G2-J0X38AW-Notebook-Review.126452.0.html
http://www.notebookcheck.net/Acer-Aspire-E5-551-T8X3-Kaveri-A10-7300-Notebook-Review.122063.0.html
http://www.notebookcheck.net/Acer-Aspire-V3-371-Notebook-Review.135831.0.html

No benchmarks from Diablo3, Dota 2, Rome II for this Broadwell device but looking to other devices it would lose in this games as well. I shall remind you of your claim: Maybe you are expecting Intel going backwards with Skylake-U graphics.

I told you "19W Kaveri hardly beat 15W Haswell in gaming" and you asked me if I'm joking. Do a reality check next time.

Have you seen the driver versions for the Kaveri used in those links you just posted ???

13.351.1002.1002

13.352.1004.1001

Just to remind you, latest AMD catalyst is 15.20.1046

 

[mikk]

Have you seen the driver versions for the Kaveri used in those links you just posted ???

13.351.1002.1002

13.352.1004.1001

Just to remind you, latest AMD catalyst is 15.20.1046

Intels 10.18.10.3977 isn't that much newer than that. Furthermore nobody denied the first Carrizo comparison with Intel because Broadwell reviews were using much older drivers....and suddenly it matters. 5% more or less from the driver won't change anything here nervertheless. And it's your job to prove your point properly which you didn't.

 

[AtenRa]

Intels 10.18.10.3977 isn't that much older than that. Furthermore nobody denied the first Carrizo comparison with Intel because Broadwell reviews were using much older drivers....and suddenly it matters. 5% more or less from the driver won't change anything here nervertheless. And it's your job to prove your point properly which you didn't.

Kaveri performance increased with drivers more than 20-30% from early 2014 to 2015. Not to mention those early drivers doesnt support Mantle.

 

[WaitingForNehalem]

Who cares about stupid carrizo, stop making every single thread about Intel vs AMD!!!! AMD isn't competitive, get over it and move on. Find something else to be fanatic about. It is a chip for pete's sake. So tired of every interesting thread derailed by the same people that have nothing to do except argue about Intel vs AMD. Where are the mods?

 

[WaitingForNehalem]

You forgot the chips on the back....

http://images.anandtech.com/doci/9090/IMG_2258_575px.jpg

Hmm you might be right. From KitGuru:

Intel’s decision to leave them running without a heatsink implies that the SSD’s controller is what is drawing the bulk of the unit’s 25W power consumption.

http://www.kitguru.net/components/s...ntel-ssd-750-series-1-2tb-pcie-nvme-review/3/

 

[Sweepr]

Intel defines 6 desktop market segments for CPU platforms, say Taiwan makers

Intel, in a bid to promote desktop CPU platforms, has divided the whole market into six segments in terms of market positioning and price: Enthusiast Tower catering to demand for gaming, video/audio content and high performance, Mainstream appealing by high performance-price ratios, all-in-one (AIO) PCs, Mini PCs (NUC), Portable AIO PCs and Compute Sticks. Mini PCs will support both Windows and Chrome OS, and the other five only Windows 8.1/10, according to Taiwan-based supply chain makers.

In early August, Intel will announce several K-series processors including Core i7-6700K, and Z170 chipsets and will unveil Skylake-S and Skylake-U series processors and H170/B150 chipsets in early September.

Intel will start mass shipping Skylake processors in October and November, while its top-end six-core and eight-core Broadwell-E processors will become available in the first quarter of 2016 using LGA 2011-3 and supporting the X99 chipsets and DDR4 memory.

Currently, Intel is seeing the Enthusiast Tower segment enjoy growing sales, while demand for NUC products and Compute Sticks is also gradually picking up.

www.digitimes.com/news/a20150729PD204.html

 

[frozentundra123456]

If intel was seriously wanting to cater to enthusiasts they would make a mainstream hex core. (Edit: even if they had to make another die and lose some profit margin. With 14nm they could certainly have die space, and they have no qualms about throwing money at mobile.) They want to cater to enthusiasts as long as it doesnt cut into their profit margins or interrupt their obsession with laptops and mobile, i.e. increasing performance per watt and making ever more powerful igps. Not saying they still dont make great chips, but the enthusiast/desktop market is definitely a low priority.

 

[Sweepr]

Anno 2070
Kaveri 19W= 31 fps
Broadwell 15W= 41.6 fps

Tomb Raider
Kaveri 19W= 51.5 fps
Broadwell 15W= 60.6 fps

Bioshock Infinite
Kaveri 19W= 46 fps
Broadwell 15W= 43.13 fps

Metro Last Light
Kaveri 19W= 20.5 fps
Broadwell 15W= 22.5 fps

Thief
Kaveri 19W= 12.5
Broadwell 15W= 20.6
http://www.notebookcheck.net/HP-EliteBook-755-G2-J0X38AW-Notebook-Review.126452.0.html
http://www.notebookcheck.net/Acer-Aspire-E5-551-T8X3-Kaveri-A10-7300-Notebook-Review.122063.0.html
http://www.notebookcheck.net/Acer-Aspire-V3-371-Notebook-Review.135831.0.html

Surprising results, honestly after years of fuzz about AMD delivering better graphics performance I forgot to check how their latest platforms stack up. Skylake-U GT2 (not even talking about GT3e) will definitely hurt them badly.

We have seen in the past they try to ignore the reality, they love to tell lies or just deny unwanted facts with thinks like the tester was paid by Intel.

Fun fact, AMD itself is using Cinebench 11.5 in their marketing slides now (Wasn't it Intel biased? What changed?) and I see nobody criticizing them for relying on 3DMark 11 like Intel did in the past. Obviously a best case scenario as you proved above with actual gaming benchmark results.

Even Kaveri 20W TDP will be faster than Skylake-U GT2 in real Games

Good joke. And considering OEM's don't give a crap about AMD some of the cheapest and most popular Carrizo notebooks might be running at lower TDPs (12-20W) and/or with single-channel RAM (just like HP is doing), you better forget about that reference platform performance bullshit in actual OEM notebooks.

 

[Abwx]

Intels 10.18.10.3977 isn't that much newer than that. Furthermore nobody denied the first Carrizo comparison with Intel because Broadwell reviews were using much older drivers....and suddenly it matters. 5% more or less from the driver won't change anything here nervertheless. And it's your job to prove your point properly which you didn't.

I took two of the links you posted, one with a decently cooled 7350B and the other with an as well cooled 5200U.

If we read through the text we can see that in the stress test CPU + GPU the power comsumption that can be sustained, and hence is the one available in games is for the 7350B :

(dropped to 33.7 watts after a short time)

And for the 5200U it can sustain its max power at :

31.5W

The 5200U has a 13.3 screen, no doubt that you know how to select devices since the Kaveris have 15.6 screens, is there no 5200U with such a screen...?..

Result is that the 5200U use roughly as much power as this Kaveri and is barely competitive , moreover given what Atenra s posted.

The quoted reviews :

http://www.notebookcheck.net/HP-EliteBook-755-G2-J0X38AW-Notebook-Review.126452.0.html

http://www.notebookcheck.net/Acer-Aspire-V3-371-Notebook-Review.135831.0.html


Anyway SKL wont change the hierarchy, there s a much closer distance between this SKU and Broadwell than between Kaveri and Carrizo, this latter did encompass two generations in a row and will likely do the same with the yet to be released Intel nex gen...

 

[Phynaz]

Lynnfield (i.e. Nehalem mainstream DT part) was a quad-core.

Still running one in my main desktop.

 

[mikk]

Kaveri performance increased with drivers more than 20-30% from early 2014 to 2015.

Another lie as usual.

 

[Enigmoid]

Another lie as usual.

Yep. If this was true it would mean that Nvidia did nothing wrong with Kepler at all and the perceived kepler sickness would be completely from AMD's gpus simply becoming better optimized.

However, it seems the gpu forum is all for nvidia messing up.

 

[Sweepr]

This is bad. Carrizo A10 8700P notebook load power up to to 40W:
http://www.notebookcheck.com/Carrizo-im-Test-Was-leistet-AMDs-A10-8700P.147092.0.html

So not only crap CPU performance compared to Broadwell-U (not to mention Skylake-U) and a questionable graphics performance advantage but also a power hog?

Here's a Core i5 5200U notebook barely above 30W:
http://www.notebookcheck.net/Dell-Latitude-3450-Notebook-Review.140811.0.html

Cinebench R15 - CPU Single 64Bit
A10 8700P (15-35W TDP): 69
Core i5 5200U (15W TDP): 108
Core i5 6200U (15W TDP, estimated): 118.8

Cinebench R15 - CPU Multi 64Bit
A10 8700P (15-35W TDP): 194
Core i5 5200U (15W TDP): 259
Core i5 6200U (15W TDP, estimated): 284.9

Cinebench 11.5 - CPU Single 64Bit
A10 8700P (15-35W TDP): 0.86
Core i5 5200U (15W TDP): 1.25
Core i5 6200U (15W TDP, estimated): 1.37

Cinebench 11.5 - CPU Multi 64Bit
A10 8700P (15-35W TDP): 2.36
Core i5 5200U (15W TDP): 2.82
Core i5 6200U (15W TDP, estimated): 3.1

WinRAR
A10 8700P (15-35W TDP): 1328
Core i5 5200U (15W TDP): 2326
Core i5 6200U (15W TDP, estimated): 2558

3DMark 11 Physics
A10 8700P (15-35W TDP): 2181
Core i5 5200U (15W TDP): 3349
Core i5 6200U (15W TDP, estimated): 3683

X264 HD Benchmark 4.0 - Pass 1
A10 8700P (15-35W TDP): 74.4
Core i5 5200U (15W TDP): 85.3
Core i5 6200U (15W TDP, estimated): 93.83

X264 HD Benchmark 4.0 - Pass 2
A10 8700P (15-35W TDP): 15.2
Core i5 5200U (15W TDP): 16.2
Core i5 6200U (15W TDP, estimated): 17.82

Geekbench 3 - 64 Bit Single-Core Score
A10 8700P (15-35W TDP): 2170
Core i5 5200U (15W TDP): 2691
Core i5 6200U (15W TDP): 2928

Geekbench 3 - 64 Bit Multi-Core Score
A10 8700P (15-35W TDP): 5105
Core i5 5200U (15W TDP): 5327
Core i5 6200U (15W TDP): 6195

 

[ShintaiDK]

Powerpoint!

 

[Sweepr]

Powerpoint!

The numbers don't lie. Don't you undertand basic physics laws Shintai?

 

[dark zero]

Well. It's game over for AMD & NVIDIA on the Mobile market. Despite their efforts, they didn't caught Intel up (in nVIDIA case maintain the massive advantage).

Carrizo is a massive dissaster and Maxwell is starting to show their ages... Only their highest tier dGPU would maintain them having some relevance.

Considering that OEM'S are somewhat retarded and ALWAYS goes for the cheapest option, Intel U alone will be the new standard.

If Broadwell is that dangerous, I fear that even the highest tier of AMD and Nvidia won't resist after Skylake since Kabylake is being marketed as the biggest GPU upgrade for a long time.

 

[Arachnotronic]

This is bad. Carrizo A10 8700P notebook load power up to to 40W:
http://www.notebookcheck.com/Carrizo-im-Test-Was-leistet-AMDs-A10-8700P.147092.0.html

So not only crap CPU performance compared to Broadwell-U (not to mention Skylake-U) and a questionable graphics performance advantage but also a power hog?

Here's a Core i5 5200U notebook barely above 30W:
http://www.notebookcheck.net/Dell-Latitude-3450-Notebook-Review.140811.0.html

Honestly, not a surprise. AMD is running at a 2 node disadvantage, it has an architecture that isn't that great, and it continues to lose market share.

If AMD had leadership products, I'm sure OEMs would appreciate the choice and more broadly deploy AMD rather than be beholden to a single, dominant vendor.

 

[AtenRa]

Another lie as usual.

Give me a brake,

AMD Catalyst 14.12 Omega

http://support.amd.com/en-us/kb-articles/Pages/AMDCatalystOmegaWINReleaseNotes.aspx

Year over Year Performance Optimizations versus AMD Catalyst™ 13.12 WHQL*

• Up to 19% more performance since launch in some games on AMD Radeon R Series GPUs
• Up to 29% more performance since launch in some games on various AMD APUs

 

[moonbogg]

So, Broadwell-E is coming Q1 2016? That means no Skylake-E until sometime in 2017 then? Oh well. Long live SandyBridge.

 

[mikk]

Give me a brake,

AMD Catalyst 14.12 Omega

http://support.amd.com/en-us/kb-articles/Pages/AMDCatalystOmegaWINReleaseNotes.aspx

No proof there, nothing more than the usual marketing blabla. You are sold into marketing claims it seems. "Up to" doesn't say anything. It requires just one cherry picked benchmark with a greater gain.

You have to prove that the games I compared improved by 20-30%. You also have to improve that in average of a big game test they improved by 20-30%. You won't be able to prove it because it's simply not the case. If there is a 5% improvement you can be happy.

 

[mikk]

http://diy.pconline.com.cn/675/6751941.html

Another i7-6700k test. Mixed results, some greater gains there but also some worse results. CB R15 score looks rather low, they were better results from earlier leaks. Gaming benchmarks possibly limited by GPU to some degree, apart from pcars maybe. Also a pity they didn't compare the integrated graphics.

 

[mscrivo]

http://diy.pconline.com.cn/675/6751941.html

Another i7-6700k test. Mixed results, some greater gains there but also some worse results. CB R15 score looks rather low, they were better results from earlier leaks. Gaming benchmarks possibly limited by GPU to some degree, apart from pcars maybe. Also a pity they didn't compare the integrated graphics.

If those numbers are legit, then Skylake-K better be a darn good overclocker and/or use much less power, otherwise it'll be a massive flop for desktop users.