Mobile kaveri

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
To give it some reference point: My a10-7850k took 290W at the wall

Well, it did run on 1.55V thou
Asrock.... rock!:hmm:

Waiting for reviews
Also, 19W TDP is overlapping with beema, is it not?
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
To give it some reference point: My a10-7850k took 290W at the wall

Well, it did run on 1.55V thou
Asrock.... rock!:hmm:

Waiting for reviews
Also, 19W TDP is overlapping with beema, is it not?

turbo and the gpu will be the differentiators.
 

Xpage

Senior member
Jun 22, 2005
459
15
81
www.riseofkingdoms.com
I hope they get some design wins and it gets out as a product. Be nic ein a HTPC too but I think mobile richland wasn't really available in the retail channel
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,926
404
126
Based on the info on the article, would anyone case to estimate how this would match up against an Intel Haswell U4200?

I'm thinking both with regards to max performance and battery life.
 

strata8

Member
Mar 5, 2013
135
0
76
Based on the info on the article, would anyone case to estimate how this would match up against an Intel Haswell U4200?

I'm thinking both with regards to max performance and battery life.

30% faster in single threaded tasks. 15% faster in multithreaded if Kaveri can only hit its base clocks but it might be a wash if it can go higher than that.

Also, 19W TDP is overlapping with beema, is it not?

Yes, CPU-wise it's very close to the A6-6310 but with a 30% higher turbo. Especially considering Steamroller and Jaguar/Puma have similar IPC as well.

A6-6310 - 4 cores, 2 GHz base, 2.4 GHz turbo, 15W TDP, 107mm2
A10-7300 - 4 cores, 2 GHz base, 3.2 GHz turbo, 19W TDP, 245mm2

The Kaveri part will have a much better GPU obviously but the comparison will still be very interesting.
 

coercitiv

Diamond Member
Jan 24, 2014
6,400
12,849
136
Based on the info on the article, would anyone case to estimate how this would match up against an Intel Haswell U4200?

I'm thinking both with regards to max performance and battery life.
I presume you're referring to i5 4200U. Performance wise they should trade blows in single threaded tasks, Kaveri should pull ahead on multithreaded tasks if turbo can be maintained. Battery life should go to the Intel platform.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,926
404
126
30% faster in single threaded tasks. 15% faster in multithreaded if Kaveri can only hit its base clocks but it might be a wash if it can go higher than that.

So are those percentages in Kaveri's or Haswell's favor? I.e. is the mobile Kaveri or Haswell the fastest?

And regardless, I'd say mobile Kaveri will be close enough to compete performance-wise with the mobile Haswell CPUs. I think that's a huge step for AMD in that case.
 

Haserath

Senior member
Sep 12, 2010
793
1
81
Based on the info on the article, would anyone case to estimate how this would match up against an Intel Haswell U4200?

I'm thinking both with regards to max performance and battery life.

4200u is 1.6 base/2.6 turbo vs 7300 2.0/3.2.

Haswell 10-20% faster single/dual. Push at 3 threads. Kaveri 10-20% faster multi.

Haswell would most likely win for battery life; it has the on package chipset and IVR.

Beema looks like it would be a much better choice for 10-20W.
 

strata8

Member
Mar 5, 2013
135
0
76
So are those percentages in Kaveri's or Haswell's favor? I.e. is the mobile Kaveri or Haswell the fastest?

And regardless, I'd say mobile Kaveri will be close enough to compete performance-wise with the mobile Haswell CPUs. I think that's a huge step for AMD in that case.

Whoops, sorry. They're in Haswell's favour.
 

Haserath

Senior member
Sep 12, 2010
793
1
81
Due to what reasons? Battery life, price, ...? Because as I understand it, Beema will quite a lot slower than mobile Kaveri.

Mobile Kaveri doesn't have enough extra performance to really make the difference for any workloads. They have similar IPC, so Beema will be close in all but the single threaded stuff. The graphics are also limited by TDP and memory performance at these power levels.

Beema costs less and, with its highly integrated nature, it uses a good bit less power.

I'd say the performance crossover point is 15W. Beema would perform better <15W. I'm only assuming, of course.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,926
404
126
Mobile Kaveri doesn't have enough extra performance to really make the difference for any workloads.

If that is true, and mobile Kaveri in turn is close to mobile Haswell in performance, then it sounds like Beema will be a very good chip for laptops. I.e. you get performance and battery life that is close enough to Haswell, but much cheaper.
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
If that is true, and mobile Kaveri in turn is close to mobile Haswell in performance, then it sounds like Beema will be a very good chip for laptops. I.e. you get performance and battery life that is close enough to Haswell, but much cheaper.

Mobile Kaveri likely won't be anywhere near haswell but will be a massive step up from richland.

Considering that a 2.6 ghz Haswell is about even with a 4.4 ghz a10-6800k or a 4 ghz a10-7850k in single thread performance (CB 11.5) haswell will have a massive singlethread performance lead.

Clocks on low power richland look really good but when you actually run the chips the real world clocks are lower than expected.

ST and MT will most definitely favour the i5-4200U.

http://www.notebookcheck.net/Intel-Core-i5-4200U-Notebook-Processor.93563.0.html

http://www.notebookcheck.net/AMD-A-Series-A10-5750M-Notebook-Processor.92882.0.html

The i5-4200U approximately matches the a10-5750m (highest bin 35W richland) in every MT test (maybe 5% weaker on average for MT).

Looking at clocks, the 7300 has no chance matching MT performance. At 19W TDP this chip is going to run base clock on the CPU under 4C load-- very aggressive turbo.
 

coercitiv

Diamond Member
Jan 24, 2014
6,400
12,849
136
Looking at clocks, the 7300 has no chance matching MT performance. At 19W TDP this chip is going to run base clock on the CPU under 4C load-- very aggressive turbo.
Turbo speeds in MT loads will indeed make or brake this chip.
 

Yuriman

Diamond Member
Jun 25, 2004
5,530
141
106
I presume you're referring to i5 4200U. Performance wise they should trade blows in single threaded tasks, Kaveri should pull ahead on multithreaded tasks if turbo can be maintained. Battery life should go to the Intel platform.

I'm sure either company could make a mobile chip with a 4ghz+ turbo. It's quite another matter if it will stay at, or even reach those clocks. Bring the GPU into the picture and what CPU clocks will we get?

Not bashing, just pointing out that max turbo clocks are nearly meaningless in a TDP and battery limited environment.
 

coercitiv

Diamond Member
Jan 24, 2014
6,400
12,849
136
I'm sure either company could make a mobile chip with a 4ghz+ turbo. It's quite another matter if it will stay at, or even reach those clocks. Bring the GPU into the picture and what CPU clocks will we get?

Not bashing, just pointing out that max turbo clocks are nearly meaningless in a TDP and battery limited environment.
The turbo clocks are extremely meaningful. For example, in the case of the i5 4200U, the chip is able to maintain 2.3Ghz CPU turbo speed when loaded with Prime 95, and will guarantee base clocks when loaded with Prime 95 & Furmark, essentially a power virus.

The way it works on long term loads is the chip will mantain the highest clocks possible as long as power usage does not exceed TDP rating. Temperature plays a role too, but only if the cooling assembly is not dissipating enough heat to keep up with the rated TDP.

Some higher clocked Haswell ULT models might not be able to maintain their full turbo clocks under this kind of extreme stress (clocks go higher, TDP is still 15W), but I believe this example is enough to show turbo clocks are important.

Bring the GPU into the picture and what CPU clocks will we get?
Well, we'll have to see what happens then. We'll be measuring performance based on what the both CPUs and GPUs have to offer.
 

insertcarehere

Senior member
Jan 17, 2013
639
607
136
This took a while, I wonder if AMD would put kaveri with 512sp igpu into the 19-25W range, that could make a very intriguing and gaming capable 'ultrabook'
 

coercitiv

Diamond Member
Jan 24, 2014
6,400
12,849
136
We should also note that i5 4200U has been replaced by i5 4300U, which bumps up CPU clocks by 300 Mhz and GPU clocks by 100Mhz. This SKU will be clearly superior when it comes to CPU performance and power consumption for a given task.

I'm really curious to see the mobile Kaveri GPU performance, but i'm not holding my breath on this one.
 

rainy

Senior member
Jul 17, 2013
508
427
136
Mobile Kaveri doesn't have enough extra performance to really make the difference for any workloads. They have similar IPC, so Beema will be close in all but the single threaded stuff. The graphics are also limited by TDP and memory performance at these power levels.

Honestly, I can't agree with you: it's true that IPC of Puma/Jaguar is quite similar to Steamroller, however in part of the tasks (e.g. video encoding) big core family would be better/faster than cat core (clock for clock of course).
In this particular case, I presume a 20-30 percent higher CPU performance because of much higher turbo in Kaveri.

On GPU side this should be an easy match: A10-7300 is a direct successor of A8-5545M (not A10-5745M as it was described by CPU-world) which means 384 SP and of course dual channel MC - A6-6310 IGP would be most probably about 3x slower.
https://en.wikipedia.org/wiki/List_of_AMD_APU_microprocessors#.22Richland.22_.282013.2C_32_nm.29_2


This took a while, I wonder if AMD would put kaveri with 512sp igpu into the 19-25W range, that could make a very intriguing and gaming capable 'ultrabook'

I'm a bit sceptical about 512 SP in mobile part and for sure not in TDP range which you've mentioned (rather 35-45W).
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
This took a while, I wonder if AMD would put kaveri with 512sp igpu into the 19-25W range, that could make a very intriguing and gaming capable 'ultrabook'

It would be basically worthless, much like HD 5000 is on HW ULV. Too tdp limited (and BW) to make a difference.

The turbo clocks are extremely meaningful. For example, in the case of the i5 4200U, the chip is able to maintain 2.3Ghz CPU turbo speed when loaded with Prime 95, and will guarantee base clocks when loaded with Prime 95 & Furmark, essentially a power virus.

The way it works on long term loads is the chip will mantain the highest clocks possible as long as power usage does not exceed TDP rating. Temperature plays a role too, but only if the cooling assembly is not dissipating enough heat to keep up with the rated TDP.

Some higher clocked Haswell ULT models might not be able to maintain their full turbo clocks under this kind of extreme stress (clocks go higher, TDP is still 15W), but I believe this example is enough to show turbo clocks are important.


Well, we'll have to see what happens then. We'll be measuring performance based on what the both CPUs and GPUs have to offer.

Hmm... A lot of the time ULV Haswell cannot maintain clocks under extreme load.



Looks like igp is prefered under load on this notebook. Short term turbo to 24W.

Haswell CPU is really efficient. Its the GPU that is disproportionately power hungry.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |