AT Mobile Kaveri CPU Performance Preview

csbin

Senior member
Feb 4, 2013
858
412
136
http://www.anandtech.com/show/8119/amd-launches-mobile-kaveri-apus/3

We don't have results for PCMark 8 for many of the systems, and the VAIO Pro 13 seems to be underperforming for some reason, but otherwise we get a pretty good idea of where things fall in terms of overall performance. Somewhat surprisingly, Kaveri actually takes the lead in the very demanding second pass of the x264 HD 5.0 encoding test. Granted, we're looking at a 35W APU vs. 15W CPU, and Intel's 35W quad-core parts would certainly retake the lead, but at least Kaveri is showing some real improvements over Richland in these tests. PCMark 7 meanwhile doesn't have any OpenCL optimizations and so the gap between AMD and Intel is a bit wider. Moving to the 3DMark results, Kaveri shows an impressive increase in performance over the Trinity/Richland GPU, which is expected. Intel's iGPU – particularly in the ULV system shown here – just doesn't stand a chance.
Overall, an Intel CPU with a discrete GPU is still faster than Kaveri in most areas, and a quad-core i7-4702QM would really distance itself from AMD's Kaveri…but the quad-core i7 CPUs tend to start at around $350, so there's not much point in discussing that comparison. My personal feeling is that unless you're really pushing a laptop hard, most of the modern CPUs/APUs are plenty fast. I wouldn't want a Kabini APU, but Trinity/Richland, Ivy Bridge, Kaveri, and Haswell are all going to be fine for everyday use. The more important element for me with a laptop is that I simply can't stand using conventional hard drives for the primary storage device any longer. Given the choice between an AMD Kaveri APU with a 256GB SSD (the Crucial MX100 is sure looking nice!) and pretty much any other laptop that had pure HDD storage, it would be a no brainer for me. In fact, laptop manufacturers would do their customers a great service if they took the cost savings of AMD's APU vs. Intel's CPU and put that into a decent SSD solution rather than chasing the lowest possible price!














 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
Very nice! It is matching or beating intel's equivalent in standard workloads and obliterates intel's equivalent in 3D, gaming and OpenCL.

No wonder OEM's are lining up with new mobile Kaveri designs: Acer, Asus, HP, Lenovo, Samsung, Toshiba, and more.

HP's EliteBooks are ready to go with top of the line features.

TAIPEI, Taiwan -- Hewlett-Packard's new EliteBooks, announcedat this week's Computex trade show here, will be the first to ship with AMD's new Kaveri Mobile APUs, or accelerated processing units.

The notebooks are expected to be available in June in the US, with prices starting from $739 to $799. For those looking for Intel-powered EliteBooks, those will debut in August.

http://www.cnet.com/news/hp-elitebooks-to-ship-with-new-amd-kaveri-mobile-apu/

So AMD equipped EliteBooks on the market before intel equipped EliteBooks?

Well done HP!
 

Erenhardt

Diamond Member
Dec 1, 2012
3,251
105
101
Would like to see a laptop with low end dgpu running dual-graphics. That could offer something we haven't seen before in mobile - good performance/$.

But apu only designs should be pretty have pretty good value aswell. Not only there are less parts, but there is no dgpu markup (and nv premium)

Also, I think fx-7400P with 2133 memory would be the best match. We've seen time and again that 512 GCN shaders need way more bandwidth.
 

Centauri

Golden Member
Dec 10, 2002
1,655
51
91
Holy crap, talk about closing the gap. This is a nice reversal from AMD; they're been pretty mum on Kaveri mobile and after it launches we find out it was a sleeper in the making.

Excellent.
 

TuxDave

Lifer
Oct 8, 2002
10,572
3
71
Very nice! It is matching or beating intel's equivalent in standard workloads and obliterates intel's equivalent in 3D, gaming and OpenCL.

I'm just going to toss this out there. Equivalent in price maybe, but not equivalent in form factor.

Granted, we're looking at a 35W APU vs. 15W ULV CPU, and Intel's 37W quad-core parts would certainly retake the lead (and cost quite a bit more), but at least Kaveri is showing some real improvements over Richland in these tests.
 
Mar 10, 2006
11,715
2,012
126
I'm just going to toss this out there. Equivalent in price maybe, but not equivalent in form factor.

Read: AMD's parts can't be sold for the same premium prices that Intel's parts go for, so they're a "great value" even though AMD's cost structure is far worse.

It's really hard to see how people can be cheering this obvious economic disaster for AMD...
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
Read: AMD's parts can't be sold for the same premium prices that Intel's parts go for, so they're a "great value" even though AMD's cost structure is far worse.

It's really hard to see how people can be cheering this obvious economic disaster for AMD...

Here is the Anandtech CPU and OverClocking sub forum, not NASDAQ

ps: you are also off topic
 

hungtran

Member
Jan 7, 2014
75
0
0
AMD must sacrifice margins in the short run (next 1-2 years) to regain market share, to not completely disappear from the map. It's not a disaster when it's strategic. And if you call that disaster, then what do you call Intel's contra-revenues?
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
AMD must sacrifice margins in the short run (next 1-2 years) to regain market share, to not completely disappear from the map. It's not a disaster when it's strategic. And if you call that disaster, then what do you call Intel's contra-revenues?

Please dont start the off topic, this is a Mobile Kaveri performance preview topic.
 
Mar 10, 2006
11,715
2,012
126
Here is the Anandtech CPU and OverClocking sub forum, not NASDAQ

ps: you are also off topic

AMD got booted from the NASDAQ actually, but I get your point.

This performance doesn't fundamentally change anything for AMD. It's still poor CPU performance, an oversized GPU that nobody will pay a premium for (since if they could afford it they'd go dGPU), and questionable power consumption since AMD apparently has an issue with letting people test the battery lives of these mobile focused parts.
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
AMD got booted from the NASDAQ actually, but I get your point.

This performance doesn't fundamentally change anything for AMD. It's still poor CPU performance, an oversized GPU that nobody will pay a premium for (since if they could afford it they'd go dGPU), and questionable power consumption since AMD apparently has an issue with letting people test the battery lives of these mobile focused parts.

Then let's not ask yourself what do you thinkg about iris pro, because if you wont even pay the price of a Kaveri mobile part and go for a dGPU, I cant even imagine what would you do instead of buying a $400 iris pro mobile sku.

TL;DR: Your argument is silly, specially because it is made to bash Kaveri when the same, if not worse, can apply to any Intel Core offering with acceptable iGP performance (as in: iris pro only sku's).
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
So we're looking at yet another AMD mobile part without the most important data:

Performance per watt and battery life. Yawn. Doesn't this seem reminiscent of the Beema and Mullins previews? Yes, yes it does.

Performance on mobile platforms is a moving target even with the same CPU used - OEMs can configure clockspeeds and turbo to get the battery life they want. One goes up the other goes down. Performance being high is the easy part. Performance being high with good battery life, another story, one which AMD isn't sharing.

Performance per watt and the resulting battery life gives the real picture of performance while in a mobile environment. Yawn, yawn, this is no different than Beema or Mullins. Performance data without battery life is meaningless. So I guess we're off to waiting for a design win to get some actual, meaningful results.
 

Roland00Address

Platinum Member
Dec 17, 2008
2,196
260
126
The total end performance is there, the total end performance for that specific tdp is not (another way to put this is efficiency.)

The market and thus the OEMs are moving to smaller and smaller form factors. This is a rought idea of where the mainstream laptops were and the thickness they had back them.

1.8 inches thick in about 2008
1.3 inches thick in about 2010
0.9 inches thick in about 2012 to today.

More expensive computers such as ultrabooks are even thinner than this (0.8 inches and below). Tablets are all sub 0.5 inches and some being 0.3 inches. To achieve the desired thickness of sub 1 inch you either need ulv style processors with sub 20w tdp, or better more expensive building materials which OEMs do not want to spend.

OEMS will go with the intel option if the intel option is a 17w part vs a 35w AMD part.

If you can not get the desired performance with the desired tdp you will only be in big bulky laptops which are not mainstream but instead considered value. That or you can beat intel in total performance, and thus raise AOSP, but what is the chances of that happening in laptops anytime soon.

Sigh AMD Mobile Kaveri is not a bad product but I don't think it is the product that people want.

Furthermore OEMS ruin the computers by not making smart decisions like having an ssd or putting a shitty screen inside the computer for all computers being $500 and up. When you can get a $128 gb ssd for $70 dollars; do you really need that i5 4300u+500gb hard drive in your 15" thick laptop or would you be better off with the i3 4120u+128gb ssd+500 gb hard drive. Sure you may lose 30% cpu performance by going with the i3 but you gain 6 times the sequential hard drive transfer speed, and 20 times the latency for any random read or random write.
 

Enigmoid

Platinum Member
Sep 27, 2012
2,907
31
91
Well all I can say is there goes any faith in PCMark 8. Clearly every test is OpenCL optimized not at all like more real world software.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,361
136
This performance doesn't fundamentally change anything for AMD. It's still poor CPU performance,

If that is poor performance then what makes the Core i5 4200U???





an oversized GPU that nobody will pay a premium for (since if they could afford it they'd go dGPU),

AMD playable, Intel not playable


AMD playable, Intel not playable


AMD playable, Intel not playable


There is a pattern emerging, no wonder Intel are trying their best to improve the iGPU performance more than the CPU all that time.

and questionable power consumption since AMD apparently has an issue with letting people test the battery lives of these mobile focused parts.

Im sure we will have power and battery life measurements of actual retail products when they will be available.
 
Mar 10, 2006
11,715
2,012
126
AMD must sacrifice margins in the short run (next 1-2 years) to regain market share, to not completely disappear from the map. It's not a disaster when it's strategic. And if you call that disaster, then what do you call Intel's contra-revenues?

$1 billion of "contra-revenue" for a company that will still generate ~$10 billion in net income this year is not a big deal and will pay off later.

How many years has AMD been "strategically" sacrificing margins and *still* losing share?

Anyway, I predict mobile Kaveri will be a flop and AMD will lose further share at the higher end of the notebook spectrum.
 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
Read: AMD's parts can't be sold for the same premium prices that Intel's parts go for, so they're a "great value" even though AMD's cost structure is far worse.

It's really hard to see how people can be cheering this obvious economic disaster for AMD...

Hey wait a minute, I thought it was all about design wins for you? You better run and tell HP that AMD's cost structure is far worse, I guess they never consulted you first! You notice they are releasing AMD EliteBooks ahead of intel's?

Side by side, AMD's Kaveri is far more appealing than whatever intel has. Kaveri is just getting rolling, and as HSA compatible software continues to roll out, AMD's mobile products make intel's look dated and much less advanced. Sure they compete on standard work loads, but fall far far behind in 3D, gaming, and when current and next gen software is utilized. Shills. They're so humorous.
 

Sweepr

Diamond Member
May 12, 2006
5,148
1,142
131
AMD got booted from the NASDAQ actually, but I get your point.

This performance doesn't fundamentally change anything for AMD. It's still poor CPU performance, an oversized GPU that nobody will pay a premium for (since if they could afford it they'd go dGPU), and questionable power consumption since AMD apparently has an issue with letting people test the battery lives of these mobile focused parts.

Yet some people are still cheering because the fastest 35W model is now slower than a 1-year old 15W Haswell-ULT chip (replaced months ago - Core i7 4600U is clocked 300MHz higher) by a smaller margin CPU-wise than Richland was. Also why comparable 37W Haswell dual-core & quad-core parts were not included in this comparison? A $225 standard voltage chip like Core i5 4310M should be way faster than the Core i7 4500U. While the iGPU performance is impressive (finally GCN @ mobile APUs) I wonder how much their advantage would be reduced had they used their fastest 17W part (running 256 GCN cores @ 424MHz) instead of top dog 512 cores @ 686MHz. That's 15W Haswell-U's real competitor, not the fastest 35W SKU.
 
Last edited:

Vesku

Diamond Member
Aug 25, 2005
3,743
28
86
Looks decent, question as always is what models will be available and their price. I've yet to see an AMD mobile launch where the OEMs have well balanced offerings. I think this is the 3rd or 4th product cycle where reviewers have stated "hey PC companies, how about taking some of that saved BoM going AMD and putting in a nice screen and/or a SSD."
 

hungtran

Member
Jan 7, 2014
75
0
0
Obviously Intel is a step ahead on mobile and many steps ahead on desktop. If only OEMs would start putting SSDs in AMD notebooks, it'll do a lot to mitigate the negative perception of the brand among regular consumers.
 

bullzz

Senior member
Jul 12, 2013
405
23
81
@AtenRa - u do understand that they are comparing a 35W CPU vs a 15W one right. that mean less weight, thinner chassis and possible better battery life
and that GPU is not even HD5000 let alone Iris
 
Last edited:
Mar 10, 2006
11,715
2,012
126
Then let's not ask yourself what do you thinkg about iris pro, because if you wont even pay the price of a Kaveri mobile part and go for a dGPU, I cant even imagine what would you do instead of buying a $400 iris pro mobile sku.

TL;DR: Your argument is silly, specially because it is made to bash Kaveri when the same, if not worse, can apply to any Intel Core offering with acceptable iGP performance (as in: iris pro only sku's).

Intel's iGPU architecture stinks, and Iris Pro isn't great either.

That Gen8 makeover can't come fast enough.
 

piesquared

Golden Member
Oct 16, 2006
1,651
473
136
Yet some people are still cheering because the fastest 35W model is now slower than a 1-year old 15W Haswell-ULT chip (replaced months ago - Core i7 4600U is clocked 300MHz higher) by a smaller margin CPU-wise than Richland was. Also why comparable 37W Haswell dual-core & quad-core parts were not included in this comparison? A $225 standard voltage chip like Core i5 4310M should be way faster than the Core i7 4500U. While the iGPU performance is impressive (finally GCN @ mobile APUs) I wonder how much their advantage would be reduced had they used their fastest 17W part (running 256 GCN cores @ 424MHz) instead of top dog 512 cores @ 686MHz. That's 15W Haswell-U's real competitor, not the fastest 35W SKU.

I thought power consumption != TDP? So how can you proclaim battery life of Kaveri products on TDP alone? Jarrod Walton seems to be hinting that mobile Kaveri will be very good, or even better than it's intel competition.
 
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/    |