The Brazos Performance Preview

[Hornet331]

Here you go,the N series single core Atom based system(with turned off panel for all systems tested) has higher idle power than (un)optimized dual core Zacate ES @ 1.6Ghz :
http://hothardware.com/Reviews/AMD-Z...review/?page=8

I don't know how he got that values cause those are really bad, atom books are idel usually in the 6-8W range (with screen on). And even the Desktop Atoms (D) idel in the 12-14W range.

Some values:
N550:
Idle: 6W (max 8W)
Load: 12W (max 14W)
http://www.notebookcheck.com/Test-Ac...0.39633.0.html

D525:
Idel: 12W (max 13,5W)
Load: 26W (max load 31W)
http://www.notebookcheck.com/Test-As...k.37498.0.html

And for duploxxx:

Again D-series isn't for netbooks. Thats why it hasn't any energy saving features. I don't know why intel does this, but you usually don't find D series in netbooks.

@consuming more then the rated TDP.. yeah if the whole system would be meant, but its only the chip itself.

http://www.notebookcheck.com/Test-So...k.33629.0.html
Idel:10W (absolute max 13W)
Load:27W (absolute max:32W)

or

http://translate.google.com/translat...53-u341cc.html
Idel:7,5W
Load:21W

Those include Wlan On, screen at full brightness, so yeah im with you actuall consumption counts much more, but I doubt you can cram in all other components in the remaining 3-9W.

[informal]

so K8 = today's mainstream performance?






good IGP + incredibly low power comsumption and affordable prices will make this a winner

1.6Ghz Bobcat is competing rather well perf. wise with 1.2Ghz CULV Core 2(45nm) based part as can be seen here:
http://www.pcper.com/article.php?aid...e=expert&pid=1
That's not bad at all taking into account how tiny the Bobcat cores are compared to Penryn and also the 2way Vs 4way OoO designs of these two products. Power draw is no contest since Zacate is in its own league.

[god_43]

lol wut.. so anad reviews didn't make since for the last 3-4 years.
Hes using this benchmark setup for pretty much 3-4 years now, i like how he gets bashed more and more for doing the same thing bractically changing nothing with his benchmark setup. Plus most of the other sides pretty much use the same benchmarks and they show nearly the same results...

If you want to critizise a review then that of legit reviews, it has no consistency, they add and remove systems or do no compression at all.



You can if you concentrate on certain sub systems like the CPU, most scores of the CinebenchR10 benchmark are very consisten accros the review sites, same goes for the x264 benchmark. But if you start to compare the whole system I agree with you.

@ reviews
Amd really did some magic with the idle power consumptionn, its really really good. Compared to the Atom platform its needs ~70% less power.
Idel runtime swill be crazy for this platform, I hope the OEM don't use this fact and save on the capacity of the batteries.

The cpu part on the other hand is not that impressive, it needs 2-2,5x (~10W compared to ~4W on atom) more power then atom and delivering more performance in the range from 10-80%, avarage performance advantage is hard to tell but i guess its somewhere ~40%. But at least this really doen't matter that much because of the awesome idel power consumption the barazos platform has and overall load power consumption is in the same range as atom.

The IGP is very good for its intended market, thought the limited memory bandwidth eats more performance then I thought.

Overall the plattform is a real winner for the ultra mobile market.
Low cost entrie desktop and lowcost main stream books (15") i am not so sure if the performance is enough, since in this segement power consumption is still not the main selling point and there are already dirty cheap Pentiums and Celerons batteling (and even some i3s) the sub 500€ market.

Now what i really want to see is how the C-50 performs, cause that is the CPU that will go against the N550.

i didn't see the legit reviews so i cant comment about them. my main problem was that he picked cpus that are not in competition with each other, that's what didn't make sense?

[Nintendork]

Is it possible to add some 256MB DDR3 1600 as sideport memory? That should help the igp performance. Or at least as an option to OEM's.

[duploxxx]

I don't know how he got that values cause those are really bad, atom books are idel usually in the 6-8W range (with screen on). And even the Desktop Atoms (D) idel in the 12-14W range.

Some values:
N550:
Idle: 6W (max 8W)
Load: 12W (max 14W)
http://www.notebookcheck.com/Test-Ac...0.39633.0.html

D525:
Idel: 12W (max 13,5W)
Load: 26W (max load 31W)
http://www.notebookcheck.com/Test-As...k.37498.0.html

And for duploxxx:

Again D-series isn't for netbooks. Thats why it hasn't any energy saving features. I don't know why intel does this, but you usually don't find D series in netbooks.

@consuming more then the rated TDP.. yeah if the whole system would be meant, but its only the chip itself.

http://www.notebookcheck.com/Test-So...k.33629.0.html
Idel:10W (absolute max 13W)
Load:27W (absolute max:32W)

or

http://translate.google.com/translat...53-u341cc.html
Idel:7,5W
Load:21W

Those include Wlan On, screen at full brightness, so yeah im with you actuall consumption counts much more, but I doubt you can cram in all other components in the remaining 3-9W.

will only be able to compare once they also have material to review....

looking at the specs of the N550 atom laptop with 1GB ram, 10.1inch and 5400 rpm i hardly doubt it will consume more...

and while the N550 will probbaly consume less then the E-350 it is slower or equal in cpu always slower in gpu... and once working with it how long do you tink the battery will last if it requires much more cpu power to do something with it...
http://www.pcper.com/article.php?aid...e=expert&pid=7

lets hope we get some good platform compares soon, all I see now is that there is more competition which can only improve the consumer choice and price.

[Hornet331]

I agree, but it kinda hard to do good plattform reveiws espeical on mobile platforms.

I hope wee see from some big OEMs books wiht bowth platforms and simmilar specs, this would provide the best comparison.

[kuroikenshi]

the ratio of perf/watt + future proof (dx11 + open gl) is what's going to make bobdog (cat I mean cat) a much better option

[qcmadness]

Is it possible to add some 256MB DDR3 1600 as sideport memory? That should help the igp performance. Or at least as an option to OEM's.

Adding another memory channel (either Sideport or main memory channel) will not give it anymore advantage. In IGP / GPU-intensive tasks, Zacate already blows out it competitors. In other tasks, the added logic will consume power.

I agree, but it kinda hard to do good plattform reveiws espeical on mobile platforms.

I hope wee see from some big OEMs books wiht bowth platforms and simmilar specs, this would provide the best comparison.

I am looking forward to more info as I am planning to buy a new nettop with Zacate inside.

[instred]

the ratio of perf/watt + future proof (dx11 + open gl) is what's going to make bobdog (cat I mean cat) a much better option

I have to agree with you there. I'm looking at a tablet PC
HP has one with a CULV and 6.5 battery life on website. So if AMD new Bobcat can get over 8 hours and be that close it is a win and I hope HP or some other manufacturer will make a tablet PC with it. Just what I'm looking for.

[spursindonesia]

This seems like a winner, make it cheap & widely available AMD, i'll buy one for my wife next year if there's a solution for less than US$ 300 budget netbook, sold in my country.

[mAJORD]

On a side note:
Kinda funny observation, suddenly you hear form the same guys, that usually proclaim single thread performance has no relevance in todays usage scenarios, are all over the superior single threaded performance of brazos compared to atom...

Yes, it is Ironic, but a key point would be that it's at this performance level that single threaded perf is actually noticable and effects user experience.

In current mainstream desktop situations (~3ghz slow cores, vs fast cores) it's almost impossible to tell the difference.. except, say, a heavy PDF doc that might be chunkier to scroll through on anything but a fast core

[generics_user]

I agree, but it kinda hard to do good plattform reveiws espeical on mobile platforms.

I hope wee see from some big OEMs books wiht bowth platforms and simmilar specs, this would provide the best comparison.

there is a slight difference between K10 @ 3ghz vs. i7 @ 3ghz and ATOM vs. bobcat...

i feel no difference at all between my 1055t @ 2.8ghz/2.0NB and the 1055t @ 3.9ghz/2.7NB in desktop user experience, in fact it hardly comes out of Cool n Quiet state when browsing the web, watching HD movies etc. (due to my OC my CPU has a CnQ clock rate of 1.1ghz)

only things were i can feel a difference: Cinema 4D (rendering only), gaming (scaling stops @ 3.2 ghz with my overclocked 6850 @ 1050/1200 in most games), 3dsMAX rendering only


on the other hand ATOM is barely able to cope with windows 7 due to its anemic single thread performance, it's even worse than a 8 year old athlon XP in this regard (Windows XP on my friends 2.4ghz P4 runs smoother than on his 1.8ghz Atom HTPC (with ION)), bobcat on the other hand seems to be slightly below Dual Core K8 performance...

you only have to take a look at single thread performance (tests) to notice that ATOM completely sucks in every way you can look at it, sure bobcat isn't a good performer but at least it gives you the possibility to do something else besides texting and browsing low res flash content on the web....

[railer]

Do aprove Single threaded performance is great. Where it's really going to shine is in gaming and it does show. I think APU's is the future. in 10 years average Joe gamer might not need a discrete graphics card.

[STEvil]

so what happens when we overclock it and give it more ram and/or play with ram timings?

[zalbard]

Looks promising. Staying tuned for the pricing. It must be much better than Atom + ION!

[Enoc]

some of the specs of systems coming from both camps...


By hectorjacob

[duploxxx]

1,6/1,8 husky 32 core nice laptop in a 35W package...

[Nintendork]

Ontario is 40nm...

[Stukov]

TDP is not rated power draw, it is relating to heat created by the chip and the rated cooling required for it.

[Mechanical Man]

Ontario is 40nm...

And husky is Llano...

[Enoc]

in that chart that I posted , the are some typos but gives you an idea of what segment AMD is going for...

LLano is a 2-4 cores, that 32 core is a typo, and Ontario is 40nm, not 32nm that is correct.

[LightSpeed]

there is a slight difference between K10 @ 3ghz vs. i7 @ 3ghz and ATOM vs. bobcat...

i feel no difference at all between my 1055t @ 2.8ghz/2.0NB and the 1055t @ 3.9ghz/2.7NB in desktop user experience, in fact it hardly comes out of Cool n Quiet state when browsing the web, watching HD movies etc. (due to my OC my CPU has a CnQ clock rate of 1.1ghz)

only things were i can feel a difference: Cinema 4D (rendering only), gaming (scaling stops @ 3.2 ghz with my overclocked 6850 @ 1050/1200 in most games), 3dsMAX rendering only


on the other hand ATOM is barely able to cope with windows 7 due to its anemic single thread performance, it's even worse than a 8 year old athlon XP in this regard (Windows XP on my friends 2.4ghz P4 runs smoother than on his 1.8ghz Atom HTPC (with ION)), bobcat on the other hand seems to be slightly below Dual Core K8 performance...

you only have to take a look at single thread performance (tests) to notice that ATOM completely sucks in every way you can look at it, sure bobcat isn't a good performer but at least it gives you the possibility to do something else besides texting and browsing low res flash content on the web....

Exactly my point. Single threaded performance becomes less important when it gets to a certain level. At the level of Atom, day to day single threaded tasks seem slow. Thats where Bobcat shines, not to mention greater multithreaded performance too while consuming less/equal power. Whats the complaint?

[-Boris-]

TDP is not rated power draw, it is relating to heat created by the chip and the rated cooling required for it.

And the heat created by the chip is more or less identical to the power it consumes.

[ajaidev]

in terms of cpu performance sandy bridge would totally ride the husky but the interesting thing is that retailers putting Ontario on the same plane as CULV Sandy bridge. The CULV sandy bridge is no slouch in either GPU or CPU terms "Less CPU than GPU" i dont know how die wise Intel will win expect use the original plan and launch a single core Sandy bridge already

I dont know why they dont put out a single core sandy bridge against the likes on Ontario, i saw really good single core performance from a few core disabled cpu i dont think single core with HT running at a high speed would run into much problems?

[qcmadness]

in terms of cpu performance sandy bridge would totally ride the husky but the interesting thing is that retailers putting Ontario on the same plane as CULV Sandy bridge. The CULV sandy bridge is no slouch in either GPU or CPU terms "Less CPU than GPU" i dont know how die wise Intel will win expect use the original plan and launch a single core Sandy bridge already

I dont know why they dont put out a single core sandy bridge against the likes on Ontario, i saw really good single core performance from a few core disabled cpu i dont think single core with HT running at a high speed would run into much problems?

Because Zacate platform is expensive.
I have heard of rumours that Intel plans 1C SB against Zacate.