AMD’s Brazos vs. Atom Thermals, Revisited
by Jarred Walton on January 14, 2011 5:35 PM ESTAMD’s Brazos vs. Atom Thermals, Revisited
Last week, we met with AMD at their CES location to see some of their upcoming systems and laptops. While they’ve also recently released several new desktop GPUs, there wasn’t anything new to discuss in that area. The same applies to their desktop CPUs—we’re all waiting to see Llano and Bulldozer. So the focus at CES was understandably on Brazos, aka the “first APU” Vision C- and E-series processors.
We’ve been critical of some of the staged platform comparisons we’ve seen in the past—as Anand put it, the onus is on AMD in this case to provide a truly representative comparison between their new product and Intel’s competing offerings. After the demonstration of their Brazos netbooks on Thursday, AMD called us back and said they wanted to let us rerun the tests to make sure we accurately represented the two platforms. See, there was a slight snafu in the initial thermal imaging comparison. Specifically, AMD thought they put out a netbook with a C-50, but the test system was actually a C-30. So, we returned….
The reason for the mix-up was simple: they had both a C-30 and C-50 system from the same OEM, and they’re basically identical (one was dark blue and the other was light blue). Given that the two C-series parts are both 9W TDP, we didn’t expect much to change, and the new testing confirmed this. We did get some better images of both the top and bottom of the three test netbooks—Atom N550 vs. C-30 and C-50. Unfortunately, stupidity on my part resulted in the loss of said images (it’s a long story…), so all we have are the thermal shots from the keyboard area and screenshots showing CPU utilization during playback along with screen captures taken with FRAPS.
_thumb.jpg)
_thumb.jpg)
_thumb.jpg)
The above gallery shows essentially the same thing as our initial testing: Brazos using its GPU uses less power and runs cooler than Atom N550 doing the decoding in software. The difference between the C-30 and C-50 is pretty much non-existent, as expected. The testing environment was not conducive to doing any form of noise comparison, so while the N550 setup was clearly warmer we couldn’t say if it was quieter or not. Battery life is looking to roughly equal Atom, so that’s good to see. Now we’re waiting for final hardware to see if we can shed any more light on the situation, as well as running our full suite of tests.
We also took the opportunity to capture a video showing the 1080p playback comparison, as that’s part of the story. The video in question is Big Buck Bunny, an open movie demo created as part of the Peach movie project. (You can read more about it on their site, though it’s old enough now that if you haven’t heard of it already there’s not much to add. Suffice it to say, the lack of any licensing issues meant BBB was all over the CES floor, and I’m tired of the short now!) This particular version is a stereoscopic rendering, so instead of the normal 24FPS the frame rate is 48FPS according to FRAPS.
I believe during playback Arcsoft TotalMedia Theater 5 is skipping half the frames, as none of the netbooks come equipped with a 3D 120Hz panel. Does that actually matter? Not that we could tell—now that we’re home from CES, I ran the regular 24FPS version of Big Buck Bunny on a different dual-core N550 netbook, and frame rates still frequently dropped into the teens. Actually, it was worse than the netbook at AMD’s demonstration, but that’s probably more to do with lack of optimizations and some bloatware that came preinstalled; but I digress….
You can see during playback that the Atom N550 periodically stutters and drops below 48FPS—and more importantly, it’s far below 24FPS as well at times. In comparison, both the Vision C-30 and C-50 Brazos/Ontario chips manage a consistent 48FPS. The C-30 does flicker between 47 and 48FPS, but again, that may simply be an artifact of using a stereoscopic 3D video on a non-3D panel. Temperatures are in line with what we reported in our earlier coverage, and the two AMD netbooks are virtually identical. CPU utilization on the dual-core C-50 is lower by about half, as expected.
Once More, With Feeling
This is essentially the killer app of Brazos compared to Atom, and it’s important to keep things in perspective. These chips have a much better IGP than Atom, but at least on the nettop side of things the faster AMD E-350 isn’t miles ahead of Atom D510 in the CPU department. When we drop clock speeds down to 1.0GHz (dual-core C-50) from 1.6GHz (E-350) and compare that to the Atom N550 (1.5GHz)… well, 62.5% of the performance of E-350 compared to 90.4% of the performance of D510 means that in some tests the N550 will probably beat the C-50 for raw CPU potential. Yeah, that’s a concern for me. The GPU is the real difference, so naturally a video decoding test is the best-case scenario. I suspect C-50 will be underpowered for most 3D games, even if the DX11 GPU inside Brazos is fast enough—it will just be the AMD equivalent of Atom + NVIDIA ION, only without as many discrete chips.
We also have to consider performance of the next tier of CPUs and IGPs. Atom is the lowest of the low hanging fruit; we have much faster chips and IGPs from both AMD and Intel, and we don’t need to move up to current generation parts like 2nd Gen Core processors. Even the old Core 2 Duo CULV chips are a darn sight faster than Atom (2x-3x faster), and bad as GMA 4500MHD is, it could do an okay job at H.264 offload. It appears that the E-350 will end up delivering performance roughly equal to the old CULV chips (probably a bit slower, to be honest). That means it will also be around the same level as the Athlon II Neo K325, only with a better IGP and apparently improved power characteristics.
The biggest point in favor of Brazos isn't performance, though. It's going to be cost. If AMD can get partners to put out $400 netbooks (hopefully without Win7 Starter and with more than 1GB RAM), that will hopefully put the nail in the current iteration of Atom. We've seen the Brazos chips, and they're extremely small—smaller even than Atom—so pricing should be very compelling. AMD also doesn't appear concerned about protecting their more expensive mobile offerings (mostly because there aren't many), so they don't have to castrate Brazos in the same way Atom has been stagnant since the first N270 rolled out. Well equipped Brazos netbooks (and nettops) in the $500 range should also be a more elegant choice than Atom + ION/NG-ION, so again AMD looks set to win several matchups.
We’re working to get Brazos hardware in for testing as soon as possible, but it looks like the biggest beneficiaries will be users that want good H.264 decoding in a 10.1” form factor, or an alternative to ION. If you’re looking for the ultimate HTPC chip, we’ll have to investigate that area in further detail, as bitstreaming support and other features are still a question mark. Right now, Brazos is shaping up to be what we all wanted from Atom last year; whether that will be enough in 2011 remains to be seen.
Facebook
Twitter
RSS
151 Comments
View All Comments
SeetheSeer - Friday, January 14, 2011 - link
I directed my post to Anand, not seeing that Jarred wrote the article. However, my point still stands when directed in a larger context to the site as a whole.krumme - Friday, January 14, 2011 - link
For loads of irrelavant multithreaded benchmarks, cpu heavy gaming benchmarks and encoding benchmarks that absolutely none of the target customers is going to use - if they know the names of the situations at all.Who cares about ordinary surfing, wav-mp3 conversion, zip, l4d type gaming and just plain ordinary office work?
Lets prepare for comparing battery life to Atom, efficiency to culv, power to core2. And let the 75mm2 on 40nm TSMC for 100usd all included stand as a minor issue compared to to abolutely catastrofic lack of bluray 3D
nitrousoxide - Friday, January 14, 2011 - link
Well that's because "ordinary surfing, audio conversion and plain ordinary office work" can not be benchmarked. A test relies on data to convince everyone because it makes no sense the test says "it works fluently in surfing and office work". Everyone believes that APU is a nice chip, but it cannot do well in benchmark, and therefore, the media has a reasonable motivation to write the test in the way Intel PR told them :)JarredWalton - Friday, January 14, 2011 - link
Sorry, but you're completely wrong. Intel said nothing to me; AMD would have loved for us to be a lot more positive, but that's not how I see things. Atom is definitely "too slow", CULV was fine about a year ago, but even web surfing demands a lot.Things I'll be happy to try and quantify: Facebook performance. I am on FB a lot, and it can be a bit sluggish even on my desktop. Sure, Bejeweled Blitz and Zuma Blitz are casual games and not something everyone needs, but I can say without doubt that they suck on Atom. Will they be better on Zacate or Ontario chips? I doubt it, simply because I don't think Flash games leverage the GPU enough. (Zuma Blitz for instance pegs one of my Core 2 Quad CPU cores at 100% when I play it.)
If "works fluently for surfing and office work" is all you want/need, you can use pretty much any computer built post-2004 and be happy. If you throw in video playback, as stated, that's the killer app for Brazos. If you're after a snappy experience, I don't think even an SSD will pull Atom or Brazos up to the level where I can do everything outside of gaming and never think, "this is taking too long." But hey, I'll be happy to try it. Give me some way to quantify this in a graph, though, because subjectively saying it "works fluently" is marketing speak for "slow but fast enough to run Windows 7". Next year, even ARM will be in that same category.
nitrousoxide - Friday, January 14, 2011 - link
I'm sorry but you pretty much misunderstood what I said. I mean, that's what's happening in my country, where all the media are downplaying APUs by comparing it to a Core i3 or Athlon or Phenom or something else that is irrelevant. I never doubted the fairness of your test on Anandtech (except the ones for Phenom II X6s) and I'm just saying a common approach taken by many media manipulated by Intel PRs.JarredWalton - Friday, January 14, 2011 - link
Sorry for the misunderstanding. I get accused of being anti-AMD so often that it's almost laughable. (See the comments in the Global Foundries/AMD article from last week.) While many people can easily get by with slower CPUs, I'm sure I'm not one of them, and I have trouble believing most of our readers are.Facebook is a huge one for example; as a parent, what happens when your kid starts to complain about all his/her favorite Facebook games running like poop on their new laptop? I don't play Farmville (anymore at least... LOL), but virtually every Facebook game that leverages graphics seems like it will kill anything less than Core 2 Duo. I suppose we can take the Apple approach and just blame Flash, but that's a cop-out as well. People are making millions of dollars by creating Facebook games, and because of the ease of creation that Flash affords, we get content from unexpected avenues.
Anyway, if anyone wants to accuse me of Intel fanboyism, read the last paragraph of my intro to an old Buyers' Guide (back in the days when AMD was actually ahead): http://www.anandtech.com/show/1550 Pretty much every BG I did up until Core 2 Duo launched said something to the effect of "AMD is the better choice". Unfortunately, the shoe is on the other foot now.
kc77 - Saturday, January 15, 2011 - link
It's seems Anandtech always gets in to trouble when it forgets the level of quality it's readers expect from it.If a Core2Duo is the bees knees then what was the purpose of Atom? Atoms purpose is the same as Brazos and that is to fit into small form factors at lower power, and lower price than what CULV processors can fit into. The bone to pick isn't that one time long ago you said something positive about AMD. The problem is that you've done an editorial with two benches and IR and from there you've interjected paragraph upon paragraph which questions the validity of Brazos even though the benches aren't really there for us to see in detail in order to agree with the articles premise or even disagree with it.
Atom wasn't "low-hanging fruit" when it didn't have a competitor, but now it is ? When has AMD ever been able to come close to Intel in power per watt in any market? I can't remember when it's ever happened.
You said this, "This is essentially the killer app of Brazos compared to Atom, and it’s important to keep things in perspective. These chips have a much better IGP than Atom, but at least on the nettop side of things the faster AMD E-350 isn’t miles ahead of Atom D510 in the CPU department. When we drop clock speeds down to 1.0GHz (dual-core C-50) from 1.6GHz (E-350) and compare that to the Atom N550 (1.5GHz)… well, 62.5% of the performance of E-350 compared to 90.4% of the performance of D510 means that in some tests the N550 will probably beat the C-50 for raw CPU potential. Yeah, that’s a concern for me. The GPU is the real difference, so naturally a video decoding test is the best-case scenario. I suspect C-50 will be underpowered for most 3D games, even if the DX11 GPU inside Brazos is fast enough—it will just be the AMD equivalent of Atom + NVIDIA ION, only without as many discrete chips."
What benchmark did you use here? What was power like? We have no idea. I can't agree with the premise because you've provided no numbers other than you downclocked a E-350 processor to 1Ghz and compared it to a DC processor running at 1.5. You shimmy over to talk about CPU prowess using what benchmark we don't know and then parry over to talk about weak GPU performance based on what? Certainly not anything earlier in the article. The logic to make certain statements is weak to say the least because they are written just by themselves with their supporting evidence being nothing more than "feeling" or suspicions, which in the later paragraphs starts talking about how great Brazos is again. Huh? (BTW less discrete chips at this price point matters.)
The problem isn't that there's some disparaging remarks against AMD. The problem is that the article goes too far with opinion without enough benches to back up what it's saying.
"Unfortunately, the shoe is on the other foot now." Based on what?
JarredWalton - Saturday, January 15, 2011 - link
Way to ignore what I've said about Atom from the beginning, but let's just grab snippets from a few of our Atom review conclusions:HP 5103 (N550): "Looking just at the performance side of things, then, the current Atom landscape fails to impress. Intel knows this, and so do the manufacturers, which is why the focus with netbooks now is more about providing an inexpensive secondary mobile device where you can access all of your regular programs and data, albeit slowly."
HP 5102 (N450): "Great Build, but Still Atom... Given we’re looking at a netbook that’s now past its prime, we’re still generally impressed with what the Mini 5102 offers. The build quality is far better than anything we’ve seen from a netbook to date, with the best keyboard in a 10” netbook we've ever used. It also comes with a matte screen that’s much easier to use outdoors, though contrast ratio is still poor. However, for $100 more than the competition this is essentially the same old Atom performance."
ASUS 1215N: "A Decent Netbook, but Atom Remains Slow... CULV will eat the dual-core Atom for lunch and not even bother spitting the bones out. You don’t even need to ask about Core i3 or its ULV equivalent."
Samsung N210: "At the end of the day, the insides of the Samsung N210 aren't likely to inspire, and they weren't supposed to. Netbooks just aren't exciting kit where hardware and performance are concerned...."
Or the first Pineview netbook review, the ASUS 1005PE: "Two Steps Forward, One Step Back... Internally, the changes haven't resulted in a system that's significantly faster, but the one change is very noticeable. Our initial tests indicate that Pineview has improved battery life on the 1005 laptops by a whopping 25%. If you want a netbook that can last 10+ hours and you want to spend less than $400, the ASUS Eee PC 1005PE has exactly what you need.... The 1201N is significantly faster than anything you're likely to see from Pine Trail netbooks. Dual-core Atom 330 trumps single-core N450, and ION pummels the GMA 3150, takes its lunch money, and sends it crying home to mama.... The latest release of Atom brings quite a few changes, but the net result isn't quite as impressive as we were hoping. We have an integrated memory controller in the CPU along with a GPU on package. Those are cost saving measures that also provide some benefits in terms of power requirements. What they apparently don't provide is a significant improvement in performance."
The Pine Trail launch was when I wanted something more like Brazos. I wanted a GPU more like ION, and we got GMA 3150. However, that was over a year ago! It's great that Brazos is better than Atom in the GPU department, but look at our nettop preview of Brazos. 1.6GHz E-350 vs. 1.66GHz D510, and the performance difference is anywhere from 10% to 75% faster for AMD. Great. Now cut the clock speed of the E-350 down to 1.0GHz, and chop the GPU clock almost in half, and I'm a lot more concerned.
E-350 is looking like a solid offering, and the Lenovo ThinkPad X120e is probably the most compelling laptop I've seen so far with Brazos. Battery life will probably be six hours instead of 10-12 (just an estimate -- it might hit eight), but performance should be about 60% faster than C-50. 60% is a huge difference; I don't think anyone would disagree with that. At 1.2GHz with the single-core C-30, some CPU tasks should still be faster on AMD than Atom N450, but 1.0GHz on an architecture that appears to be slower clock-for-clock than K8 and CULV makes me concerned.
Sorry if that's not specific enough for you. How about this. One year ago, I wrote a roundup of three CULV laptops. The title: "Intel Core 2 CULV Roundup: Who Needs Atom?" From the conclusion again: "CULV is a great way to balance performance, battery life, and cost while keeping a small size. Given the choice, we'd either go with one of the CULV models or we'd move up to CULV with switchable graphics; Atom remains more of a secondary (or tertiary) computer. Sure they cost more, but we feel the cost is justified."
I don't know how many times I have to say Atom is a lousy solution for most users before it sinks in. So when someone comes along and states that "Atom wasn't "low-hanging fruit" when it didn't have a competitor, but now it is?" you're clearly running around with blinders on. We gave exactly one editors' choice award to an Atom netbook; that was the ASUS 1001P, which happened to deliver a nice matte screen with a high contrast ratio along with all the usual Atom shortcomings. We called it our favorite netbook, and so it was, but even then we ended the review with this:
"The final caveat is that you need to stop and ask yourself: do you want a netbook, or do you want to pay ~$100 more for an 11.6" CULV machine like the Acer Aspire 1410 or Gateway EC14? If your only concerns are price, size and battery life with no regards to performance, a netbook will do the job. With over twice the computing power, a much more livable WXGA screen resolution, 802.11n/Gigabit Ethernet, and not much in the way of a size, weight, or battery life penalty, it's hard not to recommend going the CULV route."
So please don't try to pawn me off as an Intel shill, or someone that changes tune every time AMD comes out with a new part, or whatever other label you happen to have. Brazos isn't the greatest thing since sliced bread; it's merely "better than Atom", which was marginal at best. I've been pointing at CULV as the minimum target I'd recommend to most people for over a year now, so if Brazos chips can't hit CULV performance levels I'll be disappointed. Then I'll end up with, "well, CULV for $100-$150 less" and walk away saying it looks like an okay platform but there's plenty of room for improvement. And the next iteration with up to four Bobcat cores isn't exactly what I'm looking for; how about two cores in a netbook that can clock up to 2.0GHz and still not require more than 12W? That's what I want from Brazos today, because then it would be faster than CULV in every test, it would have a better GPU, and it would actually decimate Atom netbooks.
Okay, I'm done. Carry on with the flame wars.
kc77 - Saturday, January 15, 2011 - link
I don't think you get where I'm coming from. You can provide bad things you've said about Atom until the cows come home. That's not apart of my concern nor does it factor much into my statements. My issue is not that there's some bad things said about AMD. The problem is that this editorial and no other is lacking in benchmarks to support the claims that it is making.Yet somehow you're able to come up with a thesis on performance and comment with worry and suspicion when you haven't provided much in the way of benchmarks. This isn't about flaming you. This is about wondering on the lack of quality. You have your opinion for sure, but in THIS editorial there's not much to back it. That's the issue. A secondary issue is the lack of logic when making statements when they aren't backed by real benchmarking numbers. Lets take a look at your restatement.
"It's great that Brazos is better than Atom in the GPU department, but look at our nettop preview of Brazos. 1.6GHz E-350 vs. 1.66GHz D510, and the performance difference is anywhere from 10% to 75% faster for AMD. Great. Now cut the clock speed of the E-350 down to 1.0GHz, and chop the GPU clock almost in half, and I'm a lot more concerned."
Houston we have a problem! You haven't provided benches so that we can SEE it for ourselves as to why exactly you are concerned. This isn't PCWorld or Fudzilla it's Anandtech. There's a difference at least in my eyes. What CPU test did you run? Was it Photoshop? Cinebench? Angry Birds? What was is it that gave you that suspicion, worry or concern? We don't know. We will never know until you provide what metric you used.
"So please don't try to pawn me off as an Intel shill, or someone that changes tune every time AMD comes out with a new part, or whatever other label you happen to have. "
Did I say you were? Nope. Don't pass me off as just another person who is critiquing your editorial because there were unfavorable things said about AMD. I could care less. You've said other things in the past, but they were backed up by benches so there's no need to say anything. It was what it was. However you do make it easy for people to say such things when you say...."And the next iteration with up to four Bobcat cores isn't exactly what I'm looking for" BASED ON WHAT BENCHMARK!?? Good grief I'm not looking for AMD praise I'm looking for a benchmark that backs up claims like these.
whatthehey - Sunday, January 16, 2011 - link
Seriously? Is that all you've got? You people with your "Anandtech is all about Intel" are delusional. Let me try to help: "look at our nettop preview of Brazos". The article in question actually LINKED the article. Here it is again:http://www.anandtech.com/show/4023/the-brazos-perf...
PS CS4: Brazos is 9% faster than Atom.
x264 1st pass: 7% faster than Atom
x264 2nd pass: 5% faster
3dsmax: 26% faster
CB singlethreaded: 65% faster
CB multithreaded: 11% faster
Par2: 4% faster
Winrar: 6% faster
But those are all CPU intensive benchmarks! Well, yeah, that's the point of a CPU comparison. You don't benchmark a CPU by running a test that's bottlenecked by the hard drive. Hell, if Brazos is only 65% faster than Atom in the best-case result, I'm even more concerned than what Anandtech is voicing in this editorial piece! E-350 should be the lowest Brazos offering, but the C-30 and C-50 appear to be there to compete with netbooks for battery life, even if they may end up slower in everything outside of video decoding and GPU-intensive tasks.