The AMD Llano Notebook Review: Competing in the Mobile Market
by Jarred Walton & Anand Lal Shimpi on June 14, 2011 12:01 AM EST
High Detail Gaming and Asymmetrical CrossFire Misfire
Update, 8/10/2011: Just to let you know, AMD managed to get me a new BIOS to address some of the rendering issues I experienced with CrossFire. As you'll read below, I had problems in several titles, and I still take exception with the "DX10/11 only" approach. I can name dozens of good games out there that are DX9-only that released in the past year. Anyway, the updated BIOS has at least addressed the rendering errors I noticed, so retail Asymmetrical CrossFire laptops should do better. With that disclaimer out of the way, here's my initial experience from two months back.
So far, the story for Llano and gaming has been quite good. The notebook we received comes with the 6620G fGPU along with a 6630M dGPU, though, and AMD has enabled Asymmetrical CrossFire...sort of. The results for ACF in 3DMarks were interesting if only academic, so now we're going to look at how Llano performs with ACF enabled and running at our High detail settings (using an external LCD).
Just a warning before we get to the charts: this is preproduction hardware, and AMD informed us (post-review) that they stopped worrying about fixing BIOS issues on this particular laptop because it isn't going to see production. AMD sent us an updated driver late last week that was supposed to address some of the CrossFire issues, but in our experience it didn’t help and actually hurt in a few titles. Given that the heart of the problem is in the current BIOS, that might also explain why Turbo Core doesn't seem to be working as well as we would expect.
AMD also notes that the current ACF implementation only works on DX10/11 games, and at present that's their plan going forwards as the majority of software vendors state they will be moving to DX10/11. While the future might be a DX10/11 world, the fact is that many recent titles are still DX9 only. Even at our "High" settings, five of our ten titles are tested in DX9 mode (DiRT 2, L4D2, Mafia II, Mass Effect 2, and StarCraft II—lots of twos in there, I know!), so they shouldn't show any improvement...and they don't. Of those five titles, four don't have any support for DX10/11 (DiRT 2 being the exception), and even very recent, high-profile games are still shipping in DX9 form (e.g. Crysis 2, though a DX11 patch is still in the works). Not showing an improvement is one thing, but as we'll see in a moment, enabling CrossFire mode actually reduces performance by 10-15% relative to the dGPU. That's the bad news. The good news is that the other half of the games show moderate performance increases over the dGPU.
If that doesn't make the situation patently clear, CrossFire on our test unit is largely not in what we consider a working state. With that out of the way, here are the results we did managed to cobble together:
Given this is preproduction hardware that won't see a store shelf, the above results are almost meaningless. If ACF can provide at least a 30% increase on average, like what we see in TWS2, it could be useful. If it can't do at least 30%, it seems like switchable graphics with an HD 6730M would be less problematic and provide better performance. The only takeaway we have right now is that ACF is largely not working on this particular unit. Shipping hardware and drivers should be better (they could hardly be worse), but let's just do a quick discussion of the results.
If we just look at games with DX10/11 enabled, the story isn't too bad. Not accounting for the rendering issues noted below, ACF is able to boost performance by an average of 24% over the dGPU at our High settings. We didn’t include the Low and Medium results for ACF on the previous page for what should be obvious reasons, but if the results at our High settings are less than stellar, Low and Medium settings are even less impressive. Trimming our list of titles to three games (we tested TWS2 and STALKER in DX9 mode at our Low and Medium settings), ACF manages to average a 1% performance increase over the dGPU at Low and a 14% increase at Medium, but Civ5 still had to contend with rendering errors and Metro 2033 showed reduced performance.
In terms of rendering quality, ACF is very buggy on the test system; the default BIOS settings initially resulted in corrupted output for most games and 3D apps, but even with the correct settings we still encountered plenty of rendering errors. Civilization V only had one GPU rendering everything properly while units were missing on the other GPU, so you’d get a flicker every other frame with units appearing/disappearing. At higher detail settings, the corruption was even more severe. STALKER: Call of Pripyat and Total War: Shogun 2 also had rendering errors/flickering at higher quality settings. Since we didn't enable DX10/11 until our High defaults, right when ACF is supposed to start helping is where we encountered rendering issues.
Just to be clear: none of this means that Asymmetrical CrossFire is a bad idea; it just needs a lot more work on the drivers and BIOS. If/when we get a retail notebook that includes Asymmetrical CrossFire support, we’ll be sure to revisit the topic. Why ACF isn’t supported in DX9 is still a looming question, and AMD’s drivers need a much better interface for managing switchable graphics profiles. A list of all supported games with a central location to change all the settings would be a huge step up from the current UI, and users need the ability to enable/disable CrossFire support on a per-game basis if AMD wants anyone to actually use ACF. We also hope AMD rethinks their “only for DX10/DX11 modes” stance; CrossFire has worked with numerous DX9 games in the past, and what we’d like to see is ACF with the same list of supported games as regular CrossFire. If nothing else, having ACF enabled shouldn't reduce performance in DX9 titles.
In summary: we don't know if ACF will really help that much. We tested Asymmetrical CrossFire on what is, at best, beta hardware and drivers, and it didn't work very well. We want it to work, and the potential is certainly there, but we'll need to wait for a better test platform. To be continued....
Facebook
Twitter
RSS
177 Comments
View All Comments
jollyjugg - Saturday, June 18, 2011 - link
Anand,you are making such a big deal of performance between Intel and AMD machines. Most of the folks who buy laptops are not looking to a super computer level performance which matters a lot in the server world or may be even desktop world. They are looking to buy a laptop which has good performace, good battery life and are more affordable. For the kind of applications most of the people use (internet surfing, listening to musing, watching youtube, watching movies, playing solitaire etc etc), thre will hardly be any difference between the offerings from both manufacturers. The fact that you are comparing performance and battery life and rather sneeringly say that if you want a machine which is 100-200 bucks lower then you should go for Llano machine at the cost of less performance makes me see black here. Why is cost not a big deal for you. How can you absolutely say that intel performance/$ and Battery life/$ is better than AMD's metrics. If not for AMD intel will be selling these machines not for 700 but for 1000. So in reality customers see reduced cost from Intel and a further 200 discount on AMD machines. You have to be a bit impartial in your reviews and not make only big deal of performance in portable machines where battery life and cost is also equally important and compare them as such. The fact that Intel's peformance is important win for them over AMD performance is a important one for geeks and enthusiasts like you, but not for comman man in the street. For him bang for the buck is the most important metric in most of the cases. But since your website is a reputed one, whatever you say might influence the opinion of man in the street. Please dont let your or any of your staff's personal opinion cloud their ability to see the bigger picture and tell it as such to PC buyers. Be impartial in your reviews.
A Humble AMD Fan
JarredWalton - Saturday, June 18, 2011 - link
We do our best to remain impartial; if we were partial, we would strongly advocate for or against AMD. When someone calling themselves a "humble AMD fan" talks about being impartial, you've pretty much already shot yourself in the foot. I'm not an Intel fan, despite what many would like to say. Right now, Intel simply has the better processor. AMD now has the better all-in-one design if you value graphics performance, but in order to give AMD the win you have to declare GPUs as being more vital than CPUs. Right now, outside of gaming, we're missing the killer apps to make that true.What we said here is that AMD is competitive with Llano, and they are, but the pricing is really the big question. We *guess* that you might get AMD laptops for $100 to $200 less than Intel, but that's being generous. AMD says the laptop used in this review should have a target price of around $800 (because it has the 6630M in addition to Llano). At that price, this is not a clear win for AMD--not even close. Yes, Intel would be priced differently if AMD didn't exist, but you can't judge the quality of a product by what would happen if it disappeared. "Duke Nukem Forever would be an amazing game if no other FPS titles had come out in the past 15 years!"
For $800, ASUS already has the U41JF with similar graphics performance and better battery life (mostly because of the larger battery, but that's still important). Besides the MSI CX640 (i3-2310M with GT 520M for $650), there are five other laptops at Newegg that have Optimus graphics and Arrandale CPUs for under $750 (and in terms of performance, Llano is still slower than Arrandale on the CPU side). For $900, you can get a Samsung laptop with i7-2630QM and Optimus GT 540M. Dell's XPS 15 can be had with an i5-2410M and Optimus GT 525M for $800. That's what AMD has to compete with, and right now every one of those is significantly faster than Llano, offers better graphics performance than the Llano IGP, and battery life is similar or slightly worse.
Last year when I ripped on AMD's laptops for having okay performance with horrible battery life, AMD fans tried to tell me battery life didn't matter. We were looking at 2.5 hours with Athlon II/Phenom II compared to 5+ hours for Arrandale, and it "didn't matter" because people just wanted what was cheaper. Now we're looking at 6-7 hours battery life for Intel compared to 7-8 hours for Llano, at a similar price (at least going by AMD's suggestion), and battery life has suddenly become a lot more important. I recognize hypocrisy when I see it....
In summary, once more: Llano is a good step forward, but it will really depend on pricing. It cannot compete at $800. Period. Core i5-2410M is already 30-50% faster on the CPU side, offers similar battery life, and can be had with an Optimus GPU for $800. If AMD can't beat the performance or battery life, the only thing they can do is reduce pricing, so A8-3500M will need to start closer to $600, not $700. The A8-3530XM is the fastest mobile Llano chip, and it's only clocked 27% higher than the A8-3500M, likely with a $50 price premium (at least), so we can't really take that as a potential win either. 3530MX for $700 on it's own (no dGPU) would be reasonable, though, which is why I say that 3500M needs to be at the $650 (or lower) price range.
There's still the matter of getting a good quality laptop, regardless of whether it's AMD or Intel based, and that has been a seriously weak area for inexpensive laptops. Personally, if I were going out and spending my own money on a laptop right now, I'd lean heavily towards business offerings (Dell Latitude, Lenovo ThinkPad, or HP ProBook/EliteBook), just because their keyboards and build quality are so much better. That means I would be paying $1000+ for even a moderate laptop, and at that price it's no surprise that all the business offerings use Intel's CPUs. You can go the other route and buy an okay $600 laptop today, and in 18 months you replace it with another $600 laptop; the only problem is you're stuck with the crappy keyboards if you do that.
If there's any bias in my above statements, please let me know where. About the strongest bias I express is for good build quality and keyboards -- chiclet need not apply. AMD or Intel really doesn't matter to me; the question is who can offer the more compelling package overall, and determining a winner there requires listing out all the various aspects and then making a personal decision. I won't say that a faster CPU is always superior, just as a faster APU/GPU isn't always superior. They're different is all, but looking at the entire market right now the CPU will win out for the majority of users. Remove all the teenagers and 20-somethings from the population, and I'd say gamers (like myself) are less than 10% of the notebook buying population. Even with the younger generation included, I'd still say only 20% of laptops purchased will ever run anything more complex than Facebook games.
mga318 - Sunday, June 26, 2011 - link
Well, we can look at prices now.HP is currently selling their 15 inch dv6z laptop with the same processor as this one for $659 without the additional graphics card. Which means that your statement of these laptops needing to be priced closer to $600 is right on the money for what we're getting. Likewise, HP's upgrade to AMD's fastest APU is precisely what you guessed $50. So that places the A8-3530XM right at $709 without an additional graphics card. HP doesn't say exactly what card they're offering for their $50 and $100 dollar upgrades, but they're both listed as having GDDR5 instead of your DDR graphics card here:
512MB GDDR5 Radeon(TM) HD Dual Graphics [HDMI,VGA]
+$50.00
1GB GDDR5 Radeon(TM) HD Dual Graphics [HDMI, VGA]
+$100.00
So now you're getting the premium processor and graphics card for $809 (assuming you stay with the other standard components, with are 6GB RAM (speed not listed) and a cheapo 500GB 5400 RPM hard drive.
What do you think, Jared, are those competitive enough prices?
mga318 - Sunday, June 26, 2011 - link
*which are (not "with are").choikwa - Wednesday, June 29, 2011 - link
512MB GDDR5 Radeon(TM) HD Dual Graphics [HDMI,VGA]+$50.00
[6400m]
1GB GDDR5 Radeon(TM) HD Dual Graphics [HDMI, VGA]
+$100.00
[6700m]
You can verify these as
choikwa - Wednesday, June 29, 2011 - link
6400m has 160 stream procs6700m has 480 stream procs
rick1725 - Thursday, June 30, 2011 - link
Is this place full of Intel fanboys or what? The e-350 is far superior to any Atom configuration available. This has been proved over and over again. Stop flamming and get you head out of Intels overpriced a**strawhat pirates - Wednesday, July 13, 2011 - link
baka...baka...bakaaaa!!! <--- japan language... mention it to you, rick1725!!intel only win on advertising alone ...
the rest? thumbs down for InteLosers!! #boooo ...
and intel is only a theory core ...*core fuck!! there is no definite proof .. :@
FORZA AMD!! AMD till die!!
tuRnitUpsuM - Monday, July 4, 2011 - link
Im typing this on a Samsung nf210 (Atom n550) and external monitor. 4 threads Gig ram ... fold it up throw it in a rugsack. Life is good. Only two things could possibly make it better.1) the above machine (in the header)
2) same form factor but Cortex A-15 chip running off ARMv7-R instructions.
first one to market gets the CASH!!!
technology is a beautiful thing.
SMSAssembly - Tuesday, July 12, 2011 - link
We have the Llano processors for sale,AM3400DDX43GX 4-Core 1.4GHz
AM3500DDX43GX 4-Core 1.5GHz
AM3530HLX43GX 4-Core 1.9GHz
EM3000DDX22GX 2-Core 1.8GHz
Contact Jeff at SMS Assembly for inquiries
Jeff@smsassembly.com