The AMD Llano Notebook Review: Competing in the Mobile Market
by Jarred Walton & Anand Lal Shimpi on June 14, 2011 12:01 AM ESTWhat Took So Long?
AMD announced the acquisition of ATI in 2006. By 2007 AMD had a plan for CPU/GPU integration and it looked like this. The red blocks in the diagram below were GPUs, the green blocks were CPUs. Stage 1 was supposed to be dumb integration of the two (putting a CPU and GPU on the same die). The original plan called for AMD to release the first Fusion APU to come out sometime in 2008—2009. Of course that didn't happen.
Brazos, AMD's very first Fusion platform, came out in Q4 of last year. At best AMD was two years behind schedule, at worst three. So what happened?
AMD and ATI both knew that designing CPUs and GPUs were incredibly different. CPUs, at least for AMD back then, were built on a five year architecture cadence. Designers used tons of custom logic and hand layout in order to optimize for clock speed. In a general purpose microprocessor instruction latency is everything, so optimizing to lower latency wherever possible was top priority.
GPUs on the other hand come from a very different world. Drastically new architectures ship every two years, with major introductions made yearly. Very little custom logic is employed in GPU design by comparison; the architectures are highly synthesizable. Clock speed is important but it's not the end all be all. GPUs get their performance from being massively parallel, and you can always hide latency with a wide enough machine (and a parallel workload to take advantage of it).
The manufacturing strategy is also very different. Remember that at the time of the ATI acquisition, only ATI was a fabless semiconductor—AMD still owned its own fabs. ATI was used to building chips at TSMC, while AMD was fabbing everything in Dresden at what would eventually become GlobalFoundries. While the folks at GlobalFoundries have done their best to make their libraries portable for existing TSMC customers, it's not as simple as showing up with a chip design and having it work on the first go.
As much sense as AMD made when it talked about the acquisition, the two companies that came together in 2006 couldn't have been more different. The past five years have really been spent trying to make the two work together both as organizations as well as architectures.
The result really holds a lot of potential and hope for the new, unified AMD. The CPU folks learn from the GPU folks and vice versa. Let's start with APU refresh cycles. AMD CPU architectures were updated once every four or five years (K7 1999, K8 2003, K10 2007) while ATI GPUs received substantial updates yearly. The GPU folks won this battle as all AMD APUs are now built on a yearly cadence.
Chip design is also now more GPU inspired. With a yearly design cadence there's a greater focus on building easily synthesizable chips. Time to design and manufacture goes down, but so do maximum clock speeds. Given how important clock speed can be to the x86 side of the business, AMD is going to be taking more of a hybrid approach where some elements of APU designs are built the old GPU way while others use custom logic and more CPU-like layout flows.
The past few years have been very difficult for AMD but we're at the beginning of what may be a brand new company. Without the burden of expensive fabs and with the combined knowledge of two great chip companies, the new AMD has a chance but it also has a very long road ahead. Brazos was the first hint of success along that road and today we have the second. Her name is Llano.
Facebook
Twitter
RSS
177 Comments
View All Comments
whoaaaaaaaa - Tuesday, June 14, 2011 - link
A lot. I think you are very out of touch. No one buys desktops anymore, everyone uses a laptop - and quite a few people want to WOW or COD on them.luniq - Tuesday, June 14, 2011 - link
Agreed. A lot of people I know uses laptop although they rarely need to take it with them and they game too.swaaye - Tuesday, June 14, 2011 - link
I don't think he is. Most computer users don't play games. All they need is a GPU that can run the OS's UI.cotak - Tuesday, June 14, 2011 - link
The sort of people who use laptops and are pushing the CPUs hard are in the reality of the real world a limited bunch. Those who really need the power aren't going to be on a laptop anyhow like you say. Most office staff never use to the full potential of their issued computers even when they were using pentium 4s. And I actually think you'll find a lot of office workers are still humping around a P4 in their laptop bag.This is especially true as a lot of work now a days are done server side. About the only "office workers" who pound their CPU are those with big spread sheets. Even there I do not think you'll find many office workers who'd have an issue with most modern CPUs. They'll just go for a smoke or coffee break while it crunches.
And what review did you read that say there's no advantage in battery life? If you are pushing a game Llano lasts twice as long on the same battery.
voidi - Tuesday, June 14, 2011 - link
Lots of potential, they managed to find a valid niche in Intel's Netbook Hardware Portfolio. I think it's futile to argue about the success of this solution already, as the price will be the determining factor.As I see it, Llano has the potential to bring AMD back in the mobile market on a reasonable scale, but it also has the potential to utterly fail. Since it can't compete in the high-end area in either CPU or GPU performance, it will all depend on the value for your money. If the price is low enough, Llano will fill its niche quite comfortably.
Personally, I will wait for Trinity and Ivy Bridge, as both promise significant improvements and will likely be priced close to their predecessors.
jaydee - Tuesday, June 14, 2011 - link
On the one hand, I'm pleasantly surprised that AMD was able to bring the power draw of a K10/10.5 down to those levels, while adding decent GPU. It's a big step up for the to be in the discussion again for laptops. I don't think anyone could have reasonably expected them to go from Danube up to Intel's SB performance level overnight, so from that perspective, it's a big jump for them.On the other hand, it couldn't be more obvious that this product should have been competing against Arrendale about 18 months ago at the price point introduced today. I want to like this product badly, but it's so hard given it's CPU performance. Sure it's great for mid-range mobile gamers, but why would anyone else look at llano? Hopefully AMD can make just as big of a stride with BD-based mobile Fusion within the next 24 months, and then we can perhaps take AMD seriously in the bigger mobile market segments.
Gunbuster - Tuesday, June 14, 2011 - link
Will this mean the lowest common denominator HP $450 laptop of the week the everyone on a budget inevitably buys at retail when they are in a time crunch will suck less?I sure hope so.
Boissez - Tuesday, June 14, 2011 - link
I actually like the performance as it seems better balanced than in the dual-core Sandy bridges, ie., about 50% performancethe in single threads apps, 80% in multi-threaded apps and 200% in gaming. IOW better suited towards consumer needs. And battery life is good too.What AMD needs to work on though is lowering the TDP. A8-3500 performance in a 25W enveloppe would allow this to be a good alternative in the upcoming ultrabooks (especially because you can't fit a discrete GPU in those)
Boissez - Tuesday, June 14, 2011 - link
Oh yeah and make all of those APU's compatible with 1600 and 1866 Mhz RAM. RAM speed seems to be a major bottleneck in this platform.ET - Wednesday, June 15, 2011 - link
I think that even lower performance in 25W would be attractive, but it might eat into the E-350 market, unless it's priced significantly higher. I don't think it'd be a big problem for AMD to lower power to that point (certainly for A4), but they're probably waiting to see where the market goes with the current Llano APU's. I believe we'll see a 25W version later this year.