[FUDZ] The heart of AMD’s Llano is K8

[Idontcare]

The heart of AMD’s Llano is K8

The most shocking report about the second-to-come Fusion is that the CPU core behind it is the venerable K8.

http://www.fudzilla.com/processors/processors/the-heart-of-amds-llano-is-k8

Ouch, dems fighten words, I really can't believe this to be true. It makes no sense for AMD to invest resources to shrink the K8 core versus shrinking the K10 core.

 

[IntelUser2000]

Yea I think Fuad is just exaggerating. People don't think K10 is an advancement over the K8, even though per clock advancement was good. It's a somewhat modified K10.

 

[Vesku]

Yeah, judging by the 90% performance of current Athlon II mobile chips I would bet it's K10 derived. Basing this on the fact that clock rates are rumored to be similar to existing models, 1.5GHz range. Can't see a K8 power saving CPU getting that close in IPC.

 

[khon]

Llano isn't meant to be a powerful CPU anyway, so I don't see how it matters much.

Llano is the CPU you buy if you want a cheap desktop that's good enough for office programs, web surfing, videos and maybe the occasional light gaming.

It's the sort of thing very few people will put in a computer they build themselves, but lots of companies will put in their value desktops.

 

[khon]

Yeah, judging by the 90% performance of current Athlon II mobile chips I would bet it's K10 derived. Basing this on the fact that clock rates are rumored to be similar to existing models, 1.5GHz range. Can't see a K8 power saving CPU getting that close in IPC.

That's Bobcat, not Llano.

 

[Soleron]

If it's an unmodified K8 core as he claims, how could it have the power gating Anandtech already announced for it?

If you look at the die shot of Llano and compare it to Phenom II you can clearly see it's a modified Phenom II core. For the better - a ~5% IPC boost on top.

Fuad is just wrong. The misunderstanding comes from the fact that Phenom II is internally referred to as K8, same as Phenom I and Athlon 64. K9 was a cancelled design. K10 is Bulldozer. Another internal name for Phenom II's core is Greyhound+ .

@Vesku

Bobcat is the one with 90% of mainstream performance. Llano is much better.

 

[Idontcare]

Llano isn't meant to be a powerful CPU anyway, so I don't see how it matters much.

Llano is the CPU you buy if you want a cheap desktop that's good enough for office programs, web surfing, videos and maybe the occasional light gaming.

It's the sort of thing very few people will put in a computer they build themselves, but lots of companies will put in their value desktops.

what?

Llano will need to compete everywhere that SB goes in the mobile and lower mainstream desktop markets.

If it can't or doesn't then what is the purpose of the product?

To that end, it has to at least best the existing $100 Athlon X4 parts, I don't see that happening with a K8 core unless clockspeed are 4.5-5GHz.

 

[Tsavo]

There's no way that this thing is going to be based on K8. AMD wouldn't shoot itself in the feet with a move of this nature.

 

[khon]

what?

Llano will need to compete everywhere that SB goes in the mobile and lower mainstream desktop markets.

If it can't or doesn't then what is the purpose of the product?

To that end, it has to at least best the existing $100 Athlon X4 parts, I don't see that happening with a K8 core unless clockspeed are 4.5-5GHz.

Maybe I should be clearer.

I don't believe the K8 story at all. Llano will be K10 shrunk to 32nm, and with some power gating features added.

So yeah they will beat the $100 Athlon X4 parts, but that's all they'll beat. A Llano quad core won't come anywhere near a SB quad core, nor is it intended to. It's designed to be a low end product, used mostly in budget pre-built systems.

 

[Idontcare]

That's one hell of a crappy start for something that is supposed to represent the "future"...Intel showcases in their most advanced leading-edge processors and AMD just kinda tucks it into a market segment that represents the lowest margin sku's they can sell.

I'm at a bit of a loss for words here, trying to comprehend AMD's strategy on this.

"The future is fusion, if you don't believe us just buy an Intel system and see for yourself.

If you aren't overly interested in the future then hows about ya have some of our yummy bulldozer stuff?"

 

[Zstream]

That's one hell of a crappy start for something that is supposed to represent the "future"...Intel showcases in their most advanced leading-edge processors and AMD just kinda tucks it into a market segment that represents the lowest margin sku's they can sell.

I'm at a bit of a loss for words here, trying to comprehend AMD's strategy on this.

"The future is fusion, if you don't believe us just buy an Intel system and see for yourself.

If you aren't overly interested in the future then hows about ya have some of our yummy bulldozer stuff?"

You're a moderator, the link you posted has nothing in terms of information to backup his statement. Someone must be bored or something...

 

[sandorski]

Doesn't matter. If VIA took a modified Cyrix 120+ slapped a modified Virge DX on the same current sized Die and it competed against anything out there, what would it really matter?

 

[crazylocha]

Part of the problem is the fact that TSMC decided to skip a die shrink. Global foundries has 3 new plants due to be finished in next year. Once those are up and going, wave bye bye to TSMC and hopefully die shrinks/new cores should stay on release schedules. One of the main reasons for AMD going into the spin off of GBF was to prevent delays from third parties. Intel can decide to reramp any of their foundries on a whim.

For all that they have gone through behind the scenes, AMD has done a decent job staying almost with it. Looking forward to the new plants getting fired up in Germany, NY State, and finally in Saudi??(if remember right). If nothing else, the shrinks in GPU's should be very good. Can you imagine a 5970 reduced 2 sizes AND integrated together? 2 8pins and thermals of a 5770? kewlness soon to be abounding

 

[Idontcare]

You're a moderator, the link you posted has nothing in terms of information to backup his statement. Someone must be bored or something...

I'm a little daft too, you're gonna have to be a little more explicit if you want me to "get" whatever message it is you think you are sending...

 

[zokudu]

You're a moderator, the link you posted has nothing in terms of information to backup his statement. Someone must be bored or something...

• He is not a mod in CPUs and Overclocking
• He was made a mod to avoid people taking pot shots at another mod for their opinions
• This statment was completely innappropriate and unwarrented I think IDC deserves the respect he should recieve on these forums
  
  
  

 

[bryanW1995]

it comes with the territory. mod = authority figure = impolite attacks for no reason whatsoever in many cases.

Zstream, go over to video and pick a color.

 

[Daedalus685]

This was talked about on semiaccurate forums the other day (yesterday I suppose).

The issue seems to be that AMD often refers to the 10h series as code name K8 in addition to the 0Fh cores.. So pretty much everything since opteron is often referred to as K8 by their engineers.

I believe this stems from AMD giving up code numbers a while ago and never officially using them at all since Phenom came out.

Fuad heard that llano would use K8 which someone at AMD did say... but it refers to a propus+ core.. not an original athlon 64...

At any rate.. seems the fubar is all naming convention.. we can rest assured that Llano will be based on propus with power gating, and will likely still get fed to the wolves by sandy bridge if the anand preview numbers are close.

 

[Phynaz]

You're a moderator, the link you posted has nothing in terms of information to backup his statement. Someone must be bored or something...

While I think everybody around here knows I can't stand FUDz links, I think you need to go a little easier on IDC. Very few people around these boards are as reasonable as he is.

Also, a moderators job is moderate discussions. Are they also to be held to a different editorial standard?

 

[DrMrLordX]

Fuad heard that llano would use K8 which someone at AMD did say... but it refers to a propus+ core.. not an original athlon 64...

Ding ding ding, we have winnar. AMD has been pretty clear all along that Llano is an L3-less Stars quad + 480 stream processors. Power gating may or may not be a feature in all of them, but if it is, hey, good for Llano.

If there's anything that makes it run faster per clock than Propus (not taking the stream processors into account), that's news to me, though understandably that's good news.

 

[busydude]

Ding ding ding, we have winnar. AMD has been pretty clear all along that Llano is an L3-less Stars quad + 480 stream processors. Power gating may or may not be a feature in all of them, but if it is, hey, good for Llano. If there's anything that makes it run faster per clock than Propus (not taking the stream processors into account), that's news to me, though understandably that's good news.

Yeah, it is pretty much known that Llano is 'tweaked' Phenom II with powergating and turbo.

Bobcat- Netbooks/CULV's
Llano- Notebook's/Low end- Main Stream Desktop's(New Socket and not AM3+)
Bulldozer- Highend Desktops(AM3+ Socket)/ Servers

 

[DrMrLordX]

That's one hell of a crappy start for something that is supposed to represent the "future"...Intel showcases in their most advanced leading-edge processors and AMD just kinda tucks it into a market segment that represents the lowest margin sku's they can sell.

AMD has sort of been tucking products into various market segments since AM2 launched, though in some cases I think this has been due to the underperformance of some of their products since that time (Brisbane comes to mind). I think with the case of Llano, they want to get the chip into as many systems as possible as quickly while bribing developers to make use of OpenCL or DirectCompute or . . . something to modify their code so that as much fpu-intensive code as possible can be offloaded to the GPU element of the Llano APU.

Making a big deal about that now may not really be a part of their game plan. Making a big deal about what people can do with Llano/Ontario down the road once the code support is there, however . . . I dunno. Just speculating here. Obviously the code to take advantage of Llano isn't out there right now. Support for CPU + GPU based Fusion is almost non-existent at the present compared to, say CUDA-compiled apps. AMD may be waiting for developers to get on board before making much noise.

Fusion is not, and never has been about gaming: it's about moving most of the fpu functions off the general-purpose CPU and onto the specialized vector processors crammed onto the same die (or on a separate piece of silicon with its own HT link, or stuck on a PCI-e card, or . . .)

At face value, a tweaked Propus looks like it would get its arse handed to it by a Sandy Bridge product targeted at the same market segment, but a Propus with 480 stream processors doing some serious number crunching in relative harmony with the Propus cores is an entirely different ballgame. If Sandy Bridge's GPU is anything remotely like the one on Clarkdale, it will probably be revealed (at some point) that the Sandy Bridge GPU relies in no small part on the x86 cores for computing horsepower, making the Sandy Bridge GPU a poor choice for GPGPU operations.

If AMD can't get developers on board with fusion, then yeah, you've just got a Propus with a better IGP than the 890GX. Woop woop. Sandy Bridge will kill it in general computing power, and we'll see who wins the IGP battle. Maybe AMD keeps itself afloat in the same segment with better GPU horsepower, though for general purpose apps, without fusion-compliant code in the wild, that GPU horsepower might not matter.

As a side note, JF punted on a question I put to him awhile back with how APU-based Fusion is going to square with discrete graphics card-based Fusion, since right now, Fusion can either be an APU like Llano or it can be CPU + vid card working in tandem (think: Bulldozer + PCI-e implementation of Southern Islands).

If I recall correctly, the coders behind f@h commented, when developing their various GPU cores, that utilizing video cards for HPC was somewhat difficult since the turnaround time from PCI-e cards is so bad. It was like they had to code an entire module to load into the RAM on the vid card to minimize communications between the CPU and GPU - in other words, offload an algorithm and dataset to the GPU, let the GPU do all the crunching remotely, then have the GPU return the results as quickly as possible.

CPU + vid card Fusion is almost guaranteed to have the same problem since AMD has chosen not to leverage its HTX slot specification (up to this point) to facilitate superior communication between the CPU and GPU.

APU-based fusion suffers from a significant reduction in computational resource (read: fewer and slower stream processors) and no 1+ gb of memory dedicated to the stream processors involved, but since the stream processors are on the same die as the x86 cores, we can assume that the turnaround time of communications between the x86 cores and the stream processors would be much shorter than that of a CPU attempting to issue commands to a PCI-e-connected GPU. In fact, from recent uh . . . "discussions" of Barcelona, it looks like intra-die communications on K10/K10.5 chips happens about twice as fast as communications between devices connected via HT3.0.

Anybody coding to Llano will probably not have to worry about what is loaded into system memory and what is loaded into the segment of system memory cordoned off as frame buffer for the GPU. Anybody coding to Zambezi + Southern Islands will have to worry a great deal about what they keep in system memory and what they move into the vid card's dedicated memory.

Furthermore, an 80-90 cycle penalty to communicate with the stream processors still doesn't seem that awesome. I don't know if Llano will have shared l2 with the GPUs to solve that problem (I'm thinking not), but um, doesn't Bobcat have shared L2? And what if Bobcat shared its L2 with the stream processors that make up Ontario?

Doesn't that sort of play back into the idea Idontcare had about Bobcat saving the day for AMD sometime down the line . . . ?

 

[GaiaHunter]

Ding ding ding, we have winnar. AMD has been pretty clear all along that Llano is an L3-less Stars quad + 480 stream processors. Power gating may or may not be a feature in all of them, but if it is, hey, good for Llano.

Have you guys noticed these weird codenames in cat 10.8?

249,ONEGA (6750),NI TURKS
257,CASPIAN PRO (6770),NI CAICOS

Could these be the Llano GPUs?

 

[Soleron]

Have you guys noticed these weird codenames in cat 10.8?

249,ONEGA (6750),NI TURKS
257,CASPIAN PRO (6770),NI CAICOS

Could these be the Llano GPUs?

They are island codenames, and Llano's GPU is Evergreen, not SI. So probably not - good suggestion though. I think we'd see Bobcat before Llano given that we're 6+ months from the latter.

 

[GaiaHunter]

They are island codenames, and Llano's GPU is Evergreen, not SI. So probably not - good suggestion though. I think we'd see Bobcat before Llano given that we're 6+ months from the latter.

Yeah, they are island names - except they are named after russian lakes - onega river/lake, caspian sea.

So why the different code names for these 2 products?

Additionally, isn't the GPU part of CPU+GPU be quite easy to update/refresh/whatever?

 

[ViRGE]

Doesn't matter. If VIA took a modified Cyrix 120+ slapped a modified Virge DX on the same current sized Die and it competed against anything out there, what would it really matter?

I'd buy it in a heartbeat.