Richland & Kabini rumours

[Homeles]

...and you've just proved his point. Good job.

 

[Nemesis 1]

AMD hit its prelaunch targets on Bobcat pretty well, actually. Faster than Atom with better graphics than it, in a small, cheap to produce die- and it sold very well. Bodes well for the successor to Bobcat.

Which came first Atom or Bobcat . Hitting a target that isn't moving is easy . Again I don't know why intel committed to atom for 5 years in its present form .
I believe AMD could match performance but not a good thing if they are power hogs which by the info available they seem to be power hogs comparred to Intel as well as Arm options . The GPU part no longer matters thats over with haswell and haswell comes well before 22nm atom OoO .

 

[NTMBK]

6 days left . Lets see who shows their hand at CES.

This, I can get on board with.

Brazo did fine against a 5 year old product from intel . Intel was committed to the orginal atom for 5 years we all know this also . Why I don't know but they did fulfill that oblagation to whom ever it was made .

The Bonnell core was mostly unchanged since 2008, but the manufacturing process, chipset, graphics and uncore have all changed significantly. It's not the same chip it was- it's been shrunk to 32nm, gone from having a chipset which used more power than the processor itself to a fully integrated SoC, and gone from northbridge based Intel GMA graphics to on-die PowerVR graphics. The core count has doubled, and clock speeds have improved. This isn't the same Atom that launched in 2008.

 

[Nemesis 1]

...and you've just proved his point. Good job.

Really because you say so . Thats fine .If its all the same your reply said me to and nothing more. He proves my point every time he opens his mouth. Hereing about whats new from anyone is welcome news to me . But than comparring that unreleased product to existing products as long as its within their product range is fine . But comparring to someone elses old products when they also have new products coming out really is child like . wait for the products see the results crawl away licking wounds as always.

The Toms intel at 22nm article says 22nm atom 50 less power usage 60/70% better performance . It also says it will have a 4 core version . no hyper threading . So is the 60% more powerful just from 1 core ? I have a test system here But I can't say how it performs. its mind binding never the less . LOL!

 

[Homeles]

He proves my point every time he opens his mouth.

So do you.

 

[Nemesis 1]

The Bonnell core was mostly unchanged since 2008, but the manufacturing process, chipset, graphics and uncore have all changed significantly. It's not the same chip it was- it's been shrunk to 32nm, gone from having a chipset which used more power than the processor itself to a fully integrated SoC, and gone from northbridge based Intel GMA graphics to on-die PowerVR graphics. The core count has doubled, and clock speeds have improved. This isn't the same Atom that launched in 2008.

So whats this a revelation or something its still the same core and thats what were talking about . Its still inorder and the same core . Of course the process has changed intel was woking on the other things it needs befor the 22nm core Like SoC and soix.

 

[NTMBK]

But than comparring that unreleased product to existing products as long as its within their product range is fine . But comparring to someone elses old products when they also have new products coming out really is child like . wait for the products see the results crawl away licking wounds as always.

32nm Atom SoC is not an old product- it's newer than Bobcat. Clover Trail came out less than six months ago.

 

[Nemesis 1]

32nm Atom SoC is not an old product- it's newer than Bobcat. Clover Trail came out less than six months ago.

Its still the orginal core updated nothing more . The new core to be released in 2013 is all new design from ground up.

 

[NTMBK]

Its still the orginal core updated nothing more . The new core to be released in 2013 is all new design from ground up.

Yeah, the 2014 Atom is going to be a big jump. But the current Atom is still the latest and greatest Atom that Intel have brought out, and that is what is going to compete in the market against AMD. Just because Intel have neglected their CPU performance doesn't alter that.

 

[podspi]

Its still the orginal core updated nothing more . The new core to be released in 2013 is all new design from ground up.

Everything I've seen so far seems to indicate the Avoton (sp?) will be the only 22nm Atom released in 2013. Kind of a disappointment, but most OEMs haven't even managed to get their Clover Trail-based tablets out the door yet...

 

[SiliconWars]

Look I have watched your post every since conroy . No man has been more responsible for good people being banned because of your HYPE on AMD products . Every man woman and child here knows this . This is a good subject . But this constantly wanting to compare unreleased AMD products to existing Intel products is older than old . 6 days left . Lets see who shows their hand at CES. Brazo did fine against a 5 year old product from intel . Intel was committed to the orginal atom for 5 years we all know this also . Why I don't know but they did fulfill that oblagation to whom ever it was made . But now its a new product every year up to 14nm for sure. Intel says that the 14nm Atom will appear around the same time as broadwell . Unless intel has moved broadwell to 2015 . That means we will see 22nm atom and 14nm atom befor the end of 2014. So the Intel roadmaps are not saying this.

You are an absolute zoomer.

 

[mrmt]

Yeah, the 2014 Atom is going to be a big jump. But the current Atom is still the latest and greatest Atom that Intel have brought out, and that is what is going to compete in the market against AMD. Just because Intel have neglected their CPU performance doesn't alter that.

I don't think Kabini will have such an easy time as Brazos had.

First and foremost Kabini and Temash doesn't seem to have a TDP low enough to compete with Clover Trail or ARM A15, which means that AMD chips will be on tablets, but most likely niche tablets.

Kabini won't have an easy time on laptops too, as it will clash with IVB Celeron and Pentiums. The fact that AMD is redirecting part of their Brazos sales to desktops are an indicative of how tough are current conditions for AMD low-end chips, and while Brazos might improve the situation a bit, competing against Core and Trinity won't be easy.

But more important, AMD will be stuck on 28nm until at least 2015, when they will move to a half-baked 20nm process. This at the time when Intel will be on the second generation finfet at 14nm, and Atom will be on this process already.

We already saw how poor Trinity fared against Core, I don't see any reason to think that Kabini/Temash will fare any better when subjected to the same handicaps (Intel chips will have smaller die size, lower TDP and better CPU performance) regardless of how good AMD GPU is.

FWIW I think what will settle AMD's fate is how good Kabini can penetrate on the embedded market. If AMD can get good income streams there, Kabini will survive. If not it will be a matter of time until 14nm Celerons and 14nm Atoms wipe out their market share.

 

[Ventanni]

If not it will be a matter of time until 14nm Celerons and 14nm Atoms wipe out their market share.

If they're sold cheaply enough. Kabini's main advantage here will be that it's cheap.

 

[mrmt]

If they're sold cheaply enough. Kabini's main advantage here will be that it's cheap.

There is a limit on how low you can go with pricing, and AMD is already too close to this limit. They are already in the sub-40% territory, if they go to the sub-35% they are done for, unless we see a complete overhaul of their cost structure (mostly personnel).

Having this in mind, do you think AMD can get 35%+ margins competing against 14nm Core and 14nm Atom in 2014? I don't think they can.

 

[SiliconWars]

Kabini won't have an easy time on laptops too, as it will clash with IVB Celeron and Pentiums. The fact that AMD is redirecting part of their Brazos sales to desktops are an indicative of how tough are current conditions for AMD low-end chips, and while Brazos might improve the situation a bit, competing against Core and Trinity won't be easy.

I think you'll find it's the opposite case and intels IVB Celerons and Pentiums are going to be under threat from Kabini, which will have dramatically superior graphics and lower TDP.

But more important, AMD will be stuck on 28nm until at least 2015, when they will move to a half-baked 20nm process. This at the time when Intel will be on the second generation finfet at 14nm, and Atom will be on this process already.

So AMD will be on a "half baked" 20nm, but intels 14nm will be all singing, all dancing? Do you have anything to back this up?

I don't see any reason to think that Kabini/Temash will fare any better when subjected to the same handicaps (Intel chips will have smaller die size, lower TDP and better CPU performance) regardless of how good AMD GPU is.

Smaller die size and lower TDP? Kabini will not come in far above 80mm2, and will certainly have far superior battery life to any celeron or pentium class cpu, as well as overwhelmingly superior graphics.

Intel can take their pick on how to compete vs Kabini, but none of them are particularly good propositions.

ULV's are too rare and expensive to be competition.
The current Atoms are a joke and even if Silvermont is better the graphics will still be far behind Kabini.
Celerons can't compete on graphics or TDP.

 

[Homeles]

Intel's processes are always "all singing, all dancing," or at least in recent history.

TSMC had issues with 40nm, issues with 28nm supply. GloFo had a terrible 32nm process, and it's probably still not very good in a relative sense. AMD's 65nm process was part of why the original Phenom was garbage. So to have faith in Intel and have a lack of faith in the other foundries only makes sense given the historical precedent.

I haven't seen anything suggesting that GloFo's 20nm process will be "half-baked" or any other negative adjective that you want to apply to it, but it's safe to assume that it will be inferior to Intel's respective process.

 

[mrmt]

I think you'll find it's the opposite case and intels IVB Celerons and Pentiums are going to be under threat from Kabini, which will have dramatically superior graphics and lower TDP.

I don't think you really noticed but Haswell will bring the bar down to 10W, Broadwell should bring even more.

As for the dramatically superior graphics, I'll wait for the benches. IVB graphics are already beating Trinity in some cases. I don't think Haswell will have such a huge deficit compared to GNC, much less Broadwell IGPU.

So AMD will be on a "half baked" 20nm, but intels 14nm will be all singing, all dancing? Do you have anything to back this up?

GLF 20nm was practically canned, instead GLF is taking some elements of a 14nm process and mixing it with some elements of their 20nm process. Call it what you will, but this is not a process in the same league as Intel's.

Smaller die size and lower TDP? Kabini will not come in far above 80mm2, and will certainly have far superior battery life to any celeron or pentium class cpu, as well as overwhelmingly superior graphics.

Kabini will go up in the power ladder until 25W, IVB and HSW will go down until 10W. What Kabini and what IVB are you comparing? Or a 10W HSW/BRW will consume more power than a 25W Kabini?

And die size, how big do you think a 14nm Broadwell will be? IVB is already at ~100mm^2, don't think that sizes around 85-90 are far fetched. And this is just the top, by Q413/Q114 we'll have 22nm Atom at the bottom.

All in all I just don't expect Kabini to have the same success Brazos had, but I don't think Kabini will buckle until 14nm Atom and Broadwell. Where I have great hopes for Kabini is in the embedded market. AMD can make great strides there.

 

[Homeles]

Yeah, I'm definitely not seeing any reason to be parading superior graphics in Kabini. There's no information on CU counts yet.

 

[SiliconWars]

Can you at least decide what chip is supposed to be Kabini's competition then? So far it seems to be up against Atoms TDP, Haswell's cpu and graphics and Celeron's cost. Now THAT would be a chip for sure, however I get the feeling the actual chip that Kabini finds itself competing against won't be quite so balanced.

 

[mikk]

I thought Kabini has 128 SPs, frequency unknown though.

 

[mrmt]

Can you at least decide what chip is supposed to be Kabini's competition then? So far it seems to be up against Atoms TDP, Haswell's cpu and graphics and Celeron's cost. Now THAT would be a chip for sure, however I get the feeling the actual chip that Kabini finds itself competing against won't be quite so balanced.

Today Brazos isn't really Atom competitor only. Brazos also competes with Celeron/Pentium in the bottom market in notebooks, and more recently, desktops (it sells 10x more than 8C Bulldozer). So Kabini as Brazos successor will compete against Core and Atom.

 

[zlatan]

I don't think you really noticed but Haswell will bring the bar down to 10W, Broadwell should bring even more.

Haswell and Broadwell will be mainstream platforms. No chance to sell these around $50-$70. AMD will also offer mainstream product with 10-15 watt TDP. Kabini is a low power product, designed for low price, and for a small motherboard.

As for the dramatically superior graphics, I'll wait for the benches. IVB graphics are already beating Trinity in some cases. I don't think Haswell will have such a huge deficit compared to GNC, much less Broadwell IGPU.

Ivy graphics is much slower than the Trinity graphics. It may faster at low resolution and quality, but when you set the bar to medium or high than the game is still playable at Trinity but not at Ivy.
GCN is a radical architectural change compared to traditional GPUs like generation 7.x architecture from Intel. Larabee is the easiest way to understand the differences. AMD just designed a working Larrabee. GCN is actually a co-processor for x64 CPU cores.

Kabini will go up in the power ladder until 25W, IVB and HSW will go down until 10W. What Kabini and what IVB are you comparing? Or a 10W HSW/BRW will consume more power than a 25W Kabini?

There is no point to compare Kabini to Ivy Bridge or any mainstream platform. Creating a mobile product is not only depends from the TDP. You must consider the motherboard size. This is relatively large with a mainstream platform, and very small with a low power product like Kabini. With a small motherboard you can use a large accumulator, so the battery life will be perfect. With a mainstream platform this will be much lower.

All in all I just don't expect Kabini to have the same success Brazos had, but I don't think Kabini will buckle until 14nm Atom and Broadwell. Where I have great hopes for Kabini is in the embedded market. AMD can make great strides there.

The Bay Trail Atom won't be a fast product. The CPU will be good, but the iGPU will just comparable with Brazos, or slower in some perspective.

 

[Nemesis 1]

Yeah, the 2014 Atom is going to be a big jump. But the current Atom is still the latest and greatest Atom that Intel have brought out, and that is what is going to compete in the market against AMD. Just because Intel have neglected their CPU performance doesn't alter that.

Which atom is that the 22nm process or the 14nm process. Your fond of saying 2014 / Put up or shout up make a wager . I will bet that 22nm atom is in phones and on the market in 2013. You name the stakes

 

[inf64]

I thought Kabini has 128 SPs, frequency unknown though.

You are probably correct

Kabini can be coupled with discrete HD 8000 cards
...
The fastest of them all is Mars 128 LP, PRO or XT that comes with HD 8700M series. AMD now calls it Crossfire dual graphics, and I guess that Crossfire brand remains reserved for two discrete graphics cards coupled together.
...

If it matches up with 128SP discrete part next year then it will probably be close to this spec GPU-wise. Can we now say ~40-50% faster than Brazos? It sure looks like that

 

[Nemesis 1]

The Bay Trail Atom won't be a fast product. The CPU will be good, but the iGPU will just comparable with Brazos, or slower in some perspective.

Again you don't know what your talking about . The Atom phone chip will use Imagination gpu. The tablet chip will use Intels HD4000 graphics and were are talking below 8 watts chips . So you best take the cheer leading skirt off and try doing some real research.