AMD CPU speculation... and expert conjecture

[mayankleoboy1]

A look at Intel’s Latest PC Game Developer Relations Strategy
http://semiaccurate.com/2013/04/03/a-look-at-intels-latest-pc-game-developer-relations-strategy/
more wins(game devs) coming intel's way. looks like amd will have to make do with the console wins....

Err no. Just the complete opposite. Most game devs were sneering at Intel's :
1. Hardware
2. Software

 

[sarinaide]

heh heh
http://www.tomshardware.com/news/APU-Bitter-Neal-Robison-Tony-Tamasi-PlayStation,21791.html

Nvidia is Jelly they dont have boobies doing their promo's. That article was tl;dr busy staring.

 

[-Fran-]

heh heh
http://www.tomshardware.com/news/APU-Bitter-Neal-Robison-Tony-Tamasi-PlayStation,21791.html

Nvidia is Jelly they dont have boobies doing their promo's. That article was tl;dr busy staring.

What? There was an article around that picture? I didn't see anything written in there! hahaha

Anyway, don't disregard Intel/AMD biased articles, you guys just need to cut through the BS in them and get the substance.

Cheers!

 

[sarinaide]

http://semiaccurate.com/2013/04/03/a-look-at-intels-latest-pc-game-developer-relations-strategy/#.UVxB7VdBkuc

In short, nobody cares about Intel HD there are cheaper an way better options. I did say no developer takes intel HD seriously, case in point.

 

[mayankleoboy1]

http://www.phoronix.com/scan.php?page=news_item&px=MTM0MTY

Intel releases "Running Average Power Limit" control for their CPU's.


i have poor memory, but does anyone remember before Ivy Bridge launch, Intel claimed that their "mobile chips have two "power planes". And the OEM could choose on which power plane the chip ran.Sort of like a configurable TDP. So the same chip could run at 15W and could run at 30W"
Was this marketing term for TurboBoost, or was it a real tech?

 

[de5_Roy]

http://www.phoronix.com/scan.php?page=news_item&px=MTM0MTY

Intel releases "Running Average Power Limit" control for their CPU's.


i have poor memory, but does anyone remember before Ivy Bridge launch, Intel claimed that their "mobile chips have two "power planes". And the OEM could choose on which power plane the chip ran.Sort of like a configurable TDP. So the same chip could run at 15W and could run at 30W"
Was this marketing term for TurboBoost, or was it a real tech?

that was for mobile cpus iirc.

 

[mayankleoboy1]

^ was it implemenmted by OEM's ? Never heard much about that after the IB launch.

 

[sarinaide]

http://www.phoronix.com/scan.php?page=news_item&px=MTM0MTY

Intel releases "Running Average Power Limit" control for their CPU's.


i have poor memory, but does anyone remember before Ivy Bridge launch, Intel claimed that their "mobile chips have two "power planes". And the OEM could choose on which power plane the chip ran.Sort of like a configurable TDP. So the same chip could run at 15W and could run at 30W"
Was this marketing term for TurboBoost, or was it a real tech?

Sounds similar to what AMD are using now going forward with Richland. Using AMD Power technology I can set states for my APU to run different profile speeds depending on the load. Can have it run at 1.5ghz when im doing nothing but surfing. Which is why im not overly concerned about power numbers as its very subjective and circumstantial.

 

[noob2222]

The xbitlabs article says that Steamroller single core perf improvement is expected to be 5-10% . Which is not bad. The only problem i see is that AMD cant increase the default clock speeds across the whole product line, which they did with Piledriver.

So Steamroller could be 'only' 10% faster average than Piledriver. Compared to 15% Piledriver improvements over Bulldozer.

you now have no room to try and accuse anyone of not reading.

AMD also advanced single-core execution by implementing 5%-10% more efficient scheduling, incorporated higher-capacity register files and performed some other tweaks.

This does not account for any of the architecture changes, this is simply a refined scheduler.

With steamroller, AMD not only incorporated two decoders per module, but also increased instruction cache size (to lower i-cache misses by 30%), enhanced instruction pre-fetch (the number of mis-predicted branches is down by 20% compared to Bulldozer ) as well as improved max-width dispatches per thread by 25%. AMD believes that Steamroller will provide 30% improvement in ops per cycle.

5-10% scheduler improvement + 30% improvement in ops per cycle = ? single threaded IPC... Its more than 5-10%.

A look at Intel’s Latest PC Game Developer Relations Strategy
http://semiaccurate.com/2013/04/03/a-look-at-intels-latest-pc-game-developer-relations-strategy/
more wins(game devs) coming intel's way. looks like amd will have to make do with the console wins....

reading part of that article .. LOL ..

#1 Through feb 2013, Intel HD Graphics 3000 has been #1 graphics device on Steam Survey for 8 straight months
#14 >50M Steam users times 13% using Intel HD Graphics = 7,000,000 Steam Intel HD Graphics users

Nice twist of actual facts. How can you have the #1 device with only 13% of the total steam market?

Its simple, because of the variants in the rest of the pack. yes, its #1 for just that reason there is no hd3300, hd3100, but only 13% of all users are using ANY intel graphics. The rest are Nvidia or ATI, for example radeon 5xxx, 6xxx, 7xxx, ect, with each one being split again into 7970, 7950, 7890, 7770, ect... This makes it easy to see how the HD 3000 is the #1 single device used, but overall its not even used that much compared to the rest.

7,000,000 users sure, what about the other 43,000,000?

Aside from that, the article primarily focused on the graphics end.

the summary fits very well.

This list is a joke, and this move is really pathetic. [Charlie's note 5: Not as much of a joke as the "journalists" who repeated it breathlessly without asking any devs what they thought.]S|A

 

[griptwister]

"Three of top six graphics products used by game players on Steam are Intel HD Graphics"

-_-' Most impressive Intel. Had no idea! hahaha! Only reason as to why it's in the top 3 is because those people can't afford graphic cards, or are on lap tops. Not to mention... Who would be the other 3 out of the top 6?

 

[noob2222]

Spoiler

IPC = Instructions Per Cycle

All that would really mean is that the architecture's capability is going to be utilized efficiently. It does technically possess 150-175% of the horsepower of the intel chip...if it was optimized efficiently internally, it would be a more effective architecture than what intel currently offers.

yes yes... but IPC for x86 is 1 to 1.1 or up to 1.2 "instructions per clock" for a little more optimized code. "full fury" optimized code could be much more, yet staying clearly below the 2 instructions per clock per thread [makes one wonder why have CPU cores with more than 2 ALU pipes ... right ?] . And NO ONE in current software offerings, that everybody uses, has "full fury"... not even benchmarks that can be more optimized for a particular uarch than another...

[EDIT: of course in this we are talking about NON VECTOR integer code, which is the traditional and logical way of counting IPC, since architecturally x86 is NOT a "vector" uarch... though good vectorized code could in many cases pass the 2 IPC barrier... ]

Since IPC is talked so much this days, it would be nice to have a "benchmark" that would have code specifically to indicate those numbers ( 1 to 1.2 or so IPC)...so the results would be 0.x to 1.z, or something like that.

Performance depends on so much factors that is hard to beggin with, but with those numbers in mind it would be easy to understand that on x86, no CPU today is "technically or potentially", anything like 150 to 170% IPC wise more performant than another. Not even Haswell compared with Jaguar, that is, 0.9 compared with 1.2 is around ~35% in "potential"... its much more than that, by many other factors, including instructions in the super bloated pointless x86 logical extensions ( like in vector from SSE to AVX), that one chip have natively, while in another must be executed by "microcode" (painful slow)... which might indicate that the "spree to bloat" to newer instruction extensions was/is only to gain "competitive advantage", accounting with this factor( seems obvious no ?)...

High degrees of advantage/speedup only comes with high parallel data code and high parallel execution models (cache and memory and clocks and even logical instructions could never reach that >100% speedups) ... but in here we are NOT talking about CPU but more like GPGPU models...

That is where AMD has an advantage IMO, and it spells APU not CPU. But the apps for this are yet to few and too far in between to make a decisive advantage... "never settle programs" for Multimedia apps could be the salvation of AMD... but doubt those kind will appear as "common benchmark" apps anytime soon...

Well, if you have 4 cores that can run 32 IPC versus 8 cores that can run 44 IPC, I would think, even with less efficiency of code processing, you'll flat out get more operations out of the AMD chip, versus the intel.

http://www.agner.org/optimize/blog/read.php?i=49

Scroll down to "still no library that is optimal on all processors".
This is a very interesting test. this is all tested with the AMD BD cpu used, and forcing the compiler to use certain extensions for the code and run its base optimization.

The most prominent is the 32 bit mode results. The default code for 4 vecors produced a time of 360(im assuming seconds), while forcing the cpu to use AVX produced it in 99 seconds. 64 bit mode isn't as prominent, but its still ~10-20%

Now picture this "benchmark" done on the generic compiler since you won't be able to run the forced AVX mode on any cpu older than bulldozer/sandy bridge.

for intel, if it has AVX, it runs AVX code. For AMD if it has AVX, it runs SSE2 code because its not GenuineIntel. This is where 4 cores at 32ipc vs 8 cores at 44 ipc fails. no matter what, the Intel cpu will be faster unless AMD is 3x as fast as any Intel cpu.

Case in point, Cinebench is compiled on ICC. AMD gets sse2 code while Intel SB and up can run AVX. 32 Intel IPC will beat 44 AMD IPC

 

[8350rocks]

Spoiler

IPC = Instructions Per Cycle

All that would really mean is that the architecture's capability is going to be utilized efficiently. It does technically possess 150-175% of the horsepower of the intel chip...if it was optimized efficiently internally, it would be a more effective architecture than what intel currently offers.

yes yes... but IPC for x86 is 1 to 1.1 or up to 1.2 "instructions per clock" for a little more optimized code. "full fury" optimized code could be much more, yet staying clearly below the 2 instructions per clock per thread [makes one wonder why have CPU cores with more than 2 ALU pipes ... right ?] . And NO ONE in current software offerings, that everybody uses, has "full fury"... not even benchmarks that can be more optimized for a particular uarch than another...

[EDIT: of course in this we are talking about NON VECTOR integer code, which is the traditional and logical way of counting IPC, since architecturally x86 is NOT a "vector" uarch... though good vectorized code could in many cases pass the 2 IPC barrier... ]

Since IPC is talked so much this days, it would be nice to have a "benchmark" that would have code specifically to indicate those numbers ( 1 to 1.2 or so IPC)...so the results would be 0.x to 1.z, or something like that.

Performance depends on so much factors that is hard to beggin with, but with those numbers in mind it would be easy to understand that on x86, no CPU today is "technically or potentially", anything like 150 to 170% IPC wise more performant than another. Not even Haswell compared with Jaguar, that is, 0.9 compared with 1.2 is around ~35% in "potential"... its much more than that, by many other factors, including instructions in the super bloated pointless x86 logical extensions ( like in vector from SSE to AVX), that one chip have natively, while in another must be executed by "microcode" (painful slow)... which might indicate that the "spree to bloat" to newer instruction extensions was/is only to gain "competitive advantage", accounting with this factor( seems obvious no ?)...

High degrees of advantage/speedup only comes with high parallel data code and high parallel execution models (cache and memory and clocks and even logical instructions could never reach that >100% speedups) ... but in here we are NOT talking about CPU but more like GPGPU models...

That is where AMD has an advantage IMO, and it spells APU not CPU. But the apps for this are yet to few and too far in between to make a decisive advantage... "never settle programs" for Multimedia apps could be the salvation of AMD... but doubt those kind will appear as "common benchmark" apps anytime soon...

Well, if you have 4 cores that can run 32 IPC versus 8 cores that can run 44 IPC, I would think, even with less efficiency of code processing, you'll flat out get more operations out of the AMD chip, versus the intel.

http://www.agner.org/optimize/blog/read.php?i=49

Scroll down to "still no library that is optimal on all processors".
This is a very interesting test. this is all tested with the AMD BD cpu used, and forcing the compiler to use certain extensions for the code and run its base optimization.

The most prominent is the 32 bit mode results. The default code for 4 vecors produced a time of 360(im assuming seconds), while forcing the cpu to use AVX produced it in 99 seconds. 64 bit mode isn't as prominent, but its still ~10-20%

Now picture this "benchmark" done on the generic compiler since you won't be able to run the forced AVX mode on any cpu older than bulldozer/sandy bridge.

for intel, if it has AVX, it runs AVX code. For AMD if it has AVX, it runs SSE2 code because its not GenuineIntel. This is where 4 cores at 32ipc vs 8 cores at 44 ipc fails. no matter what, the Intel cpu will be faster unless AMD is 3x as fast as any Intel cpu.

Case in point, Cinebench is compiled on ICC. AMD gets sse2 code while Intel SB and up can run AVX. 32 Intel IPC will beat 44 AMD IPC

Yes, I am aware of the intel code compiler issues and the "Not intel" flag. The sad thing is, the AMD chips still perform relatively well considering the inefficient code...lol. Cinebench of 6.9 versus 7.5 is not that drastic a difference...for comparable chips...but if the code was all the same, then AMD would likely win hands down.

 

[jdwii]

I can already read the Haswell reviews :

"One year after, and the CPU perf is incremental at best. The iGPU, much improved, still has lots of room to grow. It still cant compete with Trinity APU's from AMD. If you have to build a new system, and are going to add adGPU, we wholeheartedly recommend the 4770K. If you want to build a "good enough" PC, we think that a Richland APU system would be better. "

You are giving reviewers too much unbias. They are going to read. OMG AMAZING INTEL CPU, LOOK AT THAT GPU PERFORMANCE COMPARED TO BEFORE AND IF YOU WANT TO DO WORK, CPU PERFORMANCE IS SO MUCH HIGHER THAN AMD ON ITUNES.

And Richland review will be: 6/10 "its ok" - Anandtech.

Kabini successor has Jaguar+ cores
http://www.fudzilla.com/home/item/30959-kabini-successor-has-jaguar%20-cores

A look at Intel’s Latest PC Game Developer Relations Strategy
http://semiaccurate.com/2013/04/03/a-look-at-intels-latest-pc-game-developer-relations-strategy/
more wins(game devs) coming intel's way. looks like amd will have to make do with the console wins....

heh heh
http://www.tomshardware.com/news/APU-Bitter-Neal-Robison-Tony-Tamasi-PlayStation,21791.html

Yeah that Intel article made me laugh pretty hard i like how they act they are so successful for shipping so many Intel graphics out and to tell you the truth it made me a little sad to see how many people on steam use Intel graphics.

 

[anxiousinfusion]

Let's talk iGPU naming conventions. Given;

Gen 1 = 6550D
Gen 2 = 7660D
Gen 2.1 = 8670D

Is if fair to assume that

Gen 3 = 8770D?

Also engineering sample benches are claiming ~31% iGPU performance in Richland over Trinity: http://wccftech.com/amd-a10-6700-gpu-performance-exposed-80-faster-core-i53570k-20-a105800k-3dmark-11-fire-strike/.

 

[palladin9479]

"Three of top six graphics products used by game players on Steam are Intel HD Graphics"

-_-' Most impressive Intel. Had no idea! hahaha! Only reason as to why it's in the top 3 is because those people can't afford graphic cards, or are on lap tops. Not to mention... Who would be the other 3 out of the top 6?

The only reason they came out like that is because user's can't remove the Intel HD graphics from the CPU. You could use an i3 with dual 680's (not saying it's wise to do so) and they could still say "your using Intel HD3000!".

Brilliant marketing strategy, no matter what happens they instantly win because they ship more CPU's. Anyone who has a smidgeon of PC knowledge can immediately smell the BS, it's that strong.

 

[mayankleoboy1]

^ But really, most people dont even have a smidgen of computer knowledge.
And some people do have knowledge, but dont use it : Like the "Internet echo chamber reporter".

 

[anxiousinfusion]

^ But really, most people dont even have a smidgen of computer knowledge.
And some people do have knowledge, but dont use it : Like the "Internet echo chamber reporter".

I just sold a machine last week and the lady kept referring to it as a "CPU". The people with this level of computer knowledge are probably the same ones trying to game on HD3000.

 

[sarinaide]

Let's talk iGPU naming conventions. Given;

Gen 1 = 6550D
Gen 2 = 7660D
Gen 2.1 = 8670D

Is if fair to assume that

Gen 3 = 8770D?

Also engineering sample benches are claiming ~31% iGPU performance in Richland over Trinity: http://wccftech.com/amd-a10-6700-gpu-performance-exposed-80-faster-core-i53570k-20-a105800k-3dmark-11-fire-strike/.

There is just so much ermagherd, Haswell has been relegated to "hasbeen" iGPU war is GAME, SET and MATCH, if final numbers corrolate then as far as intel developing a iGPU that can remotely keep up is all but broken. If Haswell is roughly 30-35% slower than the Devastator, Richland will widen that gap substantially and if you factor in Kaveri around the corner, by the time broadwell is out in 2015 AMD will be onto excavator and I also have reservations about Broadwell matching Trinity, by that point Intel may be about 5 years at least behind. This is why there are escalating stories of Intel and Nvidia partnerships, though it will end bad and profit sharing never works, nor are Nvidia in a position of needing to sellout. Intel will be on there own, I think the boys in team blue are now hoping and praying that SR is not a x86 beast or there is going to be a lot of "serious problems".

 

[blackkstar]

Intel and AMD have to know something that the rest of us don't, because they're both opening their flood gates of AMD FUD.

All I see is AMD winning hard with Gaming Evolved and GTX 600 series falling way behind cheaper Radeons and Nvidia spewing FUD.

All I see is a horrible 3-7% increase in CPU with Haswell and rumors of broken power and broken USB 3.0 and Intel spewing FUD.

I never thought I would say this, but out of Nvidia, Intel, and AMD, AMD is the only one that's executing well right now. That's the most messed up thing I think I've ever written.

 

[esrever]

Intel is executing well, its just it has shifted its focus to more profitable fields. Nvidia trying to do the same. AMD is left with the scraps since they can't afford to do the same as effectively.

 

[sarinaide]

Intel is executing well, its just it has shifted its focus to more profitable fields. Nvidia trying to do the same. AMD is left with the scraps since they can't afford to do the same as effectively.

Exactly how well is that going, Intel are no nearer to competing in smartphones and their 22nm process still has issues, there is also no evidence of Intel effeciently suiting the Smartphone market which is basically run rampant by Samsung, ARM, Qualcom, even Nvidia are struggling to make an impact with Tegra being a battery drainer extraodinaire. If shifting focus means substandard bread and butter and the shifted focus doesnt work out like planned, what then. There are still facts being overlooked that a very big market still remains desktop and server markets, while sales are down that is probably down to economics, people not needing to upgrade from sandy to ivy and AMD products not being sold as counter products, it is not indicative of DT dying which is actually quite a laughable claim amidst developers still prefering the platform and more and more new developments on that front, GPU's evolving at a rate of knots and of course despite dear Intel's absurd claim DT and mobilty parts are flooding the market, again the slowdown is probably down to to much to fast and intel not giving people reason to upgrade.

The next person to tell me that smartphones are replacing desktops or mobility parts, I literally will rape an angel in pure rage quitting fury, I want that person to type me a legal brief on there little smartphones within time is of the essence, have it printed and given to me faster and without mistake like what a notebook or desktop will do. The only thing a smartphone does that is useful is email on the fly, gaming wise, i have a HTC one X with a quadcore tegra which lasts about 45minutes before its battery is dead so a PS vita is still a better portable gamer and notebooks are better gaming parts than a smartphone, in general usage the desktop and notebooks make a smartphone look completely and utterly ineffective. Tablets are useful but still people rather spend on notebooks.

This is a pure case of intel finding a market that has money thinking they are just going to waltz in and take it, again nothing suggests x86 will be an efficient median and honestly who the fuck needs it on a phone which by essence of the word phone implies phoning someone to talk to them, this is not new and is very old technology now and the premise of a phone will not change. The other aspect is the reason Samsung and co are doing better than intel is due to them focusing on their bread and butter, while Intel has not given a sandy owner the need to upgrade hence why the selling less. Intel need to wake the fuck up and smell the coffee, if AMD does catch up clearly they have the most interesting marketable product right now that dearly beloved team blue and team green have nothing to replicate.

 

[JAYDEEJOHN]

Desk Top is declining tho, not ending.
Like Ive said before, AMD has to weather this economic storm, but the impacts in times like these are when sometimes the giants struggle mightily and quickly, as economies shift.
The business models that made them who they are is simply not sustainable.
Im not saying Intel will crash, but Intel has many more problems than AMD does.

The difference is this
If Intel doesnt make it into the SFF as fluently as in the past, this will hurt them.
If they try and just ride out the DT/mobile market, they cant squeeze anymore than they currently have there, and again, their business model isnt one of competition, but leading, and while cpus arent their problem, getting that blend of cpu/gpu right could be.
Tough times for all, including Intel
If only they had that 2 billion back from LRB.......

 

[sarinaide]

Desk Top is declining tho, not ending.
Like Ive said before, AMD has to weather this economic storm, but the impacts in times like these are when sometimes the giants struggle mightily and quickly, as economies shift.
The business models that made them who they are is simply not sustainable.
Im not saying Intel will crash, but Intel has many more problems than AMD does.

The difference is this
If Intel doesnt make it into the SFF as fluently as in the past, this will hurt them.
If they try and just ride out the DT/mobile market, they cant squeeze anymore than they currently have there, and again, their business model isnt one of competition, but leading, and while cpus arent their problem, getting that blend of cpu/gpu right could be.
Tough times for all, including Intel
If only they had that 2 billion back from LRB.......

Part 1:

Very correct, Intel are stumbling upon a big mountain of strife and toil, there entire operation is geared towards the saturated hand held market, as above this has led them to neglect its bread and butter. Intel sales have slowed and hence DT because Intel has given nothing to need buying and for some odd reason AMD processors are not deemed good enough.

The problem with Intel's drive is that they will be the small fish in the big pond in the now saturated handheld market. The can offer to provide chips to Apple but apple will sell them as a product for more than they paid hence make all the profits, or they will just use more efficient sources like qualcom or ARM to maximise profits instead of feeding Intel cash cow. If that drive fails it will be enough to sink a monalith the size of intel in the blink of an eye. If Intel cannot make a x86 processor better than efficient x64 processors and they suck juice delivering worse performance well that will just go balls up. I don't like that direction and it means that we can be used to seeing sub standard x86 processors from intel in DT trim.

Mobility market is a more realistic one for me and Intel is good right now in that front but AMD may start making splashes knowing they have something Intel doesn't have.

Part 2:

SFF if that is ITX intel have SFF options its just the natural and ominous competition to it is fast becoming the APU, we have seen the Jaguar cores utilized in the consoles to spectacular effect, AMD can target SFF and HTPC OEM's with the exact same candy with maybe not as highend Radeon core but even if they just recyle the PS4 specs into SFF HTPC's that will completely flush Intel out as we know Intel cannot make anything remotely competitive with the APU or its evolutionary offspring. What is also going to become a problem is HSA and paralell computing, right now its small but everything starts small and HSA is just a far better way of getting more out for less. Then there is the GPU part, while pc's move away from traditional approaches so long as intels igpu is inferior its going to remain on the back foot for all SFF systems. At this moment in history Intel needs a AMD or Nvidia discrete level card to be useful that has a power implication that hinders the drive to ULV and SFF.

I will not say that dropping die size is bad, I just think Intel have been far to aggressive on the power issue and the person that pays the price is the end user.

What do Intel have to work with;

Servers and ULV servers like AMD have looked at.
Cloud servers like AMD
DT x86 (leading the industry here, not only in performance but in Fab process technology)
Mobile x86 laptops, ultrabooks this is still a big spinner.

I am not happy with that direction at all, but I won't worry if AMD sneak a foot in the door somewhere with a product that nobody has that is cheap and efficient at what it does.

 

[mayankleoboy1]

Desk Top is declining tho, not ending.
Im not saying Intel will crash, but Intel has many more problems than AMD does.

Desktop declines. Smartphones rise. So what are we forgetting here, children ?
Servers. What is needed to serves as the backend of all the mobiles and cloud ? Servers. And guess which company rules here ?
That, and HPC.

Server market is the one all the companies want to target. AMD desperately wants to enter here. Some say that Bulldozer/Piledriver were basically designed keeping the server workloads in mind. Since x86 sort of failed here for them, AMD is trying for a seecond shot at it. This time through ARM in the form of Seamicro.

If only they had that 2 billion back from LRB.......

Dont worry, they will get those 2B and plenty more from Xeon Phi.

 

[truegenius]

The next person to tell me that smartphones are replacing desktops or mobility parts, I literally will rape an angel in pure rage quitting fury

my desktop is getting jealous of my smartphone because i spend more time with my smartphone and rarely use desktop, maybe this is a sign of [strike]my smartphone is replacing my desktop[/strike]

[strike]btw, don't forget to call me if you are going to **** an angel, first turn will be mine[/strike]

gaming wise, i have a HTC one X with a quadcore tegra which lasts about 45minutes before its battery is dead so a PS vita is still a better portable gamer

new dual/quad krait and adreno 305/320 based phones may impress you as they have better performance and better battery backup

Im not saying Intel will crash, but Intel has many more problems than AMD does.

intel clearly have x86 performance advantage with a huge margin
they only need to clock their cpu higher (they have enough tdp room) and no amd can match their cpu performance at that price bracket (example i3@4ghz will trash any amd at current price in same price bracket fir majority of tasks)

means they can go for 95w i3 , 125w i5 & i7 , >150w i7 extreme
instead of releasing many models of same series they should opt for less models and more vfm product
for example : locked and unlocked i3/5/7 @4ghz base, these will be enough and since they will have a huge performance gap (like i3 vs i5) then none of them will interfare with each other's sales
but that will surely trash amd cpu


and their "HD" or gaming igpu fever is a foolish thing imo
because this is consuming die area and tdp
thus increasing cost of production and less profit margin

imo, they should leave gaming gpus to amd and nvidia
hd or hd2000 is enough for basic purpose , and for gaming purpose majority (excluding fanboys/girls) user prefer dedicated gpu

so in short
1. clock cpus higher and privide unlocked for each series.
2. decrease number of models in same series
3. limit gpu to hd or hd2000 to save die and tdp
4. leave graphics to amd/nvidia