AMD's ARM Strategy: Crazy Or Brilliant? [View article]
Justin,
>> The Acer laptop design used for the review of the a6-1450 is pretty terrible. It has a really nice IPS touch screen, but one can buy i3/i5 acer laptops for not much more than the asking price of the Temash variant, with a non touchscreen TN panel.
Acer seems to be putting all of its high end energy around Intel's "Ultrabook" marketing push. This is another hurdle that AMD - from a financial, not a technical - perspective will need to overcome.
>>And I'm getting tired of sharing the laptop with the wife, so I'm pretty excited to see some of these new products so I have a reason to go test them out first hand :)
I am sure you will write up your first hand experience(s) with such devices, and you will always have a reader in me for your articles :)
>>I am expecting to see a good lineup of temash APUs available probably before summer is out, and hopefully some Kabini variants as well. Haswell will be closer to Christmas time won't it?
Haswell high end stuff will be out in June, with Ultrabook designs available shortly thereafter. It is "Baytrail" that is a Christmas-time launch, so AMD has a decent lead with 2 core Temash there.
Advanced Micro Devices: Is Goldman Sachs Right? [View article]
>> I also read before that the Phi is not really 1.1TFLOPS, but actually more like 500GFLOPS, and only if you actually do a multiply and an add at the same time do you get up to a TFLOP
This is how all GPU measurements are done. FMA = fused multiply add, gives you essentially two "FLOPS" for the price of one.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
Justin,
You know I think highly of your views and writing, and I know you are always respectful when you disagree, so I never got that impression.
I think Intel's "Clover Trail" is decent, but obviously the "Temash" will destroy it in terms of raw performance - it's a chip that's well suited for low end PCs which are generally much more powerful than the best tablets.
My only question is whether AMD's power states/power management is at the right level of sophistication for a "real" tablet. This is what remains to be seen, and this is why I am so carefully looking for any and all information on idle power measurements for the chip.
Products should be here soon, in high volume, so we'll have our answers soon enough!
Advanced Micro Devices: Is Goldman Sachs Right? [View article]
joezane,
Skepticism is always warranted when the company has disappointed financially (and from a product perspective) time and time again. I am always open to being pleasantly surprised, and I obviously would love to see AMD do well, but I have learned hard lessons from being overly bullish (and bearish!) on names based on what *appears* to be.
I made the call to speculate on AMD in the $2's, I put my own money on the line, and it paid off. At $4, new buyers need to be much more skeptical. That is all I'm trying to say.
Intel: Goodbye, Paul Otellini - You Will Be Missed [View article]
Jeach!
Sorry for the confusion - the micro-architecture teams for the "tick" and the "tock" are the same group. The process technology development team is a separate team that, of course, works very closely with the micro-architecture and physical design teams.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
joezane,
The big cores simply cost a lot more to develop, and it seems that "Jaguar" can really drive sales in high volume segments - the goal of any business. Why not focus on what's profitable, and cut back on what's not?
If AMD goes all in on Jaguar's successors and forgets trying to have *two* teams, then it can really have a shot of delivering a knock-out punch. But spreading its resources this thinly, and on top of that trying to stay world class in GPUs, may be a little too much.
It's better to do a few things VERY well, than to merely have "OK" solutions across the board.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
MorrisSean,
Intel will actually be in the micro-server space with its own 22nm "Atom" based part called "Avoton" along with a whole roadmap of successors. These get to market *before* AMD or anybody else's ARM based products, and could seriously disrupt any chance that ARM based solutions of gaining the massive traction that ARM touts.
But as I have said repeatedly - AMD shares will go much higher as it will take time to see how this all plays out, since this is a raging bull market and anything with a "story" and a short interest will get squeezed badly.
Structural Change In The Mobile Processor Marketplace: Intel Wins; ARM, AMD Lose [View article]
I believe that is powered by Qualcomm apps processor and modem.
Structural Change In The Mobile Processor Marketplace: Intel Wins; ARM, AMD Lose [View article]
Look up the Motorola Razr i :)
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
>> The Acer laptop design used for the review of the a6-1450 is pretty terrible. It has a really nice IPS touch screen, but one can buy i3/i5 acer laptops for not much more than the asking price of the Temash variant, with a non touchscreen TN panel.
Acer seems to be putting all of its high end energy around Intel's "Ultrabook" marketing push. This is another hurdle that AMD - from a financial, not a technical - perspective will need to overcome.
>>And I'm getting tired of sharing the laptop with the wife, so I'm pretty excited to see some of these new products so I have a reason to go test them out first hand :)
I am sure you will write up your first hand experience(s) with such devices, and you will always have a reader in me for your articles :)
>>I am expecting to see a good lineup of temash APUs available probably before summer is out, and hopefully some Kabini variants as well. Haswell will be closer to Christmas time won't it?
Haswell high end stuff will be out in June, with Ultrabook designs available shortly thereafter. It is "Baytrail" that is a Christmas-time launch, so AMD has a decent lead with 2 core Temash there.
Structural Change In The Mobile Processor Marketplace: Intel Wins; ARM, AMD Lose [View article]
Are you buying ARMH here?
Advanced Micro Devices: Is Goldman Sachs Right? [View article]
This is how all GPU measurements are done. FMA = fused multiply add, gives you essentially two "FLOPS" for the price of one.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
You know I think highly of your views and writing, and I know you are always respectful when you disagree, so I never got that impression.
I think Intel's "Clover Trail" is decent, but obviously the "Temash" will destroy it in terms of raw performance - it's a chip that's well suited for low end PCs which are generally much more powerful than the best tablets.
My only question is whether AMD's power states/power management is at the right level of sophistication for a "real" tablet. This is what remains to be seen, and this is why I am so carefully looking for any and all information on idle power measurements for the chip.
Products should be here soon, in high volume, so we'll have our answers soon enough!
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
Advanced Micro Devices: Is Goldman Sachs Right? [View article]
Skepticism is always warranted when the company has disappointed financially (and from a product perspective) time and time again. I am always open to being pleasantly surprised, and I obviously would love to see AMD do well, but I have learned hard lessons from being overly bullish (and bearish!) on names based on what *appears* to be.
I made the call to speculate on AMD in the $2's, I put my own money on the line, and it paid off. At $4, new buyers need to be much more skeptical. That is all I'm trying to say.
Structural Change In The Mobile Processor Marketplace: Intel Wins; ARM, AMD Lose [View article]
14nm FinFETs are a long, long way away.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
Excellent post, sir. Thank you.
Intel: Goodbye, Paul Otellini - You Will Be Missed [View article]
Sorry for the confusion - the micro-architecture teams for the "tick" and the "tock" are the same group. The process technology development team is a separate team that, of course, works very closely with the micro-architecture and physical design teams.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
The big cores simply cost a lot more to develop, and it seems that "Jaguar" can really drive sales in high volume segments - the goal of any business. Why not focus on what's profitable, and cut back on what's not?
If AMD goes all in on Jaguar's successors and forgets trying to have *two* teams, then it can really have a shot of delivering a knock-out punch. But spreading its resources this thinly, and on top of that trying to stay world class in GPUs, may be a little too much.
It's better to do a few things VERY well, than to merely have "OK" solutions across the board.
AMD's ARM Strategy: Crazy Or Brilliant? [View article]
Intel will actually be in the micro-server space with its own 22nm "Atom" based part called "Avoton" along with a whole roadmap of successors. These get to market *before* AMD or anybody else's ARM based products, and could seriously disrupt any chance that ARM based solutions of gaining the massive traction that ARM touts.
But as I have said repeatedly - AMD shares will go much higher as it will take time to see how this all plays out, since this is a raging bull market and anything with a "story" and a short interest will get squeezed badly.
Structural Change In The Mobile Processor Marketplace: Intel Wins; ARM, AMD Lose [View article]
Superb contribution. Welcome to SA!