or Connect
New Posts  All Forums:

Posts by afrodri

 It is more of a backward-compatibility issue than the quality of the software. Often, someone working in a given market has a lot of existing slide decks that they pull from when making new presentations, or receives slides from people in the industry. Keynote's import feature is OK, but there are still inevitable errors and manual tweaking which must be done. Eventually you reach 'critical mass' with the new software, but it takes a while.
 I've seen the A8 estimated at 2 Billion, not 3. This probably includes the 1GB LP DDR, which is a billion transistors by itself. I'm guessing the 1.4B number is just the SRAM and logic transistors.  http://www.extremetech.com/computing/189787-apples-a8-soc-analyzed-the-iphone-6-chip-is-a-2-billion-transistor-20nm-monster
 Yes, 'DDR4' and 'LPDDR4' are both different standards with very different physical implementations and protocols. LPDDR is designed for low power devices and has lower bandwidths than DDR which is designed for desktops/laptops/servers. DDR consumes more power, but gives better performance. They diverged some time ago. As I recall, the first LPDDR was similar to DDR, but they have both had separate evolutions since then. They just happen to have gone through about 4...
 Those articles are about LPDDR4, which is a different standard than DDR-4.
It should be pointed out that Micron is not the only one making LPDDR4, in fact, the presentation that Mr. Margolis focuses on, though presented by a Micron employee, is just discussing the LPDDR4 standard in general (it is a JEDEC standard, which is the body that defines memory standards like DDR and LPDDR). There is nothing in the article that indicates that Micron has a 'special' LPDDR4, just that they (like the other memory vendors) will be producing LPDDR4. All of the...
 120 hours is too high. The CPU is only part of a laptop's power draw. The DRAM, GPU, screen, power supply, wireless, etc... all take up a lot of power. Even if the CPU took zero power you might only see a 10-20% battery life improvement. I think it is also a little far to say 'ARM is basically a PPC' – they are both RISC cores, but the ISAs are very different, and, more importantly, their memory models are different. Porting can be done, but it is very non-trivial.And...
 That's not really true. 64-bit processors have wider datapaths and registers which can make a number of computations faster. Also, the ARM v8 instructions (i.e. the ones Apple is now targeting) add a range of other features: http://www.arm.com/files/downloads/ARMv8_Architecture.pdf
  I think some of it is a diversity argument - if you put all your fabs in one basket, you could be exposed to greater risk due to political changes, exchange rate fluctuations, regulatory changes, natural disasters, etc...  Also, New York probably has a lot of engineering talent which can be tapped, due to the long history of fabs in the area. And there is the marketing appeal - Apple has gotten a lot of flak over parts of its production chain being outside of the US,...
  I'm confused, Google's income is up 16% since this quarter last year, and their revenue is up a "considerable" 19%. That is flat?    Or are the numbers in the article not GAAP?
  While scaling is very dependent on the application, the idea that "...scalability of an app has nothing to do with the OS." is absurd. OS-level thread scheduling is still a very active area of research, especially for power conscious systems.  How and when the OS schedules threads, how it handles coherency and sharing, how it manages inter-thread communication and synchronization, how interrupts are handled, how threads/tasks are bound to cores, etc... all can have huge...
New Posts  All Forums: