or Connect
New Posts  All Forums:

Posts by afrodri

 While we certainly haven't hit the limit, we have reached a point of diminishing returns on the number of 'practical' transistors per chip and the impact this has on performance and power (i.e. the impact of moving form 90 to 65nm was far greater than 14 to 10nm will be). Already Moore's law seems to be slowing. For the practical limits of CMOS it is hard to say. Industry folks and the ITRS roadmap seem to get somewhat hazy after 5 or 7nm, though some might push it...
 No, The UBS report said Apple "...is favoring sole-sourcing for "many components." – "many components" implies more than one, "each component" would imply that all components are single source, which is not the case, nor has UBS said so.
 For many components, it is as simple as dissembling the watch and reading the logo on the side of the component. Other times you may have to decipher a part number on the package or even remove the chip's package and look at the silicon to know. Do this with enough watches and if you get the same supplier each time you can build up a statistical probability that they are the sole source.
Strange article. "Software without a firm release date is 'delayed' due to technical problems."   Pretty much every software project ends up taking longer than initial estimates. Particularly software which has to interact with dozens of often poorly defined protocols, random devices, and different partners.
 What, you don't add the customary 15% tip on your taxes???? /s
 Umm... definition #5 that you posted seems to indivate that "very unique" is acceptable. What were you trying to prove?
 Until Moore's Law runs out... :-)
Whoa there, I don't think Katy Huberty was criticizing Apple. The quote came from the latest earnings telcon, and she was asking what the big drivers were. Huberty has been very positive on Apple's outlook and particularly the Apple Watch.  All she did was note that R&D spending has been increasing quickly. 
 Plus a lot of those different chips are fabbed in very different processes. If you try to make DRAM in an MPU/CPU process you end up with memory that has poor density and wastes the several layers of interconnect that an MPU process has. If you try to make a processor in a DRAM process you end up with a slow, bloated processor because DRAM is optimized differently and has fewer layers of metal. Also, while putting everything on one chip offers some nice space and power...
 Interesting. That seems to indicate that a lot of the increase was on the 'Research' side, rather than on 'Development' (i.e. basic technologies rather than product design). I'm glad to see that. Research is riskier, but has a better chance of pulling off some really stunning new breakthroughs which can give Apple a competitive advantage.
New Posts  All Forums: