The power-saving potential here is tremendous. This is where Apple could set itself apart from the rest yet again.
Not to mention the money-saving potential. The iPhone 5S wireless chips at $32 are second only to the display and touch ($41) costs in the BOM. Wireless is about 17% of the total BOM cost of $191.
Apple needs to integrate it into their SOC for power savings. Qualcomm sells them discrete chips, or within one of their SOCs. This is obviously not to Apple's liking. I always thought apple would license a core of some sort from qualcomm and integrate it, but in reality this is their last big advantage they have over the A-series seeing that they do not own the compilers, OS, devices, etc.
Remember, apple is not going to just take telecom 101 and walk into a lab with a soldering iron. They will end up with what they want sooner or later.
Would be interesting to see if that team focuses on silicon on sapphire (SOS) technologies. With Apple's drive to sapphire I wouldn't be surprised. It would make for better power consumption, better signal reception, and could support enough RF band so they could build a 'one iPhone works everywhere in the world' chip.
Currently there are better SOS variants out there than current phones use. They don't get much utilization because they are still priced substantially higher and existing technologies are viewed as 'good enough'
Going beyond 'good enough' to deliver a better technology at higher price sounds pretty consistent with Apple, and with their increased volumes of sapphire capabilities it might be an avenue they can explore.
Would be interesting to see if that team focuses on silicon on sapphire (SOS) technologies. With Apple's drive to sapphire I wouldn't be surprised. It would make for better power consumption, better signal reception, and could support enough RF band so they could build a 'one iPhone works everywhere in the world' chip.
Currently there are better SOS variants out there than current phones use. They don't get much utilization because they are still priced substantially higher and existing technologies are viewed as 'good enough'
Going beyond 'good enough' to deliver a better technology at higher price sounds pretty consistent with Apple, and with their increased volumes of sapphire capabilities it might be an avenue they can explore.
Yes!
If Apple does SoS, I suspect that their first iteration will be a separate chip -- rather than part of the Ax APU. A wireless radio (Cell, WiFi, BT, GPS) would seem a natural first step.
It makes a lot of sense to eventually integrated the baseband processor into the A series CPU/GPU (to improve performance, battery life, reliability, and reduce cost). Designing their own discrete baseband processor would be the obvious incremental step along that path.
Of course they need to match the performance and capabilities of Qualcomm's global LTE chips.
Qualcomm will only keep improving their design:
Don't forget, Qualcomm and Intel already have LTE-Advanced solutions. In the case of Qualcomm their LTE-A chips have been on the market in South Korea since Q2 2013, and they'll be launching in Q2 2014 in USA with AT&T.
Here is something I heard which I believe is true about Apple's SoC design, unlike Intel and companies like Broadcom and Qualcomm, they all have to make products which work with a wide verity of application and customer and they automate much of their design. They do no have the time to thoroughly optimize the design so it all things to all people, but does not do any of it was well as it could.
Apple on the other hand only has to worry about its SoC working in Apple products and only has feature in it that only apple uses and has nothing extra to bleed off performance and resources to make it work. Where most silicon chips are auto-routed to same time, apple hand routes the chip so they are optimizing the overall performance of the chip. This has benefits form speed, heat and power which the guys who automate the task.
I believe the above to be true since I seen an analysis of a tear-down of Apple SoC, they could do the same with the Baseband chips as well and possible get ride of performance robbing piece of the design which you know exist
Apple needs to integrate it into their SOC for power savings. Qualcomm sells them discrete chips, or within one of their SOCs. This is obviously not to Apple's liking. I always thought apple would license a core of some sort from qualcomm and integrate it, but in reality this is their last big advantage they have over the A-series seeing that they do not own the compilers, OS, devices, etc.
Remember, apple is not going to just take telecom 101 and walk into a lab with a soldering iron. They will end up with what they want sooner or later.
Like the A series chip in a laptop, desktop, or server. Ha..Ha...Ha....
There is little debate over should they or shouldn't they. We all know full well that Apple will only do it themselves if they can MATCH or exceed performance, while also gaining other benefits such as cost/size/power.
There is little debate over should they or shouldn't they. We all know full well that Apple will only do it themselves if they can MATCH or exceed performance, while also gaining other benefits such as cost/size/power.
Apple's is probably motivated to make it more difficult for the competition to follow, higher margins, smaller thinner designs and long term a seat at the table to have a say on future modem development/standards.
I'm not an engineer so I'm just guessing, but these radio processors must be one of the main power hogs after the LCD and CPU. As someone alluded to above, it's likely one of the more expensive parts as well. In other words, there's a lot of motivation for Apple to take this in-house. Reductions in power consumption and cost would offset new innovations such as the sapphire glass.
I would pile on in that building your own baseband chip that is
Comments
Not to mention the money-saving potential. The iPhone 5S wireless chips at $32 are second only to the display and touch ($41) costs in the BOM. Wireless is about 17% of the total BOM cost of $191.
http://technology.ihs.com/451425/groundbreaking-iphone-5s-carries-199-bom-and-manufacturing-cost-ihs-teardown-reveals
Remember, apple is not going to just take telecom 101 and walk into a lab with a soldering iron. They will end up with what they want sooner or later.
Currently there are better SOS variants out there than current phones use. They don't get much utilization because they are still priced substantially higher and existing technologies are viewed as 'good enough'
Going beyond 'good enough' to deliver a better technology at higher price sounds pretty consistent with Apple, and with their increased volumes of sapphire capabilities it might be an avenue they can explore.
Yes!
If Apple does SoS, I suspect that their first iteration will be a separate chip -- rather than part of the Ax APU. A wireless radio (Cell, WiFi, BT, GPS) would seem a natural first step.
So it seems does Samsung.
http://www.deltapartnersgroup.com/our_insights/articles/lte-asia-article-series-part-four-samsung-swims-upstream-again-
LTE is set to be the next litigation battlefield.
I just wish some court, anywhere, would set FRAND fees in some indisputable framework which can be applied universally.
It makes a lot of sense to eventually integrated the baseband processor into the A series CPU/GPU (to improve performance, battery life, reliability, and reduce cost). Designing their own discrete baseband processor would be the obvious incremental step along that path.
Of course they need to match the performance and capabilities of Qualcomm's global LTE chips.
Qualcomm will only keep improving their design:
Don't forget, Qualcomm and Intel already have LTE-Advanced solutions. In the case of Qualcomm their LTE-A chips have been on the market in South Korea since Q2 2013, and they'll be launching in Q2 2014 in USA with AT&T.
Here is something I heard which I believe is true about Apple's SoC design, unlike Intel and companies like Broadcom and Qualcomm, they all have to make products which work with a wide verity of application and customer and they automate much of their design. They do no have the time to thoroughly optimize the design so it all things to all people, but does not do any of it was well as it could.
Apple on the other hand only has to worry about its SoC working in Apple products and only has feature in it that only apple uses and has nothing extra to bleed off performance and resources to make it work. Where most silicon chips are auto-routed to same time, apple hand routes the chip so they are optimizing the overall performance of the chip. This has benefits form speed, heat and power which the guys who automate the task.
I believe the above to be true since I seen an analysis of a tear-down of Apple SoC, they could do the same with the Baseband chips as well and possible get ride of performance robbing piece of the design which you know exist
Apple needs to integrate it into their SOC for power savings. Qualcomm sells them discrete chips, or within one of their SOCs. This is obviously not to Apple's liking. I always thought apple would license a core of some sort from qualcomm and integrate it, but in reality this is their last big advantage they have over the A-series seeing that they do not own the compilers, OS, devices, etc.
Remember, apple is not going to just take telecom 101 and walk into a lab with a soldering iron. They will end up with what they want sooner or later.
Like the A series chip in a laptop, desktop, or server. Ha..Ha...Ha....
There is little debate over should they or shouldn't they. We all know full well that Apple will only do it themselves if they can MATCH or exceed performance, while also gaining other benefits such as cost/size/power.
There is little debate over should they or shouldn't they. We all know full well that Apple will only do it themselves if they can MATCH or exceed performance, while also gaining other benefits such as cost/size/power.
Apple's is probably motivated to make it more difficult for the competition to follow, higher margins, smaller thinner designs and long term a seat at the table to have a say on future modem development/standards.
I'm not an engineer so I'm just guessing, but these radio processors must be one of the main power hogs after the LCD and CPU. As someone alluded to above, it's likely one of the more expensive parts as well. In other words, there's a lot of motivation for Apple to take this in-house. Reductions in power consumption and cost would offset new innovations such as the sapphire glass.
I would pile on in that building your own baseband chip that is
a) Designed to the rest of your chipset, and OS
delivering better internal performance (speed)
delivering better operating performance (power consumption)
b) possibly dis-integrated and 're-integrated' into your chipset
partnering with chips in a way that offloads stuff like the M7 works with the A7
c) optimized to your specific set of bands and protocols.
Introduce apple only radio based functions (patented local low power networks for things like iRing/iWatch,iEarring,iNose ring)
iBeacon V2 technology
Mesh networking
Pcell networking
I wouldn't think they would do this just to save a few bucks. Any chip Apple builds would be all about separating iDevices
from all competitors.