5 ms·

In an efficient market, companies (Intel and Qcomm in this case) would compete to produce products (5G modem) that can tailor to multiple customers for maximum

In an efficient market, companies (Intel and Qcomm in this case) would compete to produce products (5G modem) that can tailor to multiple customers for maximum volume to recoup R&D cost. This will converge in the best product at the best price. I wonder what would be the benefit of Apple producing their own chip? Especially with the threat of "peak iPhone" looming.

roenxi 8y ago
Lots of possibilities. One simple example is maybe Apple thinks they have a 'special case' problem that other 5G modem suppliers aren't going to solve. Also, theory is all very well but the timespans matter and aren't general. Someone has to physically be the first person to do a new thing. Apple's experience with the original iPhone was that competition caught up eventually. But in the years it took the competition to set up R&D, supply chains, yadda yadda, Apple could cement a dominant market position and turn crazy profits. They might see similar opportunities in other communications technologies (although not on the same scale).
the-dude 8y ago
Ensuring privacy for their users?
simonh 8y ago
The problem facing Qualcomm is that their main customer for high end modems is Apple, the units they sell to the Android manufacturers are much lower volume (at the high end) and for much less money. Apple has noticed that if they make these advanced modems themselves, the Android ecosystem doesn't have enough profit in it to make building high end modems just for that market viable. At least, it might well force Qualcomm to design down their best modems to meet lower price points at lower volumes than they can achieve now. This is aside from the advantages Apple would have in tailoring their hardware and software to each other, and guaranteeing they get the hardware features they need to achieve their end-user facing feature goals.
- martin_bech 8y ago
  And.. if they can make them even more low power, as they have the CPU/GPU and Wireless (Bluetooth/wifi). they might get an even larger advantage on battery/performance.
jakobegger 8y ago
Apple can't build the Macbooks they want because Intel doesn't deliver the chips they need. I'm sure they want to avoid that situation with their iPhone. Additionally, the problem with Qualcomm is that Apple thinks they are charging too much on patent licensing fees — by doing research and building up their own patent portfolio, they will have better leverage in license negotiations even if they use Qualcomm chips.
hobofan 8y ago
> I wonder what would be the benefit of Apple producing their own chip? Especially with the threat of "peak iPhone" looming. Isn't that exactly the answer? If you have reached market saturation, you optimize for cost.
skywhopper 8y ago
Ah, but the market is not efficient. Apple's priorities for very specific chips with specific interfaces and specific power requirements very likely don't always map to their vendors' own interests in selling fewer products to a wider variety of customers on timetables that match their own capital investment cycles. Apple is lucky to be a big enough buyer to have a lot of influence, but if Apple can spend a few hundred million to build a pipeline for radio chips that save them a few dollars each, and take less space, use less power, and are more tightly integrated with their OS and SOC, then it will pay off very very rapidly at their scale. And if it doesn't work out, then the market will still be there for them.

roenxi 8y ago

Lots of possibilities. One simple example is maybe Apple thinks they have a 'special case' problem that other 5G modem suppliers aren't going to solve. Also, theory is all very well but the timespans matter and aren't general. Someone has to physically be the first person to do a new thing. Apple's experience with the original iPhone was that competition caught up eventually. But in the years it took the competition to set up R&D, supply chains, yadda yadda, Apple could cement a dominant market position and turn crazy profits. They might see similar opportunities in other communications technologies (although not on the same scale).

the-dude 8y ago

Ensuring privacy for their users?

simonh 8y ago

The problem facing Qualcomm is that their main customer for high end modems is Apple, the units they sell to the Android manufacturers are much lower volume (at the high end) and for much less money. Apple has noticed that if they make these advanced modems themselves, the Android ecosystem doesn't have enough profit in it to make building high end modems just for that market viable. At least, it might well force Qualcomm to design down their best modems to meet lower price points at lower volumes than they can achieve now. This is aside from the advantages Apple would have in tailoring their hardware and software to each other, and guaranteeing they get the hardware features they need to achieve their end-user facing feature goals.

martin_bech 8y ago
And.. if they can make them even more low power, as they have the CPU/GPU and Wireless (Bluetooth/wifi). they might get an even larger advantage on battery/performance.

jakobegger 8y ago

Apple can't build the Macbooks they want because Intel doesn't deliver the chips they need. I'm sure they want to avoid that situation with their iPhone. Additionally, the problem with Qualcomm is that Apple thinks they are charging too much on patent licensing fees — by doing research and building up their own patent portfolio, they will have better leverage in license negotiations even if they use Qualcomm chips.

hobofan 8y ago

> I wonder what would be the benefit of Apple producing their own chip? Especially with the threat of "peak iPhone" looming. Isn't that exactly the answer? If you have reached market saturation, you optimize for cost.

skywhopper 8y ago

Ah, but the market is not efficient. Apple's priorities for very specific chips with specific interfaces and specific power requirements very likely don't always map to their vendors' own interests in selling fewer products to a wider variety of customers on timetables that match their own capital investment cycles. Apple is lucky to be a big enough buyer to have a lot of influence, but if Apple can spend a few hundred million to build a pipeline for radio chips that save them a few dollars each, and take less space, use less power, and are more tightly integrated with their OS and SOC, then it will pay off very very rapidly at their scale. And if it doesn't work out, then the market will still be there for them.