Will AMD design ARM chips?

[drizek]

But what about these $150 smartphones on Virgin Mobile like LG Optimus V? I've consistently read reviews claiming battery life is one of its weak points. Surely there is a market for SOC contributing to better run time? (especially with lower power displays on the horizon)

You're assuming that the consumers and the manufacturers actually care about having a high quality product.

You spend about $80 a month for a smartphone with a data plan. Multiply that by a 24 month contract and you get around $2000. So you can get an Optimus T for a total of $2000.99 or an Optimus S for only 10% more at $2199.99. The reality is that you would have to be pretty thick to buy a free-on-contract phone, and the people that fall for the trap aren't going to care about having an ARM 11 instead of a Cortex A5. The people that do care would happily pay the extra 10% to get 100% more performance from A9 versus A5.

 

[cbn]

You're assuming that the consumers and the manufacturers actually care about having a high quality product.

You spend about $80 a month for a smartphone with a data plan. Multiply that by a 24 month contract and you get around $2000. So you can get an Optimus T for a total of $2000.99 or an Optimus S for only 10% more at $2199.99. The reality is that you would have to be pretty thick to buy a free-on-contract phone, and the people that fall for the trap aren't going to care about having an ARM 11 instead of a Cortex A5. The people that do care would happily pay the extra 10% to get 100% more performance from A9 versus A5.

The prepaid plans are much cheaper.

The LG Optimus V is offered on Virgin Mobile ($25/5GB data/300 minutes/month).

So the total difference in cost over 24 months is much larger than you think.

• $200 subsidized contract phone + $80 month over 24 months= $2120
• $150 unsubsidized phone + $25 month over 24 months=$750
  

That makes the subsidized phone/contract service nearly three times more expensive.

 

[drizek]

The carriers are the other part of this. Virgin is owned by Sprint. If they gave Virgin phones that perform like the Evo and the Epic, they would lose out on $1200.

I bet if LG built a phone with a nice solid Cortex A5 and sold it for cheap, Sprint would not let you use it prepaid. They want to maintain the divide.

The entire industry is just messed up. I think you're more likely to see these types of phones in Europe, but again, even there, using A5 instead of A9 really won't bring prices down too much. In developing countries where people are more price sensitive, people are probably just going to stick with nokia/symbian anyway.

 

[cbn]

The carriers are the other part of this. Virgin is owned by Sprint. If they gave Virgin phones that perform like the Evo and the Epic, they would lose out on $1200.

I think the carriers will just give out better phones than Evo 4G. (eg, Cortex A15 smartphones in 2012-2013).

Although with that being said I just wonder what the ceiling of this madness will be? How fast can the processor become before the 4G network becomes the bottleneck?

The entire industry is just messed up. I think you're more likely to see these types of phones in Europe, but again, even there, using A5 instead of A9 really won't bring prices down too much. In developing countries where people are more price sensitive, people are probably just going to stick with nokia/symbian anyway.

Yep, I've consistently read that the rest of the world doesn't operate on contracts like the US does.

So in these circumstances chip cost/phone cost really matters, but the upside (I would think) is potential for much higher volumes.

Another thing to consider is how a prepaid 3G network would interact with a faster Cortex A9 processor. Is having a faster processor really useful in this instance? Or does having the extra grunt for WiFi and running offline apps make such a CPU worth it?

EDIT: With respect to phone cost of a Cortex A5 vs a Cortex A9 design: I also wonder how much mainboard design cost would factor in? I am no engineer, but from what I have read higher power designs usually require more PCB layers, etc

 

[cbn]

In a perfect world where AMD had no debt and a sizable net positive cashflow they could undertake a serious side-effort like that without putting the rest of their business in peril.

But is AMD really in a position to go all out in ARM?

Would a new CEO with a new business plan be able to secure additional funding?

 

[Rifter]

Would a new CEO with a new business plan be able to secure additional funding?

maybe if BD is a grand slam.

 

[Idontcare]

Would a new CEO with a new business plan be able to secure additional funding?

Yes, but it most likely would come in the form of being bought out and then funded internally by their new overlords.

 

[cbn]

Deleted

 

[OBLAMA2009]

why wouldnt they. arm obviously has a bright future. and if win 8 is on arm theres an opportunity for amd to come out with a high performance chip that could actually run it decently. theyd could have the arm windows market to themselves

 

[sangyup81]

AMD should not have settled with Intel if their future is in ARM

 

[drizek]

In developing countries where people are more price sensitive, people are probably just going to stick with nokia/symbian anyway.

Wow, I made a wrong call on that one.

 

[Tuna-Fish]

Wow, I made a wrong call on that one.

The developing world never bought Nokia/Symbian. The Nokia platform that owns the developing world is the S40, which is not a multitasking operating system, and can trace much of it's success to extreme simplicity. It's still selling like hot cakes to anyone who doesn't want apps in their phone. This segment is, however, narrowing quickly, and the new Nokia Meltemi platform is meant to be it's spiritual successor. I am vastly more optimistic about the success of Meltemi than I was of Nokia's previous efforts, because I believe it is being led by the same people who developed phones that sold 1.5 billion worldwide, as opposed to the "superstar" clowns who ran Maemo/Meego/whatever.

 

[drizek]

Oh. I thought S40/S60 were just two different forms of Symbian.

Ugh. Maemo was such a disappointment. I did get Android running on my N800 though before anybody really had an android phone, so that's something, I guess.

 

[Tuna-Fish]

Oh. I thought S40/S60 were just two different forms of Symbian.

No. They actually shared very little considering that they were developed under the same roof. S40 was from first principles a close-to-the-metal efficient "firmware", without most of the features people associate with operating systems. S60, later refactored to Symbian, was Nokia's attempt to make a "modern" multitasking operating system based on the Psion EPIC that Nokia acquired.