Apple A5X Die Size Measured: 162.94mm^2, Samsung 45nm LP Confirmed

45 Comments

View All Comments

• ImSpartacus - Sunday, March 18, 2012 - link

  Often times, it's best to just avoid replying to such comments. More often than not, the individual is simply "trolling" for comments.

• gorash - Saturday, March 17, 2012 - link

  I'd say the new screen is pretty overrated. It doesn't look super different.

• KoolAidMan1 - Saturday, March 17, 2012 - link

  I'd say you're pretty blind. It is almost as bad as saying that anti-aliasing doesn't make a difference in games.
  
  After a day with the iPad I want this sort of PPI in ALL of my monitors. My desktop monitors are high end NEC IPS displays, and now all I see are pixels and blockiness in text. Crossing fingers that yields get good enough to allow this sort of pixel density in desktop and laptop displays.
  
  It is insane that a display of this quality exists in a consumer product that only costs $500.

• steven75 - Sunday, March 18, 2012 - link

  Uh, mods? How about some help here.

• UpSpin - Friday, March 16, 2012 - link

  You can't just switch to a different manufacturing process. Their A5, with both CPU and GPu, is designed and optimized for 45nm. They have to invest a lot of money to make it work on 32nm or 25nm. They have to redesign both CPU and GPU. It would be stupid to do this with no further change. Intel requires a whole new Generation for such a move.
  
  For Apple it would have been idiotic, because ARM A15 technology is available now, so they better use their knowledge and time to build a new A6 based on Cortex A15 with the PowerVR6 GPU and build on 25nm. They probabley have to invest the same money and time, but get much much more.
  
  You should rather ask why Apple still uses the Cortext A9 and hasn't switched to A15 yet (look at Qualcomm, it's possible to have A15 technology ready already).

• name99 - Friday, March 16, 2012 - link

  I've said it before, I'll say it again.
  To me everything looks like A5X is a plan B because A6 (for whatever reason) was delayed.
  Look at, for example, the separate RAM that is being used, rather than PoP as in A5. I'm sure that Apple did not want this there, they were forced into it by switching to the larger A5X, and probably not being in a position, in time, to integrate double the RAM into the package.
  I also am guessing that this *externally driven* RAM is a substantial part of the requirement for a large battery. I find it hard to believe this much larger battery (which supposedly gives us much the same lifetimes for various tasks as the previous iPads) is required for the screen.
  
  People keep seeming to forget that we went through this transition before:
  The iPhone 3GS had a 1219 mAh battery, the iPhone4 had a 1420mAh battery. The iPhone 4 was generally considered to have rather better battery life than iPhone 3GS, especially in tasks like movie-watching. That seems to indicate that, contrary to the "wisdom" of web commenters, there is no intrinsic reason that a highDPI screen has to burn power.
  
  Sure, it is possible that this particular screen design burns power, in a way that the iPhone 4's does not --- but I think the onus on people who claim this is to provide proof, not to simply assert "well of course it does".

• Steelbom - Friday, March 16, 2012 - link

  But 2048x1536 is so much higher than 960x640. The GPUs will be needed more often to power that display, and the display may have needed a more powerful backlight.

• name99 - Friday, March 16, 2012 - link

  This is what I am saying. THINK.
  What determines the power usage of an LCD display. The backlight.
  So why should more pixels over the same area require more energy? The total emitted light is the same.
  
  Now in principle there COULD be effects related to more "borders" between pixels resulting in more dead area that doesn't channel light --- aperture effects. In practice,
  (a) this seems to have been taken care of in the manufacturing process --- there was an article on it flooding the web about five days ago
  (b) like I said, we have the example of the iPhone retina display. If that doesn't require a substantial boost in power over its predecessor, why should the iPad display be different?
  
  There ARE more backlight LEDs in the iPad3 screen. But that does not imply that more backlight power is being generated --- they may just be there to create a more even backlight. Certainly my iPad1, while having a quite acceptable screen, had patches where light would bleed through a pure black image more so than in other regions of the screen.

• Steelbom - Saturday, March 17, 2012 - link

  You're assuming the backlight is the same, and not more powerful. With the iPhone, they only had to pack 614,000 pixels in a small area, but with the iPad they're dealing with five times more pixels on a larger area. They even mentioned that they had to (rough remembering of it:) separate the "pixels" from the "signals" in the display and lift the former up so that signals don't get crossed. That may make it thicker, and make it need a more powerful backlight.
  
  Plus it requires more power from the GPUs to run the display. I doubt the RAM would increase power consumption by anything but a small amount.

• ssj3gohan - Thursday, March 22, 2012 - link

  Let's put some science into this thread to end the speculation.
  
  The transmissivity (amount of light it lets through) of a screen is a function of the transmissivity of the pixels and the fill factor (fraction of the total area that is actually pixels and not black). In general: higher pixel density = lower fill factor and lower transmissivity of the individual pixels. That is exactly what is going on here; the new iPad screen has significantly lower fill factor and thus, for the same luminance, needs a stronger backlight.
  
  This is the sole reason for the larger battery. The vast majority of power is sucked up by the screen even in the original iPad and iPad 2.