Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Maybe. But, again, the number being in the atomic range doesn’t mean much. A lot depends on the geometry of the transistor, and the actual size of its components, and the “node size” doesn’t tell you too much about that.

Plus we’re really in the silicon-germanium era right now (using germanium to create kinks in the conduction/valence bands to increase carrier lifetime), and switching to a different material (like GaAs/AlGaAs) doesn’t help much if what you are worried about is not being able to shrink any more.

True enough, but it might all be moot once we get down to the 500pm size, since quantum physics does weird things to the positioning of atoms. Basically, we'll have to move on from gate-based transistors.
 
  • Like
Reactions: amartinez1660
Conceivably, if nodes keep being reduced by roughly 1/2 for the next decade, we'll enter the atomic range soon:

2019: 5nm

2022: 3nm

2025? 1nm - 1.5nm?

Then we enter picometers.

2030? 500pm - 750pm? (.5nm - .75nm)

Depending on the atom, that's anywhere from 50pm - 500pm in size. Basically, silicon's days will be numbered once we reach the 2030s.
I'm not really sure what they'll use considering the current nodes are more like a marketing term and not really representative anymore. TSMC going from 7->5 mainly points at "the next generation" instead of going from 7->5 literally.

For example AMD is currently on 7nm and Intel on 10nm (or 10+) but AMD's 7nm isn't smaller than Intel's 10nm, they're just measuring differently.
 
  • Like
Reactions: amartinez1660
True enough, but it might all be moot once we get down to the 500pm size, since quantum physics does weird things to the positioning of atoms. Basically, we'll have to move on from gate-based transistors.
That’s why I’m not too worried. In 1992 I started work on my ph.d. project that involved HBT transistors on GaAs, because my advisor was convinced we were just a few years away from the end of FETs, due to quantum effects.

We keep finding ways to keep things moving, because the alternatives aren’t too appealing. :)
 
Why? iPhone wont' take advantage of this really. Its more for iPads and AS than iPhone that already can't fully utilise current chip.

There was rumour doing the rounds that Apple is secretly developing 'full fat' Mac OS for iPhone and iPad.

And?

You'll dock the iPhone/iPad to your 'Mac' or screen and have a 'Mac.'

That you'll have the ultimate in sycronicity this way. ie. A powerful iPhone 'Mac' (that blows my former 2012 iMac out the water...) that you can share full mobility with your 'base' Mac.

Imagine being able to harness a powerful iPhone and iPad at your Mac dock eg. to do 3d rendering or encoding video? A mini x-grid of power processing.

Far fetched?

I think it was Max Tech' that was demoing the iPad itself doing 8k encoding on some codecs that would bring a Mac Pro or iMac to its knees.

Regardless. The A14 will be quite the leap from the A12. In terms of 'X' power. This is the one I've been waiting for. And it should give a nice 'taste' of what's to come with AS...

...if AS doesn't beat the iPad A14x to market...

Azrael.
 
  • Like
Reactions: amartinez1660
I've got my iPhone XR and I keep my phone on a 3 to 5 year cycle depending on different factors. I'm mostly anticipating the ARM iMac's. My iMac is going to be 10 in a couple of years (2022) and I'm not interested in getting another Intel iMac when the ARM iMac's are so close to production and the length I take to replace my Mac it makes the most sense to just wait for the new fleet of ARM Macs.
 
TSMC is amazing. How come Intel can't do it?
Intels 10nm is denser than TSMCs 7nm but TSMCs 7nm+ is denser than Intels 10nm, you can't just compare them anymore.
They're more like marketing terms and different companies measure differently. Intel isn't as much behind in density as the numbers might make it appear (altho they're still behind).
 
Whilst my iMac was 'dead' it would have been nice to have hooked up an iPhone to it to boot the screen in lieu of the fried gpu...

One day?

Azrael.
 
This is awesome. But it also speaks of future lack of significant year over year performance gains as the miniaturization will soon hit a dead end. But Apple will make up for that with it’s integrated SOC technologies that handle more and more aspects of the systems such as machine learning and image processing.
 
Each time I read about TSMC / Apple using 5nm or now 3nm process for their chips, I wonder if they may start using multiple processors in devices, that use less energy and provide more processing power than the competition's single chips. Would certainly make sense with Apple's "Pro" laptops and desktops, maybe even the iPad Pros.
 
don't get too excited about whatever power consumption improvements the new chip may bring, the next iPhone that uses it will just have a smaller battery to offset the gain. Count on it.
Given that the last few years the battery has increased in size?
 
Each time I read about TSMC / Apple using 5nm or now 3nm process for their chips, I wonder if they may start using multiple processors in devices, that use less energy and provide more processing power than the competition's single chips. Would certainly make sense with Apple's "Pro" laptops and desktops, maybe even the iPad Pros.

That makes no sense. There are already multiple processors - they are just all on one chip. If you have to choose between N cores on 2 chips or on 1 chip, putting them all on 1 chip is always better. And shrinking the process node allows one to fit more transistors onto a chip (generally), so they will just keep putting more cores on the same chip, as long as doing so helps. (There are diminishing returns, as evidenced by Apple's hybrid approach that focuses more on single core performance and special-purpose cores, vs. samsung/qualcomm which tend toward more, general purpose, cores.)
 
WRT the CPU, outside of the manufacturing process, performance improvements typically come from improvements in Branch Prediction & the overall memory organization (on-chip & off-chip).

In the case of the A14, I believe the two "Pro" models will get a LPDDR5 PHY Controller !

So, tack-on another 15-20% boost in Perf for those two models !
 
  • Like
Reactions: amartinez1660
Well, device size does not mean process node size. In this case a Silicone atom is 0.2nm in length. We are approaching that limit in which single atoms become the Transistors vs now a few atoms make a transistor.
LOL Thanks; you saved me a Google
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.