A16 Bionic

exoticspice1 said:
It's miniscule improvement. That's why Apple never even highlighted the A16 GPU because it's the same. They don't even mention it being new on their website.
They just say "5-core GPU".
Click to expand...
15fps to 20 or 23 is a small improvement? TIL.

They don’t mention it because they don’t want to draw too much attention to the fact that the 14 is using an old soc.
 
Jimmyjames said:
15fps to 20 or 23 is a small improvement? TIL.
Click to expand...
Well in 1080P and 1440P its the same. Also remember iPhone 14 series got better thermal system. Any other improvement is cause of LPDRR5.

I don't need to argue with you, Apple themselves never said the GPU is new.
 
exoticspice1 said:
It's miniscule improvement. That's why Apple never even highlighted the A16 GPU because it's the same. They don't even mention it being new on their website.
They just say "5-core GPU".
Click to expand...
Well, Gfxbench website lists the median score of the iPhone 13 Pro in 4K Aztec Ruins High at 15.3 fps, and the iPhone 14 Pro at 23.0 fps, which would be a 51% improvement, hardly a minuscule improvement. I don't believe it, but I don't believe the 0%/-1% improvement either. I'll wait until the official review embargo lifts, as we're seeing contradicting leaked results.
 
exoticspice1 said:
Well in 1080P and 1440P its the same. Also remember iPhone 14 series got better thermal system. Any other improvement is cause of LPDRR5.

I don't need to argue with you, Apple themselves never said the GPU is new.
Click to expand...
Well, as far as I can see there's no difference if you look at the best bench result (both iPhone 13 Pro and 14 Pro are 54fps), but median bench results is 38fps for the 13 Pro vs 54fps for the 14 Pro.
 
exoticspice1 said:
Well in 1080P and 1440P its the same. Also remember iPhone 14 series got better thermal system. Any other improvement is cause of LPDRR5.

I don't need to argue with you, Apple themselves never said the GPU is new.
Click to expand...
Why would you exclude the most demanding resolution? All the benchmarks I’ve seen show improvement. The amount varies. Geekbench is 8% iirc and Antutu is huge. Gfxbench seems to show a significant improvement.

We’ll see how it pans out. It’s early days but I don’t see how the conclusion that the gpu is unchanged can be justified.
 
Andropov said:
Well, as far as I can see there's no difference if you look at the best bench result (both iPhone 13 Pro and 14 Pro are 54fps), but median bench results is 38fps for the 13 Pro vs 54fps for the 14 Pro.
Click to expand...
I'm going to run GFXBench on my iPhone 13 Pro now. Maybe those 13 Pro 'best' scores are pre-production A16 iPhones misreporting their model number? It's a huge deviation from the median, and it just happens that the best 'iPhone 13 Pro' result matches the median iPhone 14 Pro... could be just a coincidence, sure, but...

Another explanation (if the GPUs truly are the same) could be that the iPhone 13 Pro's best scores pop up when that test is run alone, without previous thermal throttling from other benches, and the iPhone 14 Pro just doesn't throttle due to the new thermal system so the median almost matches the best result. Idk. Still should show some improvement due to memory bandwidth improvements.
 
Andropov said:
Another explanation (if the GPUs truly are the same) could be that the iPhone 13 Pro's best scores pop up when that test is run alone, without previous thermal throttling from other benches, and the iPhone 14 Pro just doesn't throttle due to the new thermal system so the median almost matches the best result. Idk. Still should show some improvement due to memory bandwidth improvements.
Click to expand...
Maybe this is it: testing a single test (to avoid thermal throttling), I can get quite close to the median iPhone 14 Pro results: I got 20.6 fps in 4K Aztec Ruins (High Tier) Offscreen (median iPhone 13 Pro: 15.8 fps, median iPhone 14 Pro: 23fps).

The result I got the first time (running all tests) in 4K Aztec Ruins (High Tier) Offscreen is 11.5fps, even worse than the median iPhone 13 Pro result, so clearly thermal throttling plays an important role here.
 
E
Andropov said:
Maybe this is it: testing a single test (to avoid thermal throttling), I can get quite close to the median iPhone 14 Pro results: I got 20.6 fps in 4K Aztec Ruins (High Tier) Offscreen (median iPhone 13 Pro: 15.8 fps, median iPhone 14 Pro: 23fps).

The result I got the first time (running all tests) in 4K Aztec Ruins (High Tier) Offscreen is 11.5fps, even worse than the median iPhone 13 Pro result, so clearly thermal throttling plays an important role here.
Click to expand...
See, any improvement you see on iPhone 14 Pro is due to the new thermal design and LPDRR5 not because of a new GPU design.
 
exoticspice1 said:
See, any improvement you see on iPhone 14 Pro is due to the new thermal design and LPDRR5 not because of a new GPU design.
Click to expand...
Could be, but I'll still wait for proper reviews to come. Just as my 4K Aztec Ruins (High Tier) Offscreen got a low score the first time I ran it (because I ran all tests at once), the leaked iPhone 14 Pro results could be lower for the same reason. That'd also explain why there's not much of a difference between the median and best scores while in other results they're huge.
 
I dislike gfxbench, 3dmark wildlife is better imo, but has anyone had a chance to run gfxbench on their new iPhone 14 Pro (Max) yet?
 
This is my score.
61DD6CDC-B214-4331-9C0B-D84203B16FA4.png
 
diamond.g said:
I dislike gfxbench, 3dmark wildlife is better imo, but has anyone had a chance to run gfxbench on their new iPhone 14 Pro (Max) yet?
Click to expand...

First screenful of scores attached. Don’t see an easy way to export from gfxbench.
 

Attachments

  • 714C852E-A663-4F1A-9DC0-92C92E8FB291.jpeg
    714C852E-A663-4F1A-9DC0-92C92E8FB291.jpeg
    123.8 KB · Views: 54

Similar threads

Cmaier
Cost of A16
Replies
7
Views
527
Yoused
Yoused
Back
Top