Dr. Watson, 15 Sep 2023Didn't Apple themselves pretty much say that there was only a 10% performance improvement... moreBut still what ?
Switching to the new 3NM node should grant 25-30% increase in performance, AT EQUAL CLOCK SPEEDS.
Apple does the switch to 3NM, increases clock speed by 0.38GHz on the performance cores and gets 10%. That's lame to say the least.
Something's going on.
Anonymous, 15 Sep 2023All for cameras, capcut, faceberg, and tiktok. those can handle without any problems. but gaming on an iphone.. is not always ideal. drains like 28nm processor😅😅
Anonymous, 14 Sep 2023One UI use here. No any slowdown. Works pretty good on my midranger.prove if your samsung midranger could run latest one ui without any hiccups and lags. or just call it running like if it was bought new.
Huge power wasted by limited OS. Enjoy your walled garden.
Anonymous, 15 Sep 2023You are missing the point of this die shrink. You can go big and bigger like Intel or nvidia c... more"people here seems they cannot understand this is BIG...only tech people that understand silicon architecture"
YES! It quite clear you did a good job educating us, but with all these (such a power house) the owner of a 15PROMAX device only gets to enjoy 40% of this, I'm pointing to the lock down OS. It like an old man using a power housed Ferrari
Samsung uses a SOC less weaker (some says inferior) to Apple products but with the heavy tasks and day to day functionality they are able to maintain and sometimes they even keep up with Apple power house *machines
That been said there are scenarios were the Apple power housed soc loss to compete Android soc(even a yr old Gen soc frm Qualcomm by then)
And when we think that Intel and AMD haven't yet had to confront this; their transistor density is a third or so that of Apple's so their wiring is much less dense. Of course the price they pay for that is larger chips and much hotter.
potato4k, 15 Sep 2023Hopefully there will be a tech site that will do a deep dive into the cores. At this point, th... moreYou are missing the point of this die shrink. You can go big and bigger like Intel or nvidia cards when your SoC families must enter into passive phones, ipads and macs. This is a true step forward that they can increase by 10% SC and 20% gpu while go back to the smaller SoC size WHILE:
- has SUBSTANTIALLY MORE transistors at 19 billion. That's 19% more than A16.
- runs at slightly LOWER power of ~7 watts versus ~8.5 watts on A16
- increases clock speeds on the P-cores by 10%.
- P-cores have wider decode & execution units, improved branch prediction.
- GPU cores have improved efficiency, mesh shaders, and dedicated hardware ray-tracing silicon.
- has an additional GPU core for a total of 6 cores vs 5 in the A16.
- doubles the performance of the neural engine to 35 TOPS
- adds AV1 decode support
- adds USBC controller
people here seems they cannot understand this is BIG...only tech people that understand silicon architecture
Bigmeme, 14 Sep 202310% improvement in performance is pretty disappointing. The process node upgrade alone should ... moreand what will they do next year on iphone 16 when architecture will be the same if they unleash full power now?
they have to reserve some power for next year too and honestly all cpus from flagships these days are powerful enough for just about everything, companies should focus in efficiency for more batter life than benchmark gimmicks
My first comment here was under traffic by Gsmarena team (somehow I don't understand Y) but I woke up this morning seeing it has been published with 2 big red cross from Apple defenders.
I'm sorry to say "Apple defenders" but it's funny to see someone opposing your opinion without a reply(just an emoji remark) showing the level of Fanboyism and too childish such is an act.
Whenever I compliment a brand (to this set of persons) they are like "He must be a Fanboy"
Anonymous, 15 Sep 2023Why is there always a troll in this comment section? The 8 Gen 2 was already toe to toe with t... moreyou have not to know architecture to insult someones opinion
This A17 has now AV1 decoder and the best it has hardware Ray tracing in the gpu itself with its own co-processor and its own cache
This cannot be rivalled for at least 2 years..not to mention Apple bought all N3 for 2024
I expects this COMMENT SECTION to pull up heavy arguments than this but it's seems fan Boys afterall are learning from facts these days
Afterall,we all are Consumers, you won't be paid for Defense or been bais towards people's opinions.
Anonymous, 15 Sep 2023Even then, Qualcomm will take at least 2 years to reach this level of performance. As for Exym... moreWhy is there always a troll in this comment section? The 8 Gen 2 was already toe to toe with the A16, even beating it in GPU performance. There's a good chance the 8 Gen 3 will straight up beat this
taylorswift13, 14 Sep 2023the thing is new Samsung mid-rangers (with new I mean A52, A33 and A24 and above in their gene... moreJ series is officially dead. Now we have lower A series instead. My neighbor had A01. Slow phone but worked. I'm using A52 now. Still works great. Will buy flagship next.
Imaself, 15 Sep 2023Apple CPU'S Power are only to be seen on Applebench™ Then, AnTuTu Will give the lead f... moreBut then each iPhone model will come out with better efficiency in the real world because it always does and Antutu becomes meaningless
I can't wait in 2 to 3 years time for this chip to be in a base model ipad
Anonymous, 15 Sep 2023Bruh apple is lucky they had large $$$ paying tsmc so that tsmc prioritized apple over android... moreLucky? Building up the wealth that Apple has wasn't sheer luck. They've competed well on the phone market for 16 years now. It's not luck
Chidoro50, 15 Sep 2023Uhh, what price hike. There hasn't been one since the X which was 6 years agoThe X launched in India at 89000, 15 Pro launches at 134900. There is world outside US you know?
Ki124, 14 Sep 2023A17Pro is clocked at 3.78GHz while A16 was 3.4GHz
so on identical clocks the difference will... moreApple is deliberately showing modest upgrades or apple is really struggling to make bigger improvements.
Even though most likely a typo, it would be easier to make a case of the non capitalisation of "a" in the Apple
Apple Bionic A15 base was 2x 3.23 Ghz Avalanche and 4x 2.02Ghz Blizzard 5nm process with 5 GPUs increased by 1 since A14 Bionic
Apple Bionic A16 base was 2x 3.46 Ghz Everest and 4x 2.02Ghz Sawtooth 4nm process with 5 GPUs
This year the so far unnamed A17
Apple Bionic A15 base 2x 3.78 Ghz (****) 3nm process with 6 GPUs increased by 1 since A16 Bionic
The efficiency core clock speed A15 and A16 both 2.02Ghz
Apple apple has once past re-used carried forward the same efficiency core if you looked at clock speed only
This is not applicable when there is a change in process
Just by an analysis of the past increases which have little or no bearing on the future, my indefensible maths only has 4x 2.28 Ghz
The argument is restricting increases when the benchmark so far ahead of Android makes commercial sense
There are more knowledgeable people than me who will go over the actual figures when posted
Anonymous, 15 Sep 2023SD695 is light years ahead in battery life. only if used as a portable hotspot.