iPhone 15 Pro will have blistering performance, claims leaked benchmarks
A benchmark leak claiming to be for the upcoming A17 chip claim that the iPhone 15 Pro will be extremely powerful -- but there are big reasons to be skeptical.

The A17 chip in the iPhone 15 Pro will be fast, but how fast exactly?
As the generations change, Apple's chips get faster and more powerful, which is visualized in benchmark results. However, sometimes the results can be a little too high to be realistic.
One such leak arrived on Monday, in a post by "@Tech_Reve" republishing details from Weibo. The information was apparently Geekbench scores for the A17 chip, the next step in Apple's mobile chip series expected to land in the iPhone 15 Pro models.
According to the tweet, the A17 running in Geekbench 6 scored 3,019 points in the single-core test, and 7,860 on the multi-core version.
While the A17 is expected to be on the new 3nm process with speed and efficiency improvements, these numbers still seemunrealistic, at least when compared to scores of previous A-series generations.
For a start, the 3,019 score is over 500 more than the current-top A16 Bionic chip in the iPhone 14 Pro, a 20.6% jump in performance year-on-year. However, looking at the A14 against the A13 and the A13 against the A12, the jumps are 10.8% and 10.6% respectively.
The multi-core score has the same issue, but is a little bit higher at a 24% year-on-year jump, while the A16 improved 16.3% and the A15 17.4%.

Geekbench browser scores for the Pro models of iPhone, comparing A-series chips against the rumored figures.
While you would expect a performance boost each year, the amount doesn't usually jump as much as the rumored figures propose. Apple's typical 10% jump is still considerable for its chips, but 20% would be a bit of a stretch.
Add in that this is a rumor based on an unverified Weibo post with no real proof or even a Geekbench screenshot to support it, and it becomes less believable as rumors go. Based on history, it's also too early for engineering validation test iPhone 15 Pro models to be available for testing in such a manner.
Furthermore, we weren't able to find any entries in the Geekbench database that matched these results.
That's not to say Apple may come out with an A17 that is extremely powerful for this generation. AppleInsider has discussed at length Apple's chip gains over the years, and a 20% single-core performance jump has happened in the past.
Read on AppleInsider

The A17 chip in the iPhone 15 Pro will be fast, but how fast exactly?
As the generations change, Apple's chips get faster and more powerful, which is visualized in benchmark results. However, sometimes the results can be a little too high to be realistic.
One such leak arrived on Monday, in a post by "@Tech_Reve" republishing details from Weibo. The information was apparently Geekbench scores for the A17 chip, the next step in Apple's mobile chip series expected to land in the iPhone 15 Pro models.
According to the tweet, the A17 running in Geekbench 6 scored 3,019 points in the single-core test, and 7,860 on the multi-core version.
While the A17 is expected to be on the new 3nm process with speed and efficiency improvements, these numbers still seemunrealistic, at least when compared to scores of previous A-series generations.
For a start, the 3,019 score is over 500 more than the current-top A16 Bionic chip in the iPhone 14 Pro, a 20.6% jump in performance year-on-year. However, looking at the A14 against the A13 and the A13 against the A12, the jumps are 10.8% and 10.6% respectively.
The multi-core score has the same issue, but is a little bit higher at a 24% year-on-year jump, while the A16 improved 16.3% and the A15 17.4%.

Geekbench browser scores for the Pro models of iPhone, comparing A-series chips against the rumored figures.
While you would expect a performance boost each year, the amount doesn't usually jump as much as the rumored figures propose. Apple's typical 10% jump is still considerable for its chips, but 20% would be a bit of a stretch.
Add in that this is a rumor based on an unverified Weibo post with no real proof or even a Geekbench screenshot to support it, and it becomes less believable as rumors go. Based on history, it's also too early for engineering validation test iPhone 15 Pro models to be available for testing in such a manner.
Furthermore, we weren't able to find any entries in the Geekbench database that matched these results.
That's not to say Apple may come out with an A17 that is extremely powerful for this generation. AppleInsider has discussed at length Apple's chip gains over the years, and a 20% single-core performance jump has happened in the past.
Read on AppleInsider
Comments
At the higher end of an improvement this couuuld be believable with 3nm offering a big improvement over a few generations of 5nm, but this is also that timeframe of a lot of fake benchmarks, even if one shows up in the GB database sometimes it's later taken down as a fake and I don't see this one on there even.
By the way, TSMC is noting it has a 10-15% performance improvement over N5, and .58 the area density, so it's a decent node jump but not the biggest size decrease ever and the fab names are mostly just marketing, 7nm had 30-50nm fin gate lengths and pitch lengths lol
https://www.anandtech.com/show/18727/tsmcs-3nm-journey-slow-ramp-huge-investments-big-future
people are acting as it the 3nm jump (which is a massive deal in itself) is the only upgrade to A and M series this gen. It’s not.
If this score is real (or even close) they’re going to have a meltdown.
I'm not completely sure about this, as it seems reasonable to invest in chip improvements, and when they are available, use them as widely across the entire product line. Why have assembly lines for multiple products when one can make all the necessary components for all products? But as other have commented, I don't see this minor upgrade as a selling point to upgrade from 14 to 15. That said...it *is* a selling point for me, as I will be going to the 15 from my current Xr. I expect a noticeable upgrade in overall performance...even if I am not making movies or photoshopping on it.
3000 GB6 single is a wonderful score. It's a great score for a phone, a tablet, a laptop, and a desktop. Also a great score for a server, but single core performance isn't that important for servers.
I know people want to "WIN!" We've already won in terms of CPU performance. Just slow and steady. 10% to 15% every year is fine. Where people need wins are probably 16 GB of RAM base (don't give them a choice of staying at 8 GB), and low latency caches and memory. An order of magnitude increase in random storage reads/writes, so lots of room here. Probably 128 GB base storage on phones, 512 GB base storage on Macs. Lastly, 100 MByte/s upstream and downstream Internet for everyone.
Downloaded Geekbench 6 and my iPhone 13 Pro is now 2,275 for single, and 5,536 for multi. Wonder why they changed the scale to make the numbers higher, when version 4 to 5 actually adjusted the scale for lower numbers.