Anyone spending $3000+ on a laptop probably isn't penny pinching while they are looking at upgrades. Also, someone buying that 2.4 probably knows they need all the performance they can get, including extra RAM.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
16" MBP Geekbench scores so far show big differences
- Thread starter intrepidcase
- Start date
- Sort by reaction score
Anyone spending $3000+ on a laptop probably isn't penny pinching while they are looking at upgrades. Also, someone buying that 2.4 probably knows they need all the performance they can get, including extra RAM.
@intrepidcase Here's a comparison of all 16 inch models to corresponding 15 inch models.
@leman Thanks for the idea to use violin plot, it's fun!
Some information about dataset:
@leman Thanks for the idea to use violin plot, it's fun!
Some information about dataset:
- 16" i7 and i9 - search term: "MacBookPro16,1", grouped by grouped by i7-9750H, i9-9880H and i9-9980HK
- 15" i7 - search term: "MacBookPro15,1", filtered by i7-9750H
- 15" i9 - search term: "MacBookPro15,3" (Vega models only), grouped by i9-9880H and i9-9980HK
Last edited:
So the 2.3 i9 in the 16 inch just about underperforms in any case? That's a bit weird, isn't it?
My MBP 16 2.4 GH i9 32 GB 2 TB arrived today.
I got these test numbers:
Got these results:
Single Core: 1087 (worse than the 1138 for the MBP 15 with same processor)
Multi Core: 7235 (better than the 6894 for the MBP 15 with same processor and the Mac Pro 2013
Open CL Score Intel 4657
Open CL Score AMD 27440
BlackMagic Drive Scores:
Write: 2836.4 MB/s
Read: 2461.2 MB/s (not a typo)
I got these test numbers:
Got these results:
Single Core: 1087 (worse than the 1138 for the MBP 15 with same processor)
Multi Core: 7235 (better than the 6894 for the MBP 15 with same processor and the Mac Pro 2013
Open CL Score Intel 4657
Open CL Score AMD 27440
BlackMagic Drive Scores:
Write: 2836.4 MB/s
Read: 2461.2 MB/s (not a typo)
Most likely since these are standard preconfigured models that reviewers got + one can buy in store. Folks with custom orders only start receiving them about now. I am sure we will see many more benchmarks soon. I'll update the data (also polling in the 15" 2019 model) at the end of the week.
Front page is claiming a 16% increase from 2015 to 2016 🙄
www.macrumors.com
16-Inch MacBook Pro Hands-On: Benchmark and Performance Tests
Apple's new 16-inch MacBook Pro came out last week, and the new machine features a larger display, more maximum RAM, and higher maximum storage,...
www.macrumors.com
So what does that mean? 2.3 i9 not so bad after all?Front page is claiming a 16% increase from 2015 to 2016 🙄
16-Inch MacBook Pro Hands-On: Benchmark and Performance Tests
Apple's new 16-inch MacBook Pro came out last week, and the new machine features a larger display, more maximum RAM, and higher maximum storage,...
It means the front page is misleading people based on a comparison of one 15” to one 16” when the graphs on this page clearly show there is not difference when compared across multiple machines.
Just statistics. Every machine varies so if you only compare 1 machine to 1 machine you might randomly find some difference. But you shouldn’t represent that as ‘all comparable machines must therefore be like this’
I don’t know how reliable this uploaded data is. The issue is, we don’t know the testing methodology.
Completely true. But it’s real world data is another way to look at it. And I’m still believing a few thousand results showing little difference over a single result showing 16%
What I mean, these are brand new machines. People might be running the benchmark while there is still some indexing going on etc. I think the score distribution might change if we give it a couple of weeks.
Sure lets see! You could rerun your script and stratify by date even! But my guess is alot fo these machines are blank and have little indexing going on.
The problem is Geekbench doesn't really heat up the machine so the difference between the 2019 15inch and 2019 16inch should be minimal at best. Now with a longer test like cinebench where thermal could come into play it may well be 16 percent better. I would much rather see a graph plotting cinebench results over Geekbench.
Last edited:
This.The 16" is much better with sustained loads = faster over real world tasks.
And the person who commented that the GPU wasn't a big deal, yeah it's not a RTX 2080 but it's still very fast compared to what we had. It's about the speed of a D700 (single).
My Vega 64 in the iMac Pro is still over 2X faster (4 teraflops vs 11), but in real world use editing 4K video, FCPX is performing similarly on both machines. So not sure what is going on there. The iMac Pro can probably handle more streams of 4K with effects at once.
Bottom line this is the first Apple laptop that can reliably handle real-time 4K editing and playback without rendering with multiple effects applied (to one stream). The older AMD GPU just couldn't do it.
I have found this image posted on here in the past but it is only one persons results so not much of an average but should give some idea if someone could run cinebench on their 16inch.
I have found this image posted on here in the past but it is only one persons results so not much of an average but should give some idea if someone could run cinebench on their 16inch.
I was talking about Cincbench results. It you lookf back in the tread its chocked full of GB results in more detail than that!
Some additional Cinebench testing:
https://iphone.appleinsider.com/art...h-macbook-pros-thermal-management-to-the-test
https://iphone.appleinsider.com/art...h-macbook-pros-thermal-management-to-the-test
The 2.3 seems to have something strange going on though. The distribution doesn't really overlap on the high end the way the 2.4s do. It really seems to me that the 2.4 behaves about as you'd expect under this theory, with a little bit more "mass" in the violin plot towards the bottom end, but with a higher median and similar peak. The 2.3s though are just lower, across the board.