Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

First M2 benchmarks

Pressure

Pressure

macrumors 603
May 30, 2006
5,180
1,544
Denmark
MayaUser said:
the gpu is 50% better or even double in some instances...from my understanding Minecraft is fully M1/M2 compatible
On pc side i never accounted almost double the performance from 1 generation to another
Click to expand...
Only the newest version of Minecraft (1.19) uses the AArch64 Java version, unless you already are running something like MultiMC.
 
EugW

EugW

macrumors G5
Original poster
Jun 18, 2017
14,899
12,872
13" MacBook Pro comparison: Under load, the M2 fan sometimes becomes noticeably noisy unlike the M1 fan.

 
Pressure

Pressure

macrumors 603
May 30, 2006
5,180
1,544
Denmark
EugW said:
13" MacBook Pro comparison: Under load, the M2 fan sometimes becomes noticeably noisy unlike the M1 fan.

Click to expand...
A new cooling solution like in the larger MacBook Pro models are probably coming in the future.

Remember that they haven't really changed anything on this model compared to the M1.
 
T

theorist9

macrumors 68040
May 28, 2015
3,880
3,060
JimmyjamesEU said:
If these testers had access to Spec cpu 2017, it probably would push the cpu. Alas it costs $1200 I think.

I’ll have to go back and watch again, but I don’t recall Max showing cpu freq for other tests.
Click to expand...
Pretty close, it's actually $1000. Which seems relatively little for any site that does regular benchmark testing and has a decent viewership, like Linus Tech Tips, CNet, Mobile Tech Review, MacWorld/PCWorld, Ars Technica, etc. So it's surprising that anandtech seems to be the only site (or at least the only English-language site I know of) that uses it.

1656572928441.png
 
M

mr_roboto

macrumors 6502a
Sep 30, 2020
856
1,866
theorist9 said:
Pretty close, it's actually $1000. Which seems relatively little for any site that does regular benchmark testing and has a decent viewership, like Linus Tech Tips, CNet, Mobile Tech Review, MacWorld/PCWorld, Ars Technica, etc. So it's surprising that anandtech seems to be the only site (or at least the only English-language site I know of) that uses it.
Click to expand...
Running SPEC is not just a matter of paying $1000 and pushing the go button - it's supplied in source form. It's up to you to get it compiled, and even doing that takes considerable care as you have to look into the best optimization flags for the purpose of the runs you're going to do. Full SPEC runs also take considerable time to complete.

Most tech press simply doesn't have the technical expertise in house to figure out how to compile SPEC, much less do a competent job of running it.
 
  • Like
Reactions: T'hain Esh Kelch, theorist9, hans1972 and 2 others
L

leman

macrumors Core
Oct 14, 2008
19,521
19,677
mr_roboto said:
It's up to you to get it compiled, and even doing that takes considerable care as you have to look into the best optimization flags for the purpose of the runs you're going to do.
Click to expand...

Which by the way is also the reason why one has to take great care interpreting SPEC results, especially if they come from an interested party. The ability to choose the optimal optimisation flags is great for supercomputers where you own the software and the system and can pretty much tailor it to whatever you run it on. But for the general customer we get absurd situations where Intel runs SPEC on it's benchmark-rewriting compiler (while penalising codegen on competitor hardware), each subtest using a different set of optimal flags, so you get these phenomenal crazy high scores, but the practical relevance for precompiled real-world software the users will see is zero.
 
  • Like
Reactions: theorist9, CWallace, JimmyjamesEU and 2 others
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom