M2 Pro 10 core vs M1 Pro 10 core

[jdb8167]

In all honesty, I would expect both E and P cores to have some improvements. I believe that the M2 MBA runs a bit hotter and uses more power then the M1, so we have that to go on.

As far as I can tell the M2 uses the same power as the M1. If it is higher I can’t tell when doing the exact same tasks on the M2 that I did on the M1.

As for hotter I’ve always been confused by how people determined that. If someone is using a laser temperature meter against the case, how does the change in materials modify the results?

If they are using one of the menu bar temperature monitors, how do they know that the internal temperature sensors are measuring the same thing as the M1’s sensors. The names all changed and the sensors are completely undocumented.

To get back to the previous point, it seems to be all speculation.

 

[Alex W.]

I have a MBP with the M1 Pro chip, but I am within the 2 weeks return window.

Did you get it cheaper? A deal?

Honestly, return it and get the newer one if its the same price -- if you got a deal on the refurbished unit or something keep it.

M2/M1 is very similar 20% in CPU perf (best case) isnt much same with the GPU.

 

[Sagnet]

Did you get it cheaper? A deal?

Honestly, return it and get the newer one if its the same price -- if you got a deal on the refurbished unit or something keep it.

M2/M1 is very similar 20% in CPU perf (best case) isnt much same with the GPU.

I bought the M1 Pro at full price, which is $3100 (tax inc) where I live. The M2 Pro is $3250 (tax inc). The costs are covered by my employer though, so it is not a personal cost for me.

Again, performance is not what's most important to me. Thermals, power consumption and battery life are equally important.

 

[Alex W.]

I think the m1 pro maxed CPU/GPU load will last longer than an M2 pro maxed out from what I understand; claims of 1hr better battery are on lower power applications; I believe the M2 Pro pulls more maxed out to make its claims with 20%/30% better perf on the same Nm process even if it is matured.

Honestly, man, if work paid for it, get the better unit. If it isn't a huge issue with them but if it causes a mess or looks bad, keep the PRO. It's a great machine and will last for a long, long long time (they're also identical units screen and design-wise).

 

[jdb8167]

I think the m1 pro maxed CPU/GPU load will last longer than an M2 pro maxed out from what I understand; claims of 1hr better battery are on lower power applications; I believe the M2 Pro pulls more maxed out to make its claims with 20%/30% better perf on the same Nm process even if it is matured.

Sure. But the M2 will also do more work than the M1 during that time. If you are using the GPU flat-out the difference in performance is more like 40% rather than 20%-30%. CPU is less. About 12%-19%. So for most tasks, the CPU/GPU get back to low power faster on the M2 for the same task ultimately saving power.

 

[itsboi]

What's $150 if you're spending $2k+ on a laptop? M2 Pro will obviously be more appealing down the line in terms of resale. The M1 Pro is a 15 month old chip, lacks the upgraded Wifi, BT, and HDMI + has less GPU cores. By those Geekbench scores, they're similar in multi-core but the M2 Pro has the edge in single. Power consumption and battery life differences are probably negligible.

 

[Sagnet]

I think the m1 pro maxed CPU/GPU load will last longer than an M2 pro maxed out from what I understand; claims of 1hr better battery are on lower power applications; I believe the M2 Pro pulls more maxed out to make its claims with 20%/30% better perf on the same Nm process even if it is matured.

Honestly, man, if work paid for it, get the better unit. If it isn't a huge issue with them but if it causes a mess or looks bad, keep the PRO. It's a great machine and will last for a long, long long time (they're also identical units screen and design-wise).

What's $150 if you're spending $2k+ on a laptop? M2 Pro will obviously be more appealing down the line in terms of resale. The M1 Pro is a 15 month old chip.

It's actually $3k+. The M2 is only better for me, if it doesn't come with more heating and fan noise than the M1.

 

[haralds]

Most of our systems loaf most of the time. After getting my Mac Studio M1 Ultra last year, I decided not to look for a few years to avoid core-envy. In general use, it will be hard to tell the incremental improvements.
This one was replacing a fully decked-out 2012 Mac Pro 3GHz with some upgrades (Radeon Graphics, PCie M2 SSD.) But for most uses, I cannot really tell the difference. Exception are Xcode compiles and Logic Pro.
There there is that Cemu WiiU emulator running under Rosetta2. That does at all not perform as well on the ten year old box.
I also do not miss the lights dimming, when turning on the Mac Pro.

 

[kzly]

@Sagnet, what did you end up with?

 

[Sagnet]

@Sagnet, what did you end up with?

The M2 Pro, and have not looked back.

 

[Basic75]

I think the two additional efficiency cores of the M2 Pro will help lower energy load compared to the M1 Pro if the system is not taxed.

In which scenarios are 2 E-cores not enough AND 4 E-cores enough AND the system doesn't decide to use the P-cores anyhow because of the applications'/processes' QoS settings?

 

[quarkysg]

In which scenarios are 2 E-cores not enough AND 4 E-cores enough AND the system doesn't decide to use the P-cores anyhow because of the applications'/processes' QoS settings?

I think the question would be is the E-cores good enough that there's no need need of P-Cores most of the time. I frequently see all 4 E-cores used and P-cores dormant for my M1 Mini with only Safari active. With E-core sipping power even at full tilt, it makes sense to have more of them, especially for battery powered devices.

 

[Sydde]

With E-core sipping power even at full tilt, it makes sense to have more of them, especially for battery powered devices.

From what I see, it looks like the M2 E cores are around 10% faster than the M1 E cores.

 

[Sumo999]

As a frame of reference, one Einstein on the Forum concocted a Mathematica routine as a benchmark for Intel and then had someone run it on an M1. He concluded in the name of science that Apple Silicon performance claims are bogus, despite never having used a silicon computer. Conclusion: this kind of stuff is sort of pointless with regard to real world use. I can just honestly attest that Logic on a plain M1 machine runs crazy rings around my Intel box, no contest.

 

[Chris_99]

The M2 Pro, and have not looked back.

Hi Sagnet, since you got the M2pro 10/16 .....can you share your user experience? about fan noise? thermals? heating? battery life for light and battery life for heavy usage? what is your overall assessment ? Iam also looking into this 14 inch M2 10/16 sod with 32gb ram and 1/2 TB storage. so I would love to hear about your first hand user experience .
What are the temperatures on the cores for you? Are you fully satisfied ? or would you spec something different now?
Thanks you!
Chris