8-Core 14-Inch MacBook Pro Around 20% Slower Than 10-Core Models in Multi-Core Benchmark

[ProfessionalFan]

Anyone getting the base, binned-chip models of these is nuts.

Well I guess I’m nuts then. I’ll add it to the list of things I’ve been called here the past two weeks for purchases I made. I’m a fool, an idiot, a nut, and I need help. Keep them coming, everyone.

Considering I plan to get the next revision, getting the base seemed wise. Especially considering the current M1 MBP has never been taxed in the year I’ve had it.

 

[Taz Mangus]

Tomorrow is the day we get to see all sort of benchmarks and tests showing the capability of the M1 Pro and the M1 Max.

 

[tbirdparis]

I’m more interested to find out how general efficiency pans out. With Logic Pro X beavering away, connected to a large external monitor, the last thing you want is fans whirring away to get rid of heat.

This has been the hated bugbear of Logic Pro users since forever - the dreaded fans at full speed the second you start using an external display (or more than one). The reason this has always been the case on Intel MacBook Pro's comes down to the dedicated GPU's inside those machines, with their high-wattage demands and all the heat they generate. It seems pretty likely that this is about to become a thing of the past with these new machines.

Also, even aside from the far lower power use and heat generation of Apple Silicon in general, I think it's no accident they've thickened up the chassis on these new machines even if it's not by much. Given what a disaster they had with thermals and throttling on the previous design since 2016, it looks like they're taking no chances whatsoever to make sure there's plenty of headroom for the chips to heat up under load and still not have to push the cooling system hard to deal with it. After all, they've set an expectation now with the first M1 machines that the fans (on the ones that even have them at all) are barely audible even when the machines are pushed hard. So it wouldn't look good at all if the upper-tier MacBook Pro's run hot and loud and yet are premium models. I guess we'll see in the next few days when the reviews and tests start coming out. But I'd be very surprised indeed if there's fan noise anywhere near as bad as the outgoing Intel models they're replacing.

 

[SimonQ]

Naw, save your money for next year I'd wait till you actually feel the need it in your current system. This is the beginning and Apple will probably work out some kinks in this model like they did in your 2017 over my 2016's issues If you're really wanting to upgrade, get the 10 but the same money will get you a bit more performance in a year when it counts compared to your current system.

Thank you

 

[mjtomlin71]

Doubtfull.

A "pure" M2 MBP (replacing the 13" M1) makes no sense and it seems that Apple wants the Macs on a 2 year schedule.

So some more M1??? (Mini, big iMac) in spring followed by M2 consumer HW (based on A16??) in the 3rd quarter and M2Pro/Max stuff in 23.
All skipping the odd-numbered Axx designs.

Could be wrong would like to be wrong, but they introduced A14 (M1) based SoCs 1 month after the A15.....

I’d guess two more new M1’s (Ultra and Extreme) during WWDC. M1 Ultra will debut in the new iMac ”pro” released then. And the M1 Extreme in the Mac Pro released at the end of the year.

In the Fall the M2, based on A16 cores.

 

[mjtomlin71]

Accoring to these my own calculations, the 14" 10 core cpu verison, has a 58% higher geekbenck score (12700) - than
the Macbook Pro M1 13" (7382). Valid gainz imho.

That’s actually a 72% increase.

 

[Fomalhaut]

Astonishing that the M1Max seems not to be faster than the M1Pro.

Why? The CPU parts of the SoCs are the same. It's the GPU that changes (and maximum RAM capacity due to the larger SoC package)

 

[2adOpin2]

It seems to me that lots of people who are ordering maxed out M1 Max machines and think that they will run cool and without fan noise while using the laptops to the max will be in for disappointment as they don't seem to understand basic physics. That a maxed 16 in requires a 140 watt charger should be a clear indication that if pushing the CPU and GPU hard while running multi monitors, the laptop will be drawing more than 100 watts so it will at least be a 100 watt heater. Of course it will get hot and the fans will need to run hard under those conditions. It will be a laptop doing desktop level of work. Even with the efficiency of Apple Silicon, there will still be heat. Hopefully, the fans will be less loud than the 2019 world and won't spin up simply if a monitor is attached. But that heat isn't going to be disappearing by magic. What I'm hoping for is more power yet something somewhat better noise and heat wise compared to a maxed 2019 model. I won't be complaining if I can settle for that.

 

[mjtomlin71]

The only point where M1 Max/Pro appears to shine is performance on battery, which many professional users rarely use. You can also experience less noise and less thermal throttling compared to mobile high-performance i9 but other than that, I expect similar performance to high-end mobile workstations, based on initial benchmarks. We will see if there is any sub-2,500$ PC laptop on the market that can achieve better CPU performance. Then, you have GPU performance which comes near, but does not match mobile RTX 3080 according to Apple itself. Again you have less noise and less thermal throttling due to the efficiency of the M1 but definitely the new MBPs do not blow out of the water mobile existing workstations from Lenovo, Dell and other major OEMs. The efficiency of Apple Silicon is definitely something very exciting when it comes down to iPhone/iPad/MBA, but for devices like the iMac, or even high-end MBPs, Im sure many people are ready to sacrifice 3-4 hours of battery life, and some fan noise more frequently, but to have vastly superior performance compared to Windows-equivalent laptop. 20 hours battery life is an overkill. It is good to have a few hours of batter life for meeting, while traveling and such, but two-days battery life for a laptop is absolutely not necessary, especially when you sacrifice so much potential performance.

Bottom line:
M1 Pro/Max are not better chips than Intel's best mobile CPU in terms of raw performance. M1 Pro/Max are not better chips than Nvidia's best mobile GPU in terms of raw performance. With M1 Pro/Max you gain incredible battery life, less noise and less thermal throttling, and of course MacOS optimization, hardware acceleration for certain task and shared memory which may be quite useful for certain workflows. These pros alone are in my opinion a little disappointing. M1 is definitely the best ultrabooks CPU/GPU and outperform any low-voltage CPU/integrated GPU on the market, but the M1 Pro/Max do not appear to do the same in their respective categories, which are mid-tier performance laptops, and high-end mobile workstations. You still have the MacOS/Apple Silicon gaming handicap, software compatibility issue across platforms, and other cons that are hard to ignore.

You should’ve stopped at the first line, because that’s exactly what Apple set out to do. Create chips with the best performance per watt there is. All those other metrics you talk about are pointless. No one expected Apple to outperform the top of the line Intel or AMD CPUs in raw power. Especially in a first generation design. However, they do outperform anything Intel or AMD have today that can be put inside these systems with their thermal constraints.

 

[DeepIn2U]

I'm curious if the die is actually different on this model or if it's a chip binning strategy where one or two of the cores are defective.

I'm more curious why all of these have the same single-core performance and the reasoning behind keeping it virtually the same giving 6 months of deveopment and production has passed. Is there any process and apps in MacOS 11.x of all of Apple's applications or any major third party applications where the single-core performance improvements will not be beneficial by chip improvements??

If not, even for Windows or linux why is Geekbench even still having this as a measurement of chip performance?

 

[SeaFox]

The 14" with the 10/16 M1pro and 1 TB SSD is the same price as the base 16" (with 10/16 M1pro and 500 GB SSD).
Unless you really want a 14" size and storage it seems like it makes more sense just to buy the 16" and get more screen. You also get the 140W power adapter thrown in instead of the 96W.

 

[EugW]

I'm more curious why all of these have the same single-core performance and the reasoning behind keeping it virtually the same giving 6 months of deveopment and production has passed. Is there any process and apps in MacOS 11.x of all of Apple's applications or any major third party applications where the single-core performance improvements will not be beneficial by chip improvements??

If not, even for Windows or linux why is Geekbench even still having this as a measurement of chip performance?

? The single-core chip performance is the same for all these chips is because because the core is the same core.

The reason single-core is measured is because it assesses the speed of that core design. M2 will likely have a slightly faster single-core speed, probably in the range of 1900 or so I'm guessing.

 

[wilberforce]

nvm

The 14" with the 10/16 M1pro and 1 TB SSD is the same price as the base 16" (with 10/16 M1pro and 500 GB SSD).
Unless you really want a 14" size and storage it seems like it makes more sense just to buy the 16" and get more screen. You also get the 140W power adapter thrown in instead of the 96W.

I agree, unless you want a smaller machine (as you say). I really like the size of my 2015 13" MBP, and the new 14" M1 MBP is identical in overall dimensions. Previously I had a larger (17") laptop and hated it (heavy and cumbersome).

One of the really nice things about these M1 MPBs is you can get the same performance from either size, so it simply comes down to a personal size preference. Previously, the 13" Intel MBPs had much worse performance than the 16" Intel MBPs, so one felt punished by wanting the smaller machine, as though that was the "cheap" MBP.

 

[ikramerica]

? The single-core chip performance is the same for all these chips is because because the core is the same core.

The reason single-core is measured is because it assesses the speed of that core design. M2 will likely have a slightly faster single-core speed, probably in the range of 1900 or so I'm guessing.

Yeah, but in 1 year no progress has been made? That’s the thing bothering me. Not even a “turbo” mode enabled on a streamlined M1e/+ core.

Looking back at the powerpc transitition, a lot of progress was made in 12 months. Then again the first 601 wasn’t very good, and the M1 is very good.

 

[Christopher Kim]

I agree, unless you want a smaller machine (as you say). I really like the size of my 2015 13" MBP, and the new 14" M1 MBP is identical in overall dimensions. Previously I had a larger (17") laptop and hated it (heavy and cumbersome).

One of the really nice things about these M1 MPBs is you can get the same performance from either size, so it simply comes down to a personal size preference. Previously, the 13" Intel MBPs had much worse performance than the 16" Intel MBPs, so one felt punished by wanting the smaller machine, as though that was the "cheap" MBP.

Yep - as someone with the 13" MBP the last 5 years, I've come to really like this size. Compared to my wife's 2015 15" MBP, it really feels more portable and easy to take around.

I thought I was going to get this year's 16", but now that it's no longer the 13" being worse performance than the 16", but it's really more like iPhone 13 Pro & Pro Max or iPad Pro 11 / 12.9. Internals the same, so you can optimize for what size you prefer.

 

[Christopher Kim]

Yeah, but in 1 year no progress has been made? That’s the thing bothering me. Not even a “turbo” mode enabled on a streamlined M1e/+ core.

Looking back at the powerpc transitition, a lot of progress was made in 12 months. Then again the first 601 wasn’t very good, and the M1 is very good.

I think many ppl suspect that this chip is still "based on the A14" eg. M1 from last year, as it wasn't supposed to come out this late. Was supposed to come out less than 1 year after the M1 debut (I suspect at least before the A15-based iPhone 13 Pro came out). And because of the COVID-related delays, we're still getting this now, even more than 1yr later

 

[businessnumbersmoneypeopl]

nvm

I agree, unless you want a smaller machine (as you say). I really like the size of my 2015 13" MBP, and the new 14" M1 MBP is identical in overall dimensions. Previously I had a larger (17") laptop and hated it (heavy and cumbersome).

One of the really nice things about these M1 MPBs is you can get the same performance from either size, so it simply comes down to a personal size preference. Previously, the 13" Intel MBPs had much worse performance than the 16" Intel MBPs, so one felt punished by wanting the smaller machine, as though that was the "cheap" MBP.

Another way to look at this is that the 14" base model with 32gb memory and 1Tb is only 100$ more expensive than the 16" base model with 16gb and 512gb. And yet, the multicore geekbench score of the base model is twice as high as the 2020 Intel flagship MBP – more than fast enough for most people, also in the future.

 

[yitwail]

Yeah, but in 1 year no progress has been made? That’s the thing bothering me. Not even a “turbo” mode enabled on a streamlined M1e/+ core.

Looking back at the powerpc transitition, a lot of progress was made in 12 months. Then again the first 601 wasn’t very good, and the M1 is very good.

That’s a narrow view of progress. M1 Pro and Max have numerous new features, like LPDDR5 ram, 200/400 GBs bus, video encode/decode acceleration, multiple monitor support, additional usb-c ports, 2x/4x gpu cores, etc. Apple wants to improve “real world” usability as well as benchmark scores.

 

[ikramerica]

I think many ppl suspect that this chip is still "based on the A14" eg. M1 from last year, as it wasn't supposed to come out this late. Was supposed to come out less than 1 year after the M1 debut (I suspect at least before the A15-based iPhone 13 Pro came out). And because of the COVID-related delays, we're still getting this now, even more than 1yr later

That’s a valid hypothesis. We are seeing such delays in automobiles, appliances, etc.

I believe that’s why the mini pro is delayed. With limited production output in chips, the MBPs took priority.

 

[ikramerica]

That’s a narrow view of progress. M1 Pro and Max have numerous new features, like LPDDR5 ram, 200/400 GBs bus, hardware video encode/decode acceleration, multiple monitor support, additional usb-c ports, 2x/4x gpu cores, etc. Apple wants to improve “real world” usability as well as benchmark scores.

Yeah? At a huge increased cost.

Progress in the chip world is delivering more at the same price point, not just charging more for more features a year later.

At minimum the M1 should have been replaced with a + with access to more memory or at a higher clock or something meaningful.

 

[EugW]

Yeah, but in 1 year no progress has been made? That’s the thing bothering me. Not even a “turbo” mode enabled on a streamlined M1e/+ core.

Looking back at the powerpc transitition, a lot of progress was made in 12 months. Then again the first 601 wasn’t very good, and the M1 is very good.

AFAIK, these chip designs are effectively three years old. There is no “1 year progress” in this regard. It doesn’t work that way. Significant modifications done now would show up in chips in 2024 or whatever.

The intent was to have the same A14 core powering the entire line, including iPhone, iPad, MacBook Pros, iMacs, and Mac Pros, but not all at once. As Apple said, they expect a 2 year transition period, and we’ve only just hit the halfway mark. The second half of the transition will include bigger iMacs, the souped up Mac mini, and the Mac Pro, all next year, again powered by the same core.

At the same time though, M2 is also coming, and so likely will M2 Pro/Max and derivatives eventually.

Mind you, I personally have been ahead of the game since 2017. My MacBook runs m3. ?

 

[yitwail]

Yeah? At a huge increased cost.

Progress in the chip world is delivering more at the same price point, not just charging more for more features a year later.

At minimum the M1 should have been replaced with a + with access to more memory or at a higher clock or something meaningful.

It’s a pro machine. LPDDR5 ram isn’t cheap, and neither is a mini LED 14” screen that varies from 24hz to 120hz. Not too many base machines out there that start with 32mb ram either, so you upgrade like I did. Heck, Amazon has $50 discount on base 14” and Adorama might have a bigger discount, though you’d have to settle for 16mb, so with patience you can narrow the price gap.

 

[michelb76]

It's a smart line-up. The base model is not great, and probably has lesser chips. The top model is 20% faster but only 15% more expensive than the middle model, thus a no-brainer.. Strictly speaking, it makes little sense to buy anything other than the top model.

 

[JippaLippa]

With ALL the on-chip SRAM & DDR PHY Controller enhancements Apple R&D has added into even the low-end M1 Pro, it is, IMO, an absolutely killer CPU !

My ONLY complaint is the cost of obtaining it !

Will need to wait for the M2 MacBook Pro, out mid-2022 ?

That should be ~20% slower (performance-wise) than the low-end of the M1 Pro.

OK by me, for the price differential ! ... $1.3K vs $2K

If priced right, & NOT performance-crippled in ANY way, the M2 MacBook Pro could be the single-best-selling Mac ever !

The macbook pros will always have the pro/max variants of the chips, and I totally expect the 13" being discontinued soon.

If you’re not a “power user,” I’ll bet the regular 13 inch M1 would be fine for you

Agreed.
I got the M1 Max 10/32 with 64GB as my workflow is very resource intensive (motion graphics design), however a friend of mine who makes music and does some light photoshop work is desperate for a replacement for his 2014 mbpro, but has very little money (relatively speaking).
I suggested a 16GB 8 core macbook air, as it'll be plenty fast to drive a Daw, and it costs way less than what an M2 model, with the liquid retina XDR display, will end up being.

 

[businessnumbersmoneypeopl]

It's a smart line-up. The base model is not great, and probably has lesser chips. The top model is 20% faster but only 15% more expensive than the middle model, thus a no-brainer.. Strictly speaking, it makes little sense to buy anything other than the top model.

The base model is twice as fast as the former Intel flagship model. Few people ever need such power now or in the foreseeable future. Apple really made a major leap with its in-house SoCs. The logic of going for the top model was a no-brainer in the Intel era but not so much anymore (for many users, anyway). If one has extra money, I would rather invest it in memory or SSD space.