Going from 2019 base 16" macbook pro to base 14" macbook pro... worth it or a true upgrade?

[jazzneel]

Hi,

I currently have a 2019 base 16" macbook pro, and was thinking about going to a 14" macbook pro base model, but unsure if this is really an upgrade or not overall (from a performance perspective).

It would be nice to have a smaller laptop (and the additional ports), but I do want a true upgrade as well. It looks like the m1 pro CPU is much better/faster than my i7, but the 14 core gpu is about on par with my 5300m. Is this accurate or am I missing something? I do play some games like starcraft 2, other various items so it would be nice to have something that runs better (but I know intel vs apple silicon is an issue too).

Thanks!

 

[ahurst]

Hi,

I currently have a 2019 base 16" macbook pro, and was thinking about going to a 14" macbook pro base model, but unsure if this is really an upgrade or not overall (from a performance perspective).

It would be nice to have a smaller laptop (and the additional ports), but I do want a true upgrade as well. It looks like the m1 pro CPU is much better/faster than my i7, but the 14 core gpu is about on par with my 5300m. Is this accurate or am I missing something? I do play some games like starcraft 2, other various items so it would be nice to have something that runs better (but I know intel vs apple silicon is an issue too).

Thanks!

My lab manager did a similar upgrade a few months back (2019 16" MBP to 2021 16" M1 Pro MBP) after the former spontaneously died. He's not a power user in the traditional sense (no heavy photo/video/3D work or anything), but he says it's been a revelation to have a laptop that basically never gets hot or whirrs up the fans every time he fires up a Teams meeting, and that the battery life is massively improved.

Also, as the current owner of a 14" MBP, I'd say the screen is going to be a nice upgrade over the older MBP as well: being able to watch movies/shows in native 24p without any judder is very slick, and the Mini-LED panel looks fantastic.

 

[PauloSera]

Hi,

I currently have a 2019 base 16" macbook pro, and was thinking about going to a 14" macbook pro base model, but unsure if this is really an upgrade or not overall (from a performance perspective).

It would be nice to have a smaller laptop (and the additional ports), but I do want a true upgrade as well. It looks like the m1 pro CPU is much better/faster than my i7, but the 14 core gpu is about on par with my 5300m. Is this accurate or am I missing something? I do play some games like starcraft 2, other various items so it would be nice to have something that runs better (but I know intel vs apple silicon is an issue too).

Thanks!

The old machine gets white hot under a small amount of load, sounds like its going take off at any second, and is dead after 45 minutes on the battery.

The new machine is cool to the touch no matter what you do with it, is entirely silent, and lasts all day on the battery.

There really is no comparison. It is a gigantic upgrade. You're focused on the wrong points.

 

[LinkRS]

Hi,

I currently have a 2019 base 16" macbook pro, and was thinking about going to a 14" macbook pro base model, but unsure if this is really an upgrade or not overall (from a performance perspective).

It would be nice to have a smaller laptop (and the additional ports), but I do want a true upgrade as well. It looks like the m1 pro CPU is much better/faster than my i7, but the 14 core gpu is about on par with my 5300m. Is this accurate or am I missing something? I do play some games like starcraft 2, other various items so it would be nice to have something that runs better (but I know intel vs apple silicon is an issue too).

Thanks!

Howdy jazzneel,

Not sure what you mean by "true upgrade," but I made a similar switch just about a year ago. Went from a CTO 2019 16" MacBook Pro with i9 2.3 GHz, 16GB RAM, 1TB SSD, and the 8 GB 5500m upgrade, to a CTO 2021 16" Macbook Pro M1 MAX with 32GB RAM, and 1 TB SSD. To be honest, from a performance perspective, I didn't notice a huge performance upgrade from my older i9. In some circumstances it seems a little quicker, but about the same for most things. If I was able to leverage the encoders and the NE in the MAX, I know that would be a huge upgrade, but my typical workload does not. I was hoping that iMovie would take advantage of the M1 MAX, but it didn't make much difference.

Do not take this as a negative, or even a critical review, just stating my personal experience with this. Things that are significantly better than the older model are battery life and unplugged performance. These two things alone almost make me think it was worth the upgrade. I have never been able to use my MBP unplugged for fear of battery life, and most significantly performance down turn when away from the wall kept me from doing it often. With the 2021 MBP, performance is exactly the same plugged in or on battery. Under this circumstance, you will notice a performance improvement.

Lastly I want to briefly mention app support and games....... For Blizzard games, with the exception of World Of Warcraft, still run on OpenGL. Apple's OpenGL drivers are terrible, and this made games using it run with graphical errors and lower than expected performance. This is a known issue, and not new. Since Apple decided to drop OpenGL and use Metal, they just seemed to stop updating the OpenGL stack. This is a "hidden" side-benefit to M1-based Macs. Since M1 Macs do not run OpenGL natively, when an OpenGl application is started, it is translated (I assume by Rosetta 2?) to Metal to run natively on Apple Silicon. This means that in some cases your OpenGL games will run better on the M1 Macs, than on Intel Macs. At worst, I have seen the same performance between both systems, with the added benefit that there is less graphical corruption on M1.

App support is another potential gotcha. Before you make the plunge and upgrade, check and make sure all of your apps will work. Ideally, they will have been updated to Universal binaries at least. Rosetta 2 is great, and runs well, but not the ideal situation.

Good luck!
Rich S.

 

[hardax]

I did this exact upgrade and dont regret it. the simple fact that the 2019 was a jet engine every time I even thought about doing some higher end Acrobat work (Im in the desktop publishing business and do extensive PDF editing) made this a worthy upgrade. Day to day app performance seem on par but everything is snappier and dead silent.

 

[Beau10]

Hi,

I currently have a 2019 base 16" macbook pro, and was thinking about going to a 14" macbook pro base model, but unsure if this is really an upgrade or not overall (from a performance perspective).

It would be nice to have a smaller laptop (and the additional ports), but I do want a true upgrade as well. It looks like the m1 pro CPU is much better/faster than my i7, but the 14 core gpu is about on par with my 5300m. Is this accurate or am I missing something? I do play some games like starcraft 2, other various items so it would be nice to have something that runs better (but I know intel vs apple silicon is an issue too).

Thanks!

I had a fully maxxed 2019 16" from work and got a base 14" and the 14" absolutely obliterates it for my use, which is primarily programming - the extra speed and reduced heat is off the charts. I don't game though and would think keeping w/ Intel would be more flexible, and if not, looking at at least the lower end max option to get a real boost in that regard.

That said, unless you're getting it at a deep discount I'd probably wait a bit, rumor mill has indicated recently that the M2 machines are coming next quarter.

 

[jterp7]

before bfriday microcenter had high spec 14" models for 7-900 off. But we're probably close enough for you to wait and see at this point

 

[wyrdness]

My work Macbook is a 2019 2.6 GHz 6-Core Intel Core i7, whilst my personal laptop is a M1 Macbook Air.
I'd swap this roasting hot (and thermally throttled) Intel machine for Apple Silicon in an instant.

 

[LinkRS]

I had a fully maxxed 2019 16" from work and got a base 14" and the 14" absolutely obliterates it for my use, which is primarily programming - the extra speed and reduced heat is off the charts. I don't game though and would think keeping w/ Intel would be more flexible, and if not, looking at at least the lower end max option to get a real boost in that regard.

That said, unless you're getting it at a deep discount I'd probably wait a bit, rumor mill has indicated recently that the M2 machines are coming next quarter.

Howdy Beau10,

I am curious about your experience with the 14" M1 with programming. I too use my MacBook Pro primarily for programming, using a mixture of Unreal Engine, Xcode, VS Code, Visual Studio, and VIM (when I have to ). I discount the speed of compilation as it is Apples and oranges (pun intended) compiling ARM code and X64 from my perspective. XCode performs the exact same for me, as does VS Code and VIM. Unreal Engine does not run as well, it was already slow in the 2019 MBP, but having to go through Rosetta makes it that much worse. I am kind of stuck until (if?) Epic embraces Apple Silicon support. What has improved so substantially for you? Thanks!

Rich S.

 

[GrumpyCoder]

Did this and went from a maxed out 16" Intel (64GB, 2TB) to a 16" M1 Max (24GPU cores, 64GB, 2TB). As far as performance it's pretty much a draw. Went 24 GPU core for a little longer battery life. I gained much longer battery life with low and average loads. When running under full load CPU+GPU the battery life of AS Macs drops just as fast, don't kid yourself. You get about one hour, give or take (see below linked video). Noise level on the same utilization level is much lower. During normal browsing/reading/writing both Intel and M1 Max are silent. The Intel ramps up fans much quicker. The M1 Max isn't always silent though, start stressing CPU+GPU simultaneously it becomes easily audible sitting on a desk, but nowhere near Intel noise levels. In such cases, the 14" is also louder than the 16". Am I happy with it? Yes. Would I upgrade again? Yes. If you pick the 14" or 16" is up to you. The 14" handles a little better being more compact (weight isn't the problem), as a result it has a little less battery life, it's clocked down compared to the 16", it's noisier under heavy CPU+GPU load. Most users won't experience this as they never stress it high enough. The rest is pretty much equal, ignoring the obvious display size difference. That being said, I don't think you can go wrong with either.

Unreal Engine does not run as well, it was already slow in the 2019 MBP, but having to go through Rosetta makes it that much worse. I am kind of stuck until (if?) Epic embraces Apple Silicon support.

UE5.1 supports AS, you can't directly download it though. You have to build it from scratch with Xcode. That being said, Intel via Rosetta and AS native perform about equal. The Intel version has more low level FPS dips, avg. and max frame rate is about the same on both. Nanite technically works on the Intel/Rosetta version, but crashes frequently (practically it's unusable). Nanite doesn't work at all on the native AS version. Lightmass works on the Intel/Rosetta version, it does not work on the native AS version, so you won't be able to bake lightmaps.

The 5.1 AS version is experimental, see https://docs.unrealengine.com/5.1/en-US/unreal-engine-5.1-release-notes/ ini the Platform section. If you can live with this, build 5.1 AS and enjoy.

 

[LinkRS]

UE5.1 supports AS, you can't directly download it though. You have to build it from scratch with Xcode. That being said, Intel via Rosetta and AS native perform about equal. The Intel version has more low level FPS dips, avg. and max frame rate is about the same on both. Nanite technically works on the Intel/Rosetta version, but crashes frequently (practically it's unusable). Nanite doesn't work at all on the native AS version. Lightmass works on the Intel/Rosetta version, it does not work on the native AS version, so you won't be able to bake lightmaps.

The 5.1 AS version is experimental, see https://docs.unrealengine.com/5.1/en-US/unreal-engine-5.1-release-notes/ ini the Platform section. If you can live with this, build 5.1 AS and enjoy.

Thanks! Last time I checked, the AS support was "coming soon." Haven't bothered to check in awhile. Was considering checking out Unity, but not sure I want to take that route. I much prefer C++. Appreciate the link!

 

[Beau10]

Howdy Beau10,

I am curious about your experience with the 14" M1 with programming. I too use my MacBook Pro primarily for programming, using a mixture of Unreal Engine, Xcode, VS Code, Visual Studio, and VIM (when I have to ). I discount the speed of compilation as it is Apples and oranges (pun intended) compiling ARM code and X64 from my perspective. XCode performs the exact same for me, as does VS Code and VIM. Unreal Engine does not run as well, it was already slow in the 2019 MBP, but having to go through Rosetta makes it that much worse. I am kind of stuck until (if?) Epic embraces Apple Silicon support. What has improved so substantially for you? Thanks!

Rich S.

I don't have the 16" anymore (work upgraded to a 16" M1), but I did when I initially purchased the 14". I used it primarily for React dev. Things like running our test suite which pushes the CPU to the limit nearly halved in time and doing a build with rollup was significantly faster. It also *seemed* like VSCode was a fair bit more responsive. The Intel machine was regularly in a stutter/locked up type state when really pushed that the 14" seems to suffer less from, possibly from heat soak... and not having a toaster in the lap anymore is quite nice.

Maybe obliterate is too strong of a term, but the overall experience is notably better in day to day going from a maxxed out -> base model so I was quite happy with the change, and esp. more so when my work upgraded the intel to the M1. I tend to only use the 14" for personal programming tasks now.

 

[LinkRS]

I don't have the 16" anymore (work upgraded to a 16" M1), but I did when I initially purchased the 14". I used it primarily for React dev. Things like running our test suite which pushes the CPU to the limit nearly halved in time and doing a build with rollup was significantly faster. It also *seemed* like VSCode was a fair bit more responsive. The Intel machine was regularly in a stutter/locked up type state when really pushed that the 14" seems to suffer less from, possibly from heat soak... and not having a toaster in the lap anymore is quite nice.

Maybe obliterate is too strong of a term, but the overall experience is notably better in day to day going from a maxxed out -> base model so I was quite happy with the change, and esp. more so when my work upgraded the intel to the M1. I tend to only use the 14" for personal programming tasks now.

Thanks for your reply! I very rarely used my 2019 MBP away from the wall, as when I did, performance was simply terrible outside simple productivity type activities. I've not used React myself, but have experience with AngularJS, but could see how that would be faster on the M1. Appreciate the response!

 

[GrumpyCoder]

Was considering checking out Unity, but not sure I want to take that route. I much prefer C++.

We're using both for different projects. At least on the Mac, while Unity is not quirk and error free, it's a better developer experience in my opinion... for now. You won't get the same graphics quality from the engine as with UE, but UE is gimped on the Mac right now. Everything equal, M1 Max has about half the FPS as a laptop 3070Ti in Windows/Linux. The sky is the limit, we're running 4090s in desktops now. The rest is personal preference.

 

[Beau10]

Thanks for your reply! I very rarely used my 2019 MBP away from the wall, as when I did, performance was simply terrible outside simple productivity type activities. I've not used React myself, but have experience with AngularJS, but could see how that would be faster on the M1. Appreciate the response!

For sure... fwiw I wasn't aware is a difference between battery/plugged aside from the profile settings (any links to how the perf profile changed? googled a bit and couldn't find anything) but I did use it almost entirely on battery. It was close to 1000 cycles (and 76% life!) in about 2 years before relinquishing.

 

[LinkRS]

For sure... fwiw I wasn't aware is a difference between battery/plugged aside from the profile settings (any links to how the perf profile changed? googled a bit and couldn't find anything) but I did use it almost entirely on battery. It was close to 1000 cycles (and 76% life!) in about 2 years before relinquishing.

Hi Beau10,

Are you referring to the M1-based MBP? The only behavior that changes when not-plugged in is how long it takes before it goes to sleep or dims the screen. All of those are under Settings. The CPU/GPU perform the same regardless of plugged-in state. That is why it feels so much faster than the 2019 when unplugged. If you your use is mostly unplugged, the M1-based MBP will be a much better experience, across the board .

 

[Beau10]

Hi Beau10,

Are you referring to the M1-based MBP? The only behavior that changes when not-plugged in is how long it takes before it goes to sleep or dims the screen. All of those are under Settings. The CPU/GPU perform the same regardless of plugged-in state. That is why it feels so much faster than the 2019 when unplugged. If you your use is mostly unplugged, the M1-based MBP will be a much better experience, across the board .

No, I meant the prior intel version.

 

[LinkRS]

No, I meant the prior intel version.

The Intel version throttles the CPU down as far as possible, depending on the model to as low as 800Mhz (called Intel SpeedStep https://en.wikipedia.org/wiki/SpeedStep), as well as lower screen brightness and tries to put the storage disk to "sleep" aggressively. In the case of Macs with discrete graphics, it would also disable the dGPU in favor of the iGPU which can take framerates. Some applications (such as Unreal Engine) force the dGPU to be on at all times, regardless of the plugged-in state, causing the battery life to suffer.

 

[theorist9]

The old machine gets white hot under a small amount of load, sounds like its going take off at any second, and is dead after 45 minutes on the battery.

The new machine is cool to the touch no matter what you do with it, is entirely silent, and lasts all day on the battery.

There really is no comparison. It is a gigantic upgrade. You're focused on the wrong points.

Nope, he specifically said he was interested in performance. Nothing wrong with telling him about other benefits of the new machine, but it's just typical internet obnoxiousness when someone asks about X and someone presumes to say that's wrong, they should instead be interested in Y. Folks should show enough respect for posters to answer what they ask, like the others above have done.

 

[Beau10]

The Intel version throttles the CPU down as far as possible, depending on the model to as low as 800Mhz (called Intel SpeedStep https://en.wikipedia.org/wiki/SpeedStep), as well as lower screen brightness and tries to put the storage disk to "sleep" aggressively. In the case of Macs with discrete graphics, it would also disable the dGPU in favor of the iGPU which can take framerates. Some applications (such as Unreal Engine) force the dGPU to be on at all times, regardless of the plugged-in state, causing the battery life to suffer.

Oh yeah, I remember messing with this w/ a third party app back in the day to shut off the GPU and played around with the clockrate. Aside from GPU switching as needed to preserve battery, was there anything performance related that Apple actually did (did they actually downclock when on battery)?

 

[LinkRS]

Oh yeah, I remember messing with this w/ a third party app back in the day to shut off the GPU and played around with the clockrate. Aside from GPU switching as needed to preserve battery, was there anything performance related that Apple actually did (did they actually downclock when on battery)?

SpeedStep and even AMD has a similar technology for their GPUs (but I don't remember what it is called?), both downclock the CPU/GPU whenever the system is idling to preserve battery life. Under Windows, this was exposed in power settings, and you could tweak this behavior, but I do not recall being able to do this from the interface in macOS. I *think* the Intel PowerGadget app that you can download for macOS allows some tweaking as well, but the default behavior is based in firmware and results in the CPU being aggressively throttled when running on battery. Intel used to advertise that newer version of SpeedStep had more incremental frequency steps, and could clock up and down very rapidly to the point that you wouldn't notice. However, I noticed. Responsiveness is more laggy, and battery life suffers when unplugged. I *believe* based on experience that Apple made the most aggressive versions of SpeedStep default on battery Mac systems. This is one of the reasons that Apple Intel Macs had much better battery life than comparable Windows laptops (other reasons are better optimizations of background tasks, and better control of what can run in the background). Conversely, some performance oriented Windows laptops use the least aggressive settings of SpeedStep by default. Granted regardless of the firmware defaults, the OS can override these settings to some extent. I would not be surprised if you could do some aggressive tweaking in the terminal for Intel SpeedStep settings. Hope this answered your question?

 

[PauloSera]

Nope, he specifically said he was interested in performance. Nothing wrong with telling him about other benefits of the new machine, but it's just typical internet obnoxiousness when someone asks about X and someone presumes to say that's wrong, they should instead be interested in Y. Folks should show enough respect for posters to answer what they ask, like the others above have done.

Nah, providing actual wisdom is more important than answering the wrong questions. Not everyone knows enough to ask the right questions.

 

[theorist9]

Nah, providing actual wisdom is more important than answering the wrong questions. Not everyone knows enough to ask the right questions.

But it's not the wrong question. Performance is important to this poster. Compare your answer to, for example, posters #4, #5, and #10, all of whom show the capability and decency to answer the poster's question about performance (not much difference when plugged, significant difference on battery), explain the main benefit is noise and heat, and do it in a friendly, collegial manner (i.e. without talking down to them). That's real wisdom.

Too often people on this forum are more interested in using it as a platform to attempt to show they are superior to others rather than providing actual help.

 

[Beau10]

But it's not the wrong question. Performance is important to this poster. Compare your answer to posters #4, #5, and #10, all of whom show the capability and decency to answer the poster's question about performance (not much difference when plugged, significant difference on battery), explain the main benefit is noise and heat, and do it in a friendly, collegial manner (i.e. without talking down to them). That's real wisdom.

Too often people on this forum are more interested in using it as a platform to attempt to show they are superior to others rather than providing actual help.

What about my responses, which directly address perf (though CPU not GPU bound) and day to day improvement in user experience in a professional context due to said perf? I even provided an example of a benchmark that was nearly halved, and that was against i9, not an i7. It's also not evident that plugged vs. unplugged would have any bearing. It's real wisdom to not cherry pick to support a stance.

Actual benchmarks show significant improvements in many areas - those are factual and repeatable. But of course that wont translate into day to day for everyone, and yes, GPU tasks are going to be conditionally sideways unless one upgrades to a Max variant.

 

[theorist9]

What about my responses, which directly address perf (though CPU not GPU bound) and day to day improvement in user experience in a professional context due to said perf? I even provided an example of a benchmark that was nearly halved, and that was with an i9 not an i7. It's also not evident that plugged vs. unplugged would have any bearing.

Actual benchmarks show a bloodbath - those are factual and repeatable. But of course that wont translate into day to day for everyone.

Ah, sorry, I missed yours. I was just pulling a few examples. Your post was of course fine as well . And interesting, because it shows that, while many don't see a significant difference in performance, on certain specific tasks AS can be strikingly faster.

I tested a key program use, Mathematica, and found there's not a lot of difference in performance between a 2019 i9 iMac and an M1 Max. Here I created my own benchmark to compare my iMac with AS, corresponding to the types of computations I routinely do, and asked some posters I knew from another forum to run it on their M1 Max's. [These tasks are all single-threaded.] You can see it averages only 2% to 18% faster for suites of symbolic computations (first table) and 21% faster for graphing. I also gave it an image-processing task, and it was 46% slower.

So I suppose the real take-home message is that the only way to tell whether a machine is significantly faster for your use case is to test it with your actual workflow.