Which Apple MacBook Pro 14 config is right?

[Beau10]

I don't think it'd change things that much, so I suspect it's the RAM that's killing you,

There's not anything the OP is doing that is remotely impacting RAM that much.

Again, not internet strangers - MacRumors own guide on the matter, one more time... with plenty of videos and evidence of how these machines perform under load.

Most likely it's caused by Teams being a runaway process. I've seen this happen several times on my base 14".

16GB vs. 32GB MacBook Pro: How Much RAM is Enough?

Apple's high-end 14-inch and 16-inch MacBook Pro offers the M1 Pro chip with 16GB of RAM as standard, but you can upgrade to 32GB of RAM for...

www.macrumors.com

 

[wilberforce]

I have done it

MacBook Pro 14 | 8 CORE base | 32GB

Got returns till start of Jan so will see if I need the 10 core or not. If I look at my daily stats at the moment I don't really hit over 50% at any point but RAM I do swap and max out at times if screen sharing, on camera etc on MS Teams

MS Teams is the devil < They need to optimise it

It sounds that for your particular use case you have your own stats that indicate more RAM might help but more processor might not, so that seems a good choice.
Your own personal stats and usage is much more important and useful than input from others (including myself) that have different use cases.
The 8/14 core MBP M1 Pro already has 50% more performance cores and 75% more GPU cores than what you have right now, plus a faster SSD, so by increasing RAM to 32GB it should be an overall significant upgrade, I think. This base 14" MBP is faster than my mid-spec 2020 iMac which cost ~$3000.

 

[leifp]

I have done it

MacBook Pro 14 | 8 CORE base | 32GB

Got returns till start of Jan so will see if I need the 10 core or not. If I look at my daily stats at the moment I don't really hit over 50% at any point but RAM I do swap and max out at times if screen sharing, on camera etc on MS Teams

MS Teams is the devil < They need to optimise it

That sounds like the correct compromise from my perspective. Nothing you use is CPU intensive. RAM exceeds present requirements and should hold up well even if you add lots of additional apps/Safari tabs... the GPU enhancement vs M1 is probably the most important.

Since I'm not an expert here either, take my free advice at face value. EDIT: *more accurately: report back whether you keep it or not and why/not*

(I have dual 4K displays on my M1 mini, a couple of dozen tabs in Safari, Mail, Slack, Vivaldi, PDF software (both Preview and PDF Expert), Word, and Excel open all the time. Occasionally Zoom, Teams, etc as well. CPU load: <=50% performance cores, rarely any efficiency cores utilized. RAM load: 3 of 16GB free *at worst*. Until I fire up the photo editing software(s)...) [Measured with iStat Menus]

 

[wowee]

Yes, I know, but I have got up to 22GB of swap usage using Photoshop and Lightroom, with some lagginess on interactive masking, although the processors are not maxed out, - and I don't have the infamous memory leak. So I can understand there could be some benefit in more RAM. Just trying to be honest rather than justify my own purchase decision. If I were to buy it again I would get 32GB RAM, rather than more cores, at least for me personally.

I agree. I think more ram is better than more cores in this case.

 

[wilberforce]

There's not anything the OP is doing that is remotely impacting RAM that much.

Again, not internet strangers - MacRumors own guide on the matter, one more time... with plenty of videos and evidence of how these machines perform under load.

Most likely it's caused by Teams being a runaway process. I've seen this happen several times on my base 14".

16GB vs. 32GB MacBook Pro: How Much RAM is Enough?

Apple's high-end 14-inch and 16-inch MacBook Pro offers the M1 Pro chip with 16GB of RAM as standard, but you can upgrade to 32GB of RAM for...

www.macrumors.com

I think you make a very good point, and this is useful information that helps people not spend unnecessarily on upgrades that make no difference whatsover, but just donate extra cash to Apple.

On the other hand, most of these initial batch of comparison videos tend to repeat each other using pretty much the same benchmarks and usage conditions, so unsurprisingly come to similar conclusions.

Now, however, that the new MBPs have been out several weeks, some people are finding that, under certain conditions, their experiences are not always quite the same. For example, large or many graphical edits, Teams, multiple displays. This is not to detract from the value of these comparison videos (in avoiding unnecessary expense), but can provide additional perspective, but hopefully not just additional confusion.

I would like to say: the 14-day Apple return period is fantastic, and people should take advantage of it more intensively. Rather than maxing out upgrades due to uncertainty and doubt: take your best guess at a more moderate upgrade, and try it for a week. Then you will know much better if you need to upgrade further. Unless cost is not very important, in which case go for it.

 

[Beau10]

Now, however, that the new MBPs have been out several weeks, some people are finding that, under certain conditions, their experiences are not always quite the same. For example, large or many graphical edits, Teams, multiple displays. This is not to detract from the value of these comparison videos (in avoiding unnecessary expense), but can provide additional perspective, but hopefully not just additional confusion.

Has there been any evidence that these issues are due to RAM constraints? Benchmarks have shown with even heavy page swap there has been minimal measured slowdown.

I suspect the vast majority of the cases it's a runaway process that's borking the machine. The only slowdowns I've experienced on my 14" have been due to this - I monitor using iStat and see this happening frequently, and I think this is particularly a nasty issue for non-ARM code that may be running on your machine. Teams absolutely is because of this and not RAM, I've seen it with my own eyes.

 

[wilberforce]

Has there been any evidence that these issues are due to RAM constraints? Benchmarks have shown with even heavy page swap there has been minimal measured slowdown.

I suspect the vast majority of the cases it's a runaway process that's borking the machine. The only slowdowns I've experienced on my 14" have been due to this - I monitor using iStat and see this happening frequently, and I think this is particularly a nasty issue for non-ARM code that may be running on your machine. Teams absolutely is because of this and not RAM, I've seen it with my own eyes.

I agree with you, actually 100% agree with you, but seeing you asked: "has there been any evidence that these issues are due to RAM constraints," here is a comparison video that does show potential benefit of 32GB RAM.
Now, I realize this is a special niche usage case, and almost hesitate to post this so as not to mislead others or invite (deserved) criticism, but it may be of interest:

TL,DR: quote: "if you are a photographer doing large composites, large files.... get the base, upgrade the RAM, upgrade the SSD, and you are all set. ...showing clear benefits and differences between 16GB and 32GB RAM especially if you are a photographer..."

 

[Beau10]

I agree with you, actually 100% agree with you, but seeing you asked: "has there been any evidence that these issues are due to RAM constraints," here is a comparison video that does show potential benefit of 32GB RAM.
Now, I realize this is a special niche usage case, and almost hesitate to post this so as not to mislead others or invite (deserved) criticism, but it may be of interest:

For sure - anything where you need a very large active working buffer it makes absolute sense. Very large photos/videos, machine learning w/ a large training set, CAD, video game dev, large number of concurrent docker VMs, things like that.

I didn't mean to mean to imply there aren't any use cases, just those use cases are rather extreme, even taking into account most photo and video editing needs that would traditionally require 32 gb or larger, can in fact be done w/ minimal impact. Basically if your memory needs are rather large but granular in nature, memory management and swap speed are crazy good.

The OP is dealing w/ a situation in pretty much the opposite direction.

 

[jonobin]

Don't judge ram usage, it's known that this version of Monterey has some memory leaks

 

[wilberforce]

Don't judge ram usage, it's known that this version of Monterey has some memory leaks

This is a good caution, high memory usage could be due to a Monterey memory leak.

In my case, so far, I have not seen evidence of memory leaks. My memory usage on Adobe apps is repeatable, even after a restart. There are no processes taking unexpected and growing amounts of RAM, nothing I would want to kill.

Now it may be that Adobe has terrible memory management, but that is a different story. I am certainly not going to count on Adobe "fixing" their memory usage in the future.

So at this point, I am judging RAM usage. I don't have anything else to judge.

 

[upandown]

There's plenty of evidence in benchmarks and whatnot that RAM doesn't have quite the impact w/ these machines, in fact, for most workloads, even fairly intense ones that historically benefit from 32 gigs, minimal impact. We're not talking like 10-20% slower, more like 1% if anything even measurable and certainly *way* beyond what the OP is doing.

Even MacRumor's own guide on this matter to generally avoid upgrading RAM unless you get the Max (well - you have no say in the matter then .

16GB vs. 32GB MacBook Pro: How Much RAM is Enough?

Apple's high-end 14-inch and 16-inch MacBook Pro offers the M1 Pro chip with 16GB of RAM as standard, but you can upgrade to 32GB of RAM for...

www.macrumors.com

Yeah this is hard to ignore. Seems pretty clear that the 16gb performs almost identically to the 32gb based on these tests and real world intensive uses. All because the memory system is very different then any Intel system.

 

[Tooney]

I went for the MBP 14 with 32GB RAM and have updated the desk....