Future proofing, what I personally think of it.

[aevan]

Not true. Year over year new MacOS releases (with new features) and new software increase demand on processor, memory, and storage.

By your thinking, Apple should still be producing 4GB base memory models.

I purchased 16GB back in 2013, today I'm going with 32GB as demands have increased significantly despite me not changing my workflows.

No, they haven’t - in practical terms. True, macOS demands increase, but not in ways you’d notice.. macOS Big Sur basically runs the same as OS X Mavericks on my wife’s 2013 MBP.

I’ve been seening literally the same arguments on these forums about RAM for the past 8 years.

“Future proofing” is just a justification method for buying more spec than most people need. And those who do need it, know exactly why they need it and for what workflows and it’s not for future proofing, it’s for what they’re doing now.

 

[CarbonCycles]

I view Apple and Intel based chips makers like auto manufacturers. Apple is the Toyota of the auto world (note I did not say Ferrari, Bentley, etc. as there are very boutique makers out there that cater to this line specifically). Everyone else I consider the Chevys and Fords.

Apple holds incredible resale value because they don't have endless discounts or end of year sales to entice customers in to the stores, which ultimately devalues EVERYONES car. Rather, they have a very slow upgrade cycle, TYPICALLY (emphasized as their quality seems to be iffy lately) very well built/reliable, and introduce new products with substantial changes in technology (e.g., new hybrid engine used on the Tundra TRD Pro truck that was recently introduced)...very much like Toyota's business model.

With that said, unless you're in a very high computational or dynamic work load situation, it doesn't make sense to "future proof" your laptop because technology is morphing at exponential rates with new architectures, cameras, biometrics...software (driven by consumer choices) is pushing hardware in ways that are unpredictable. I wouldn't be surprised if we see a quantum processor for commercial use in the next 15 years...we can all speculate how that would change the landscape.

Rather, it makes more sense to purchase a system that slightly exceeds your base usage be it web browsing, publications, accounting, etc.and pocket the extra cash for that next refresh cycle that has the latest technology that will satisfy your needs at that time. There is nothing worse than running tasks on 5 year old hardware that takes 25x more time to complete because you assumed the tasks you were doing 5 years ago will be the same today. Btw, I have personally fell into this "future-proofing" mindset when I maxed out a 15" MBP many years ago...total waste of money. When that system failed, I purchased the first 13" MBP base configuration using the ARM architectures to get me by, which it did very nicely as long as the software could be recompiled. Fast forward to this week and now I'm in a better financial situation to splurge a bit on the new 16" MBP, which satisfies 85% of my needs. The other 15% (what you guys could call "future proofing") could be addressed by spending a bit more, but I just consider it a waste of money as the next refresh cycle will address that.

Flip side comment: I would completely urge you to future-proof (i.e., "hedge") your purchase if you are using your laptop for LOCAL development work that requires dynamic load that are very bursty or sustained processing at very high clock speeds (video/photo processing, machine learning, software engineering, etc).

Anyway, TLDR version: Think of your laptop as a car purchase...it helps frame the reference a bit more.

 

[ascender]

No, they haven’t - in practical terms. True, macOS demands increase, but not in ways you’d notice.. macOS Big Sur basically runs the same as OS X Mavericks on my wife’s 2013 MBP.

I’ve been seening literally the same arguments on these forums about RAM for the past 8 years.

“Future proofing” is just a justification method for buying more spec than most people need. And those who do need it, know exactly why they need it and for what workflows and it’s not for future proofing, it’s for what they’re doing now.

I was having this exact discussion yesterday. For most people, if they need 16GB+ for their workflow, they'll already know.

There's obviously no harm in speccing as much RAM as you can afford if budget is no option, but if you look back to the previous-gen MBP, when it launched and for every revision since, people have been saying things like "16GB is going to be the minimum in 3-5 years time as macOS and programmes are going to get more demanding and bloated". Well, here we are 5 years later and 8GB is still more than enough for lots of people. I'd say that with the M1 chip and fast SSDs, entry level machines like the Air and Mini have never been as capable as they are now.

 

[babyexercise]

I view Apple and Intel based chips makers like auto manufacturers. Apple is the Toyota of the auto world (note I did not say Ferrari, Bentley, etc. as there are very boutique makers out there that cater to this line specifically). Everyone else I consider the Chevys and Fords.

Apple holds incredible resale value because they don't have endless discounts or end of year sales to entice customers in to the stores, which ultimately devalues EVERYONES car. Rather, they have a very slow upgrade cycle, TYPICALLY (emphasized as their quality seems to be iffy lately) very well built/reliable, and introduce new products with substantial changes in technology (e.g., new hybrid engine used on the Tundra TRD Pro truck that was recently introduced)...very much like Toyota's business model.

With that said, unless you're in a very high computational or dynamic work load situation, it doesn't make sense to "future proof" your laptop because technology is morphing at exponential rates with new architectures, cameras, biometrics...software (driven by consumer choices) is pushing hardware in ways that are unpredictable. I wouldn't be surprised if we see a quantum processor for commercial use in the next 15 years...we can all speculate how that would change the landscape.

Rather, it makes more sense to purchase a system that slightly exceeds your base usage be it web browsing, publications, accounting, etc.and pocket the extra cash for that next refresh cycle that has the latest technology that will satisfy your needs at that time. There is nothing worse than running tasks on 5 year old hardware that takes 25x more time to complete because you assumed the tasks you were doing 5 years ago will be the same today. Btw, I have personally fell into this "future-proofing" mindset when I maxed out a 15" MBP many years ago...total waste of money. When that system failed, I purchased the first 13" MBP base configuration using the ARM architectures to get me by, which it did very nicely as long as the software could be recompiled. Fast forward to this week and now I'm in a better financial situation to splurge a bit on the new 16" MBP, which satisfies 85% of my needs. The other 15% (what you guys could call "future proofing") could be addressed by spending a bit more, but I just consider it a waste of money as the next refresh cycle will address that.

Flip side comment: I would completely urge you to future-proof (i.e., "hedge") your purchase if you are using your laptop for LOCAL development work that requires dynamic load that are very bursty or sustained processing at very high clock speeds (video/photo processing, machine learning, software engineering, etc).

Anyway, TLDR version: Think of your laptop as a car purchase...it helps frame the reference a bit more.

Chromebook is Toyota, cheap and you get your work done.

MacBook Pro is Maserati, 1 MacBook Pro can be like 10 or more Chromebook price, it is much faster and more expensive than most people need, and you just love the logo and design. You have enough money to just buy a new model later and feel good anyway, who care future proofing when you have a lot of money to spend anyway.

May be Mac Pro is Ferrari grade.

 

[tdbrown75]

No, they haven’t - in practical terms. True, macOS demands increase, but not in ways you’d notice.. macOS Big Sur basically runs the same as OS X Mavericks on my wife’s 2013 MBP.

I’ve been seening literally the same arguments on these forums about RAM for the past 8 years.

“Future proofing” is just a justification method for buying more spec than most people need. And those who do need it, know exactly why they need it and for what workflows and it’s not for future proofing, it’s for what they’re doing now.

It's not "future proofing" which is a stupid term. It's anticipating needing additional capacity (compute, network, storage, RAM) based on growth projections. If requirements never increase, why does Apple no longer offer a 4GB RAM option as a base model?

It's the same thing as saying "I only use 490GB of storage now, so I'll never use more than 500GB."

And along those lines, why bother upgrading at all? I mean most are not pushing their processors to 100%, so you don't "need" another processor. Can't deny the steady growth of requirements over time, and it's not limited (but does include) memory (speed and volume).

Tim

 

[tdbrown75]

Anyway, TLDR version: Think of your laptop as a car purchase...it helps frame the reference a bit more.

I like this mentality. I drive a Tesla Performance Model 3 that does 0-60 in three seconds. Do I NEED to go 0-60 in three seconds? No. Can I afford it, and do I enjoy the extra speed and performance? Hell yes.

Tim

 

[aevan]

It's not "future proofing" which is a stupid term. It's anticipating needing additional capacity (compute, network, storage, RAM) based on growth projections. If requirements never increase, why does Apple no longer offer a 4GB RAM option as a base model?

It's the same thing as saying "I only use 490GB of storage now, so I'll never use more than 500GB."

And along those lines, why bother upgrading at all? I mean most are not pushing their processors to 100%, so you don't "need" another processor. Can't deny the steady growth of requirements over time, and it's not limited (but does include) memory (speed and volume).

Tim

I’m not denying the steady growth, I’m denying you’ll see much difference in RAM requirements in next 5-6 years. True, Apple doesn’t sell 4Gb configurations any more, but they sold them for more than a decade. One day they won’t sell 8Gb configurations anymore. Then 16. Then 32. But that’s not coming in 5 years.

As for why people upgrade - well, people upgrade for a lot of reasons. There are professions and workflows where a 5% speed increase means more profit. There are enthusiasts that like to run benchmarks. And sometimes, there are groundbreaking leaps like the M1 chips. But it really depends on what you do and what kind of performance do you need.

If 16Gb is enough for you today, chances are - it will be enough for the next five years. If it’s not, then you need more today. And it’s ok to get more than you need, it’s fine. Just don’t try to justify it with talks how “in a few years, you’ll need twice the RAM you have today, even though today you have enough.” No, you won’t.

I like this mentality. I drive a Tesla Performance Model 3 that does 0-60 in three seconds. Do I NEED to go 0-60 in three seconds? No. Can I afford it, and do I enjoy the extra speed and performance? Hell yes.

Tim

That’s a perfectly fine argument. “Wanting” and “needing” is not the same. If you want it - and can afford it - by all means, go and get it. It will be better, for sure. Hey, if I could comfortably afford a fully maxed out 64Gb M1 Max - I’d get it! Just no need to justify it with this “future proofing” nonsense. Because it is, for the most part, nonsense.

 

[tdbrown75]

Just no need to justify it with this “future proofing” nonsense. Because it is, for the most part, nonsense.

Agreed, but that's not an excuse to ignore capacity planning and does not account for architectural changes like a shared CPU/Video memory pool.

Perhaps the real problem is that “future proofing” is a stupid term to use, because we all know there's no such thing.

There are requirements, a steady increase in demand overtime, and a need to understand where you yourself fit on that curve. Same with housing, cars, energy, etc.

Tim

 

[redheeler]

I future-proofed my 16” 2019 MacBook Pro with the Core i9 and 32 GB of RAM two years ago because I didn’t expect Apple to start the transition to ARM chips so soon afterward (the rumors had been circulating for years by that point so it was natural to be skeptical of an exact timeline). Whether future-proofing is a good idea depends on a lot that isn’t really knowable until it’s already too late.

 

[leovilma]

I went with 32GB memory partly for “future proofing” against new OS versions down the road, that may require more memory and run more tasks for added services.

I also went 32GB due to the unified memory model— assumed it would be better to share more memory between CPU and GPU than less. Especially since there are more cores all around…

Maybe some of the above are flawed logic, but that was my basis for the upgraded memory. I upgraded to 10/16 out of a desire to buy what was actually presented. A little underhanded having $1999 models cores not cited in performance comparisons.

I feel the same. But then again 16gb has been standard for 15" pros for 10 years now and I still don't feel I need more. Still I think there will be a sudden leap towards to more memory hungry apps before I plan to upgrade again. So I "need" 32gb and since thats 50% of cost of upgrading to Max, I might as well just go all out. Not going with more than 1TB though, rather have external Thunderbolt drive for that money.

 

[RUGERMAN]

Not so. When OS support for you machine expires, the machine will continue to run on the last supported version, and may do so for many years. Apple does not have a remote kill-switch to disable your machine.

That's true, I have a late 09 iMac that only runs High Sierra but I use it every day and its on 24/7 the only thing I did to it was add 4 g of ram runs fine no problems.

 

[badsimian]

I feel the same. But then again 16gb has been standard for 15" pros for 10 years now and I still don't feel I need more. Still I think there will be a sudden leap towards to more memory hungry apps before I plan to upgrade again. So I "need" 32gb and since thats 50% of cost of upgrading to Max, I might as well just go all out. Not going with more than 1TB though, rather have external Thunderbolt drive for that money.

But that assume the first 50% of that cost has the same value as the second. The Max chip is not twice as fast as the Pro for normal processing. So unless you have very specific workloads which require GPU then I don’t think that jump is worth it. The 16-32 jump in RAM I think is much more valuable to a wider range of people than Pro -> Max.

 

[UnexpectedHuman]

I have been using a 2018 15” MacBookPro with 32GB and discrete GPU with another 4GB. In my normal work setup, the memory is just barely enough that I can run (without paging) Xcode, one VM (Windows or older macOS), one heavy-weight media editor (eg Affinity Photo) and the usual background Safari/Mail/etc. The main problem are the VMs where, for example, myWindows 10 system has 12GB of RAM assigned.

With the new laptops, you do not even have the discrete GPU, so 32GB is potentially even more restrictive. I can easily see why people might want 64GB today, even if VMs are not currently (yet) viable on the new machines.

However you look at it, these machines are way better than my laptop. No hated touch-bar with all the necessary software workarounds to avoid the stupid virtual escape key. No butterfly keyboard. Finally the return of an SD slot and HDMI for when travelling. And decent thermals too, plus a major step up in performance - all at pretty much the same price point. What is not to like?

 

[theluggage]

Perhaps the real problem is that “future proofing” is a stupid term to use, because we all know there's no such thing.

This. My 2011 17" MBP, which served me for about 7 years, did so because a couple of years in it turned out that the "next big thing" in PC performance was the SSD, and it just so happened that it was very easy to swap out the HD for an SSD - which gave it a dramatic new lease of life. No reasonable amount of maxing-out of RAM, HD or CPU options at the time of purchase would have helped had the HD not been replaceable mid-life. (I can't remember if SSD was an option at all at the time, but in 2011 it would have been small and hyper-expensive). Now, user upgradeability is a whole other can of worms, but I don't think it's what this thread is about.

And yes, RAM and storage requirements are still growing but we're talking about minimum viable RAM doubling over 10 years c.f. the good old days when it doubled every 18 months... and typical storage space has probably reduced because of the switch to SSD, which is still more expensive per GB than spinning rust.

I think "present proofing" is a better term for the issue here: you can't guess what your future needs will be but you do need to understand your present workflow at a deeper level than "I do photoshop" or "web development"... and don't get too distracted by the fact that any M1 will probably out perform your 5-year-old Intel laptop - unless your budget is the main constraint, look for the optimum performance now. Analysing your workflow should tell you where diminishing returns will set in.

Budget is a sensitive issue - if you just don't have the money that's end of argument - but otherwise if you're doing anything even remotely professional then your time is probably far more expensive than any computer equipment (apart from wheels, display stands and cleaning cloths) so there's really no point on skimping on hardware.

If your current workflow benefits from 32GB RAM then don't be talked into settling for 16GB: your work may even run faster on a 16GB M1 than on your old Mac but - if RAM is a limiting factor - it will likely run faster still with 32. Likewise, if you're work is GPU intensive then it will likely benefit from a M1 Max - if not you're wasting your money (unless you really need 3 XDR displays). If if your work isn't optimised for serious multi-threading you may not even need an M1 Pro...

 

[dead flag blues]

My idea of 'future proofing' is getting both a new Mac and a new PC every 2 years.

 

[haruhiko]

My idea of 'future proofing' is getting both a new Mac and a new PC every 2 years.

Future proofing is just a nicer term for “I want it”.

 

[dead flag blues]

Future proofing is just a nicer term for “I want it”.

"Just buy the damn thing and get on with it."

 

[Fomalhaut]

I am also mildly irritated by the term "future-proofing", mostly I think, because the concept shows a naive hope that you can stop the clock on the march of technological progress. It's a similar sentiment expressed by people who are offended that a new improved model of something appears within a few months of their own purchase.

Everything you own, even your own body, will get old, die, and turn to dust in time....there is no future-proofing...

What does make sense, is to take a serious look at your current and *likely future requirements*, based on observation of past patterns.

Unfortunately, unlike cloud compute instances that can be resized based on demand, you can't do this with physical hardware, and especially not with Apple hardware.

I do a fair bit of work sizing server machines, and I'm always looking for the sweet spot of performance per dollar, without spending unnecessary amounts. You need to consider your workloads: are they limited by memory, compute power, storage size & bandwidth, network bandwidth, GPU power?

In the Apple world, I think the most important decision is RAM - because you can't change it, and under-sizing can have a big impact on OS and application performance. You can't really have too much, but you can over-spend. I have an old Dell Xeon 6-core workstation that I filled with 64GB of (quite expensive) ECC RAM. I never used it all (even with Windows bloat) and when the RAM failed (yes, all 4 16GB DIMMs!), I replaced it with 32GB because that was the right size for me.

Next is storage, again because you can't change the internal SSD. But you can add very fast external drives now for less money, so it's about having a good plan for your storage "budget" for the life of the machine, and deciding how much of an inconvenience having external storage is. For a very mobile person with large amounts of data, you will probably want more internal storage. Consider also the storage bandwidth - you don't need 7.5GB/s SSDs to play your music collection. A 1990's HDD would be sufficient. Enterprise storage does just this, and has different tiers of storage - maximum speed for OS, applications, and current data sets, slower storage for infrequently used files, and slowest storage (and largest) for archives. Apple's Fusion drives were an example of this for the consumer.

Then comes compute power. I think we're at the point where single-threaded performance is going to be more or less equal whichever configuration you buy, so you need to consider multi-core capacity. This will depend on your applications and workloads. You need to measure current usage and determine whether it would be significantly better with more CPU cores. There is no benefit to having a machine run at 95% idle most of the time, unless you have very variable workloads during which you really need the extra power. (some cloud-compute types take these patterns into account with "burstable" vCPUs that permit short periods of load, but run at much lower averages, for lower cost).

And finally, and possibly most importantly, you need to honestly assess how your needs are likely to change over the expected lifetime of the machine.

Have previous upgrades to your chosen software required more memory to run? Are you working with larger data sets (e.g. bigger video files) that require faster or larger storage?

A final thought is that even if you've done all of the above and logically chosen an optimal configuration based on empirical data, you might still be bothered by the fact that there are newer machines with other benefits (e.g. better screens, cameras, custom processors etc.). You need to realistically assess whether you are a person who enjoys having "the latest thing", or is happy to use their purchases until they fail or are no longer fit for purpose. You may be both, of course - I love having new computers, but couldn't care less about TVs, cars and a whole host of other things...

 

[arvinsim]

Apple should have had a base model of 32GB RAM for the 14". I bet a lot of people would snap that up in store.

 

[ian87w]

Just something I want to throw into the group, and don’t expect everyone to agree with:

I see the term ”future proofing” pop up everywhere again, now that the new MacBooks released.
People looking at the order page, dazzled by the options and prices, seeking for advice on a forum like this and getting talked into ‘this’ and ‘that’ because in some years they might need it.

In my opinion there is no such thing as future proofing by ‘overspeccing’, just in case you might need it later. Get what you need now, or for the foreseeable future and save the remaining cash for an upgrade in 3 years, instead of 5-6 years. Sell or trade-in your current machine and add some extra cash to get the newest thing. In 5 years, your 5 year old Macbook is going to be sluggish anyway, even with the added RAM (you didnt really need, nor utilized in the first place). You honestly think those people that spec’d into a beefed up, “future proof” 2019 MacBook Pro aren’t considering a 2021 model now, with all the fancy bells and whistles?

My opinion is: look at your current situation, you don’t expect it to change anytime soon? Then just get specs accordingly to your personal situation and need case. That said, if money isn’t a thing, by all means get the highest specs you can get and enjoy.

Food for thought perhaps,

Peace out,

Rick

I think it's normal for less techie person to try to get the "best" bang for the buck, considering the high price of Apple products. As such, the idea of "future proofing" to justify the high cost in that "it will be used for longer time." I'm seeing this most often with people who don't actually know much about tech (whether they want to admit it or not). Can't blame them, but nowadays, these same people are also making tech youtube videos and pretend they know what they're talking about, making things more confusing for others. That's why there are so many confusion, even here.

And yes, easiest way is to see the current computer one is using and to analyze how it is sufficient and how it is lacking. That can give a hint on which part to upgrade and put money on, and which part that don't need to be fancy.

 

[shenoka]

I think it's normal for less techie person to try to get the "best" bang for the buck, considering the high price of Apple products. As such, the idea of "future proofing" to justify the high cost in that "it will be used for longer time." I'm seeing this most often with people who don't actually know much about tech (whether they want to admit it or not). Can't blame them, but nowadays, these same people are also making tech youtube videos and pretend they know what they're talking about, making things more confusing for others. That's why there are so many confusion, even here.

And yes, easiest way is to see the current computer one is using and to analyze how it is sufficient and how it is lacking. That can give a hint on which part to upgrade and put money on, and which part that don't need to be fancy.

A lot of generalizations here... What if this isn't our main computer and our work computer is the one we code on? But, we also want something we can code on if we want to.

 

[Sf844]

Agreed. In 5 years the cheap model will run circles around the current top spec model. It makes no sense. Just buy a discounted basic configuration and upgrade every 3-5 years. You will be better off: less devaluation, better value, more warranty and so on.

 

[KVB]

Not true. Year over year new MacOS releases (with new features) and new software increase demand on processor, memory, and storage.

By your thinking, Apple should still be producing 4GB base memory models.

I purchased 16GB back in 2013, today I'm going with 32GB as demands have increased significantly despite me not changing my workflows.

Tim

just bridge the L2 with graphite #2 pencil

 

[Beau10]

In the Apple world, I think the most important decision is RAM - because you can't change it, and under-sizing can have a big impact on OS and application performance. You can't really have too much, but you can over-spend

Agreed, that's why I got the base model - the most important option, the one that really gives it a bit of real longevity, is too pricey against the cost the machinery itself. In a couple more revs, either the default will be 32 gigs, or it will be a $200 upgrade instead - and on top of that, the machine will be significantly more performant in a variety of other ways.

Essentially as the original thesis of this thread implies, you'll be ahead... sell your computer for top dollar because it's Apple, esp. as you're not underwater is pricey options that have marginal returns, and be more current with the needs and whims of the current market AFA power.

 

[Sammy in SoCal]

How unwise would it be to just buy the M1 pro(November 2020 release) with Touch Bar right now starting at 1199$(education pricing) now. Is it’s resale and usefulness in the dump with these new releases? I’m not a video editor/developer/software engineer