Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Upgrade from M4Pro to M5Pro?
- Thread starter Superrenz
- Start date
- Sort by reaction score
AdmiralKirk
macrumors member
Only you know what your use case is. The reality is that the M5 Pro will shave 30 seconds to a minute off demanding tasks according to youtube tests. Are you using your M4 Pro for your business where 30 seconds to a minute off demanding tasks saves a lot of time every day, and is worth a few thousand $ to you?
Nope. Sure not… the only thing that might be worth considering is gaming. But the M4 Pro is fast enough - and I don’t care about the difference between 70 and 90 fps.
FreakinEurekan
macrumors G3
Only if you don’t have enough RAM or storage, and need a new machine anyway.
Basic75
macrumors 68030
The time to buy a new computer is when the old one stops doing what you need it to do, or doesn't do it well enough anymore. Only you can tell whether that's the case. Buying a new computer just because it exists is consumerism. If your hobby is buying new computers then go for it. Otherwise look at it as a tool and replace as needed.
Apple_Robert
Contributor
throAU
macrumors G4
Depends.
I have both M4 based and M5 based machines (M4 max macbook pro and M5 ipad). Also have M1 generation.
Is M5 generation faster at single thread? yes. Is the GPU better in terms of same tier vs. same tier? yes.
But 1 generation is a hard sell because of the massive hit you’ll take getting rid of the M4.
I’d say that unless your job relies on maximum performance and you will recoup the money via improved productivity - it’s a waste. And i very much doubt that’s the case.
Run the machine for 3 years IMHO. The M5 is great. M6 will be better and M7 will be even better.
Wait for M7. That’s what i’m doing with my own money.
Ditto for M1 vs. M2, M2 vs M3, etc. Single jumps aren’t worth it unless there’s some very specific niche use case that is massively accelerated (e.g., you heavily depend on say, AI inference in the case of M4 to M5… or say, ray tracing going from M3 to M4 or M5). 2 generations barely.
M5 is aimed at M1/M2 users really.
Last edited:
throAU
macrumors G4
Nah you can instantly tell the difference. Is it huge? Is it worth it? Probably not, but if you can’t tell the difference between different generations of processor within a couple of minutes of using the thing… not sure what to tell you…
10% is generally subtly discernible by most people and apple is typically hitting 10-15% improvement per generation.
Basic75
macrumors 68030
That doesn't sound right. I remember that to be at least 50% in general.
raythompsontn
macrumors 65816
Based on what time to accomplish the task? Does the task take an hour? Does the task take three hours, 45 seconds? Giving a time savings, without the actual time to process, is akin to manufactures stating a product is 30% better with no frame of reference. 30% better than what? 30 seconds faster than what?
Mostly 4Xstrategy or RPG like Baldur‘s Gate. Cyberpunk is way up on the to do list.
turbineseaplane
macrumors Penryn
Depending upon the cost to swap, I would say yes.
The SSD speed boosts are no joke, as are the GPU improvements (if you'll benefit from either).
The SSD speed boosts are no joke, as are the GPU improvements (if you'll benefit from either).
sw1tcher
macrumors 604
throAU
macrumors G4
50 percent is massive and way beyond “noticeable”.
Don’t confuse “noticeable” with “worth upgrading”.
enricoclaudio
macrumors 6502a
I did upgrade from M4 Pro to M5 Pro because I got a really good trade in value from Best Buy and also wanted a bigger screen. My M4 Pro was 14” and got the 16” M5 Pro. Ended up paying only $900 after trade in with 18 months financing with Best Buy.
1BadManVan
macrumors 68040
I agree with this, this gen isn't really meant for M4 users to make the jump, unless you under spec'd your current M4 for your use case. I went from an M2 to an M5, and really the bigger reason for the upgrade was because Apple held down prices this year, not sure they will do the same for next year. The big boost in performance was just a bonus as I had no complaints about my M2. Though it was more noticeable then I thought it would be even for just basic tasks, much snappier
Personally I'd hold off for a couple of reasons, neither of which might be applicable to you but here are the reasons I would skip the M5 series if I already had a decent M4 series MacBook...
1 - The M6 MacBook Pros are rumoured to be getting a chassis redesign that rumours say will make them thinner and lighter. If portability matters to you (it does to me) that might be of interest even if you don't care about other rumoured stuff like OLED touchscreen screen. Maybe you have enough spare cash to not really care about upgrading every year but if it was me then, rather than spend money on an upgrade this year, I would prefer to keep that cash in my bank account for a potential purchase next year once we know what the M6 generation will deliver (with even more performance than the M5 series being pretty much a given as well)>.
2 - I'm currently a bit conflicted about how much memory is now appropriate for a 2026 Mac purchase that one expects to keep for at least 2 or 3 years and I suspect that the answer for at least some people is that right now we just don't really know. Some people are completely disinterested in the rumoured upcoming AI features in which case I don't think that RAM size is such an unknown but if you are someone who thinks that AI features might become an important part of your computing experience in the future then this is a rapidly evolving field and I'm not sure we really know how that will affect RAM requirements. I think it's fairly safe to assume that Apple will make a fair amount of use of local LLMs but how will it size those? What will the minimum memory requirements be for whatever Apple deems a minimum acceptable performance? Will different models (parameters and/or quantization) be installed on a system depending on memory size and how much might that affect performance (quality and speed of results) depending on how memory constraints determine the model used? Will big strides be made within the next year in reducing model sizes such that it becomes a non-issue and say 24GB of RAM in no way constrains the Apple Intelligence experience? Because of that I would be much more confident choosing my longer-term RAM size requirements in 2027 than I am in 2026 which for me would be another reason to delay my next purchase until 2027.
Disclaimer: I am about to buy my first ever Mac after about 35 years of Windows use so I am in a totally different situation to you and despite my 2 points above I am 100% not patient enough o wait for the M6 series and so will be buying my first ever M5 Mac in a couple of weeks. I am however not going to go mad with the spec because I won't be at all surprised if I end up upgrading it probably not with the M6 series in 2027 but very possibly with the M7 series in a couple of years time both because I would value something as thin and light as I can get (and I don't want an Air) and if AI developments over the next 12 to 24 months do deliver functionality that is of interest to me and it does end up making 24GB a bit on the tight side in terms of performance then I can step up the RAM size at that point.
1 - The M6 MacBook Pros are rumoured to be getting a chassis redesign that rumours say will make them thinner and lighter. If portability matters to you (it does to me) that might be of interest even if you don't care about other rumoured stuff like OLED touchscreen screen. Maybe you have enough spare cash to not really care about upgrading every year but if it was me then, rather than spend money on an upgrade this year, I would prefer to keep that cash in my bank account for a potential purchase next year once we know what the M6 generation will deliver (with even more performance than the M5 series being pretty much a given as well)>.
2 - I'm currently a bit conflicted about how much memory is now appropriate for a 2026 Mac purchase that one expects to keep for at least 2 or 3 years and I suspect that the answer for at least some people is that right now we just don't really know. Some people are completely disinterested in the rumoured upcoming AI features in which case I don't think that RAM size is such an unknown but if you are someone who thinks that AI features might become an important part of your computing experience in the future then this is a rapidly evolving field and I'm not sure we really know how that will affect RAM requirements. I think it's fairly safe to assume that Apple will make a fair amount of use of local LLMs but how will it size those? What will the minimum memory requirements be for whatever Apple deems a minimum acceptable performance? Will different models (parameters and/or quantization) be installed on a system depending on memory size and how much might that affect performance (quality and speed of results) depending on how memory constraints determine the model used? Will big strides be made within the next year in reducing model sizes such that it becomes a non-issue and say 24GB of RAM in no way constrains the Apple Intelligence experience? Because of that I would be much more confident choosing my longer-term RAM size requirements in 2027 than I am in 2026 which for me would be another reason to delay my next purchase until 2027.
Disclaimer: I am about to buy my first ever Mac after about 35 years of Windows use so I am in a totally different situation to you and despite my 2 points above I am 100% not patient enough o wait for the M6 series and so will be buying my first ever M5 Mac in a couple of weeks. I am however not going to go mad with the spec because I won't be at all surprised if I end up upgrading it probably not with the M6 series in 2027 but very possibly with the M7 series in a couple of years time both because I would value something as thin and light as I can get (and I don't want an Air) and if AI developments over the next 12 to 24 months do deliver functionality that is of interest to me and it does end up making 24GB a bit on the tight side in terms of performance then I can step up the RAM size at that point.
throAU
macrumors G4
Worth noting: if you HAVE under specced your use case, that's something else entirely. Screen upgrades/storage upgrades/etc. are all valid reasons to jump, but if its just "but new shiny thing!" then there's always new shiny thing.
The fact that we're talking Pro machines here and not max leads me to believe that no the difference won't make you any more money so its just a want really. At which point its like.... do what you want.
But as above I do think doing single step upgrades is excessive, unless you're literally being held back generating real income with the thing entirely due to needing maximum performance, in which case you should be buying 128GB RAM Max and upgrading to the new version of that as soon as practicable.
Last edited:
New shiny things are always tempting… but you’re totally right: It won’t make me more money.. therefore it stays in the store.
Alameda
macrumors 68000
I just upgraded from an M1 Pro to an M5 Pro. It took that long for the upgrade to be justifiable for me. And even then, the M1 Pro did everything I wanted it to do; I could have kept using it, but I was getting some slowdowns and glitches while doing intensive work, so I got a computer with 48GB of RAM and lots more CPU cores so I won't have to worry about it.
My college-age kids are still using 8GB MacBook Air M1 laptops. They are perfectly happy with them.
My point is that I don't think a one year improvement is worth upgrading for. It's only when you add the increments together over the course of several years that it is noticeable.