Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

KDPlazed

macrumors newbie
Original poster
Mar 17, 2021
26
11
Recently, I was fortunate enough to save some money for a computer for college. However, I was wondering if it would be more ideal to purchase an Intel based Mac vs a arm based mac. I have heard a lot of news about the excessive wear on the SSD of Arm Based Mac's, and that's got me concerned as I want to atleast have my laptop last me for 3-4 years. Could anyone tell me what their ssd usage has been with few months of use to help me decide whether to purchase an Arm based mac or an Intel one?
 

MBAir2010

macrumors 604
May 30, 2018
6,975
6,354
there
ARM BASED
intel's chips are problematic on newer MacBook's and arm is apple's new architecture and future.
3-4 years?
i have a MacBook air that runs very good from late 2010, yours should last longer!
 
  • Like
Reactions: Quackers

Quackers

macrumors 68000
Sep 18, 2013
1,938
708
Manchester, UK
If mine (M1 MBA) doesn't last 3 or 4 years I'll be very miffed!
I expect it to easily manage 5 years and hopefully considerably more.
 

leman

macrumors Core
Oct 14, 2008
19,494
19,631
If you are looking at an entry-level Mac (basically, MacBook Air), there is no reason whatsoever, at all, to buy an Intel one. The next Intel Mac costs almost twice as much, is slower, has worse battery life and as @jasoncarle points out, has no future. And if you care about x86 compatibility, get a PC laptop. That ship has sailed.
 

dmccloud

macrumors 68040
Sep 7, 2009
3,122
1,883
Anchorage, AK
The SSD "issue" is nothing more than a handful of people being overly concerned and basing those concerns around SSD characteristics from 5+ years ago. Modern SSDs have insanely high lifespans, so those people complaining the loudest are complaining about 5 years' lifespan in the worst-case scenarios. The M-series Macs are far more power-efficient, outperform even an i9 MBP in some scenarios, and run cool and quiet while kicking the butt of most x86 machines (Mac and Windows) manufactured in the last 5 years.
 

KDPlazed

macrumors newbie
Original poster
Mar 17, 2021
26
11
Oh ok that definitely clears up some concern. I guess it would make sense as the M1 Macs probably have a higher read/write speeds compared to the Intel based mac's.

Thanks for the clearing my concerns!
 

padams35

macrumors 6502a
Nov 10, 2016
501
345
The biggest quirk is the M1 does not support booting into Linux or Windows. If you go M1 and have classes that require Windows software or Virtual Box images you may be forced to use the campus computer lab.
 
  • Like
Reactions: mi7chy

trevpimp

macrumors 6502a
Apr 16, 2009
697
301
Inside A Mac Box
I would be more worried how Apple would consider performance for past intel models. Apple will definitely make use of faster silicon chips and proceeding forward with more advance software performance an Intel CPU wouldn't be able to perform as fast with apple silicon improvements
 

satcomer

Suspended
Feb 19, 2008
9,115
1,977
The Finger Lakes Region
I keep thinking back in day with PPC Linux was running native on it! So just a recompile of the code can live again! So the future look good with Linux and some have proven they can run Microsoft Arm version! However no one shown one running some arm software on it!
 

hsiunghsiung

macrumors member
Dec 27, 2020
36
17
I would go M1 mac for price, performance, and future proofing.

I would only go intel if:
1) You want to bootcamp into Windows at some point for sure
2) Use an EGPU (if you already own one. Might not be worth getting one just for it cuz it may be bottlenecked either way)
3) Want to use 32 bit apps, or apps that require 32bit to authenticate like older Adobe software ( ex. lightroom 6.14 standalone requires 32 bit to authenticate. You can time capsule it to install in already authenticated form in operating systems from catalina and onwards though)

If you decide on intel because of any of those 3 points, then you'll need to do research to determine which intel mac to get. Some points to consider:
1) If you want 32 bit app handling, you'll need a mac that can run Mojave as it is the last one that allows it. This rules out T2 chip macs I think, and I think also rules out macbook pro 16 as it cannot downgrade to that? I could be wrong
2) Look up which macs have butterfly keyboard issues as it has made things really inconvenient for some owners. I think macbook pro 15 2015 was the last 15 inch with good keyboard. With macbook pro 16, I think 2018/2019 has the improved keyboard
 

KDPlazed

macrumors newbie
Original poster
Mar 17, 2021
26
11
I totally agree with everyone on the best value for a good macbook with unparellal speed compared to the Intel-based ones. However, I notice that many individuals who frequently use their macbook are experiencing huge read/write speeds, indicating a shorter life span of M1-based SSD's. Moreover, Apple has not addressed this problem makes it even more concerning to me. Especially with the intel mac's being discounted heavily and with continued support from Apple, would it not make sense to invest in an Intel Mac due to its guaranteed lasting of its ssd?
 

Homy

macrumors 68020
Jan 14, 2006
2,450
2,364
Sweden
Care to give a couple sentence summary we can read in a few seconds vs everyone spending a quarter hour watching videos?

You want the info served to you so you can watch cat videos? One of the videos is only under 7 min. You can always fast forward or jump to the end for a summary. There is a huge thread about this here https://forums.macrumors.com/threads/ssd-swap-high-usage-of-terabytes-written.2284893/

Short answer is no, it's not an issue. Third party software used to measure the written data can show false info and you should buy 16 GB RAM or bigger SSD if you can. If you want to learn more interesting stuff you should watch the whole videos.
 
Last edited:

jdb8167

macrumors 601
Nov 17, 2008
4,854
4,594
Short answer is no, it's not an issue. Third party software used to measure the written data can show false info and you should buy 16 GB RAM or bigger SSD if you can. If you want to learn more interesting stuff you should watch the whole videos.
The highlighted section is almost certainly wrong. I rewrote a portion of smartmontools for NVMe SSDs using only Apple approved APIs and I got identical results as smartmontools. If the data is false info it is Apple’s fault.

Details here: https://forums.macrumors.com/threads/ssd-swap-high-usage-of-terabytes-written.2284893/post-29749045
 
  • Like
Reactions: deaglecat

KDPlazed

macrumors newbie
Original poster
Mar 17, 2021
26
11
Yeah, I am looking at the entire thread about the high ssd swap usage, and people are saying that even with the software written with Apple's API indicate the high ssd usage. Which is very concerning to me, because I don't want a mac that becomes unusable after 1-2 years. In addition, there is no way to replace the SSD because of Apple's crappy idea of soldering the SSD and ram.
 

jdb8167

macrumors 601
Nov 17, 2008
4,854
4,594
Yeah, I am looking at the entire thread about the high ssd swap usage, and people are saying that even with the software written with Apple's API indicate the high ssd usage. Which is very concerning to me, because I don't want a mac that becomes unusable after 1-2 years. In addition, there is no way to replace the SSD because of Apple's crappy idea of soldering the SSD and ram.
Most people aren’t seeing a high number of writes. I’m at less than 9 TBW since I bought my M1 MBA at release in November. The tools still show 0% used. Once you get the M1 Mac, use it for a month normally and then check the TBW. A month is probably enough time to set your workload pattern. It is likely you won’t see any problem. If you do, then apply some of the mitigations found in these forums.

Edit: Apple solders the SSD and RAM for performance reasons. For the SSD for battery life and RAM for both battery and performance. An M.2 NVMe has an on-board SSD controller. Apple has their own on the M1 SoC that is 5nm with all the advantages that entails. The RAM is lpddr4x which is the highest performance for power but it isn’t available on a DIMM. It is also soldered on the same chip carrier as the M1 to obtain the lowest possible latency.
 
Last edited:

KDPlazed

macrumors newbie
Original poster
Mar 17, 2021
26
11
Most people aren’t seeing a high number of writes. I’m at less than 9 TBW since I bought my M1 MBA at release in November. The tools still show 0% used. Once you get the M1 Mac, use it for a month normally and then check the TBW. A month is probably enough time to set your workload pattern. It is likely you won’t see any problem. If you do, then apply some of the mitigation’s found in these forums.
Yeah, I think I might just do that and see how the ssd holds up. I have been looking through several threads and videos, and the causation for the excessive SSD usage is Rosetta 2. I think if I stick with just M1-optimized apps without using Rosetta 2, my ssd should just be fine.
 

Homy

macrumors 68020
Jan 14, 2006
2,450
2,364
Sweden
The highlighted section is almost certainly wrong. I rewrote a portion of smartmontools for NVMe SSDs using only Apple approved APIs and I got identical results as smartmontools. If the data is false info it is Apple’s fault.

Details here: https://forums.macrumors.com/threads/ssd-swap-high-usage-of-terabytes-written.2284893/post-29749045

Interesting! I said "can", meaning not all. Have you watched the videos? Do you think there is a problem? The TBW numbers given by manufacturers are estimations and in fact many SSDs last many times longer. There is a formula you can use to estimate the life span: https://www.compuram.de/blog/en/the...es-it-last-and-what-can-be-done-to-take-care/

"For example, the Samsung SSD 850 PRO SATA is stated to be “built to handle 150 terabytes written (TBW), which equates to a 40 GB daily read/write workload over a ten-year period.” Samsung promises that the product is “withstanding up to 600 terabytes written (TBW).”

 

deeddawg

macrumors G5
Jun 14, 2010
12,466
6,563
US
You want the info served to you so you can watch cat videos? one of the videos is only under 7 min. You can always fast forward or jump to the end for a summary. There is a huge thread about this here https://forums.macrumors.com/threads/ssd-swap-high-usage-of-terabytes-written.2284893/

Short answer is no, it's not an issue. Third party software used to measure the written data can show false info and you should buy 16 GB RAM or bigger SSD if you can. If you want to learn more interesting stuff you should watch the whole videos.

Excellent, thanks for the summary.

Now other forum participants can quickly know what you were trying to convey without wasting a bunch of time scrubbing through videos hoping to find a summary somewhere (if one even exists).
 
  • Like
Reactions: coffeeboat

Homy

macrumors 68020
Jan 14, 2006
2,450
2,364
Sweden
Excellent, thanks for the summary.

Now other forum participants can quickly know what you were trying to convey without wasting a bunch of time scrubbing through videos hoping to find a summary somewhere (if one even exists).

If one even exists? So you didn't even watch any of them and never cared about the problem yourself but are more concerned about not wasting other people's time? How do you know then that I'm not wrong? Am I not "wasting" my time to do your job by writing what the videos are about instead of you caring to watch them?

I guess it depends on your view on tech information. Do you consider reading newspapers, books or watching the news or movies as a waste of time? Do you ask others to tell you about the content of a book or a movie or do you want to read/watch for yourself and decide later?

Those are very good and educating videos and not "waste of time" to me and many others. That is how you learn more about your computer and can get a deeper understanding for the problem. I don't understand how someone can consider those short videos a waste of time and at the same time hang around tech forums like this one looking for answers. I can tell that you visited the other thread with over 80 pages since you liked some of the posts there when you instead could watch the 6 min long video above and get your answer. Isn't that a waste of time?

Another reason is if you watch them you can decide for yourself and use it as a valid source instead of mentioning me, a dude on internet, as the source. That is what I was trying to convey, watch and decide actively, because it depends on how you use your Mac and in some cases it can become an issue.

You can also find the videos much easier than you can find my post in the future if you forget something, but I guess not everyone is interested in details about their problem. They just want a quick answer. If people can't spend 5 min on understanding and solving such a problem then maybe they shouldn't worry about it to begin with?

Also worth mentioning that you often learn and remember better if you use different senses at the same time. Sound and picture are more powerful than only written words.
 
Last edited:
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.