When I had originally asked this question, I assumed that this information may have been provided via a leak but it seems not. I had not watched the Apple event so I assumed there may have been some hints given but no. What you've witnessed above is people arguing over things that have not been confirmed yet.
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.
M2 Mac Mini Storage: 256GB vs 512GB single NAND?
- Thread starter Mity
- Start date
- Sort by reaction score
Gotcha - yeah, I was looking for the same info via a leak and was just checking again and came across your post. I did a lot of research before buying my M2 Air with 512GB and am happy with the performance. I watch the memory pressure on my laptop and it does go yellow occasionally, so it’s nice to swap to faster SSD. I also didn’t want to always have a nearly full SSD with only 256GB, which also hurts performance with the single chip.
Did you listen to what you quoted? It says the they didn't see any noticeable difference in real world behavior, that most people are using this for web surfing and zoom calls, and the M2 outperforms the M1 and and then basically says maybe you want a bigger SSD so you can hold more files-- not because of speed.
I've watched the entire video. They did a web surfing test and suggested an increase to 512GB to hold files.
"I think it's a very capable machine for what it is though I would recommend that if this is going to be your laptop of choice for a long time and you're worried about the SSD and the speeds being a problem, I just think that 256 is not enough if you're going to be keeping files on this for a long time and if you don't want to get rid of a lot of things, please bump up to 512..."
I said exactly why it's not geared towards professional use. The HIGHER SPEC ones are. The base spec one is NOT. No professional can get by with 256GB of storage. No matter if its 100 GB/s speeds.
A 2010 Mac Pro is still usable by professionals. And the 256GB MacBook Air has over 3 times....yes OVER 3 TIMES the SSD performance as the Mac Pro.
It's marketing talk. My Mac Studio CANNOT achieve 7GB/s SSD performance. But Apple said so in their marketing that it would. A maxed out MacBook Air would be a decent professional system. But the absolute base spec? No.
Why did Apple use the base M1 and M2 to compare battery life? Why not compare the exact same model with more RAM and larger SSD?
.... I thought we were talking about SSD performance and professional performance of a system, not battery life? Higher spec SoC use more battery.
The point is that performance numbers were cherrypicked to make the machine seem more competent in every way. Apple won't even compare the base M1 and base M2 in simple photo/video editing tests.
Professional reviewers probably already have these machines. The reviews will come out on the release day or shortly thereafter so, most likely before Feb 1.
Even the base spec one can do "simple photo/video editing". As I said, I am able to achieve these tasks on my 2010 Mac Pro that is limited by SATA 2 speeds, which is 250MB/s Max by the way if you did not know. Perfectly capable of handling my video editing and photoshop work.
Good for you - if it works for you, keep using it. But how does that prove that the base M2 Air is a better machine than the base M1 Air? It doesn't. No one thinks so.
Here is The Verge's take on it:
All of those configuration options add up to a noticeably more expensive computer than before, and there’s very good reason to avoid the base model entirely. Apple confirmed to me that just like the base model of the 13-inch MacBook Pro M2, the base Air’s 256GB of storage is stored on a single NAND chip instead of two like on the M1 models or new M2 models with 512GB or more storage. That can make the storage perform half as fast as even the older base M1 Air’s and will slow things down whenever you try to copy large files around or multitask enough to max out the 8GB of RAM and force it to use swap memory. It’s a disappointing regression and really means the only models I feel comfortable recommending start at $1,500.
We were able to test a base model Air with 8GB of RAM and 256GB of storage and sure enough, the storage is a lot slower than the prior M1 Air or M2 models with 512GB. Unless you’re able to get the base model M2 Air on sale for a sizable discount, I’d stay away from it.
Yes because the SSD is the only thing important. Let's ignore all the other differences between M1 and M2 like media encoders, higher clocks, different CPU/GPU configuration etc. Nah, its worse ONLY due to ONE component.
Let me ask you something. Do you actually do any video or photo work? 256GB is NOT enough in any way. Not even if it offered 100 GB/s performance, 256 is just NOT ENOUGH even for "simple video/photo editing". The base spec is geared towards very very light users. Just hopping on Facebook, or typing up their college papers.
Last edited:
When you have the base model with 8GB RAM and swap starts taxing the SSD, it matters!
Why no talk of encoders, etc.? In your own words, in post 36: "I said exactly why it's not geared towards professional use." Why do those things now matter if the target audience aren't professionals? I never made any claims about the M2 chip. Go back to my first post, where did I say the M1 chip was better than the M2 chip?
Let me be very clear: I think the base M1 Air is a better machine overall than the base M2 Air. There is virtually no difference in performance unless you use the SSD like a normal human being. A person has to decide if paying $300 more for the webcam, and slightly larger screen is worth it to them. I don't. The base M1 Air has a better SSD than the base M2 Air and since there is little performance gain from the M2, then it just makes sense to buy the M1.
Edit: Regarding your comment about video editing, yes people do use the base M1 for video editing.
Last edited:
The MacRumors video you posted thought so.
And they gave benchmarks showing the M2 faster than the M1 when doing real world file transfers here:
And you're "witnessing" people arguing with you over details that haven't been released yet because you're insisting on them doing so.
Is websurfing a legitimate use case or not, because you seem to keep saying it isn't but then keep talking about chrome tabs. When people point out to you that 256GB of storage might be a bit light for video editing anyway, you argued against it by highlight the 2TB model Apple benchmarked.
This is the problem with this whole thread. You want to hear a certain response back, and when you don't hear that response you're not sure how to handle it and give contradictory replies.
And is it the speed of the SSD you're concerned about or the space because if your question is whether the 256GB Mini will store more than 256GB, then no it won't. If your question is whether the 256GB M2 Mini will be noticeably slower in typical use than a 256GB M1 Mini, the answer is only in the most narrow of circumstances. Likewise, versus the 512GB M2 Mini, only if you have a way of doing sustained access to the internal drive with no other bottleneck which is just not very common.
For example the Blackmagic tests that everyone is pointing to says that you can write to the M2 drive at 2305MB/s on the 512GB drive and 1362MB/s on the 256GB drive-- that's abstracting away the CPU and everything else and only pushing random data to the drive without doing any work on that data at all. So let's say you have a near empty drive with something like 200GB free and then want to fill it with random numbers. Ballparking at 1000MB per GB, because I don't know what's using what standard, it takes a minute more to write the file to the 256GB drive-- and then your drive is full. When you ask yourself how many times you're going to write a 200GB file on a 256GB drive but do no other work on the data, you start to realize how little time in your day you're losing to this.
You're very unlikely to see a difference with transferring files to external drives because only the fastest TB3/4 drives will exceed the 1400MB/sec mark anyway. And how much time does someone spend pushing files onto and off of an internal drive? If you're doing it more than once a day, and if your external drive is faster than the internal one, then just work off the TB drive.
I have yet to see a legitimate benchmark on swap, but I've seen plenty of people assume swap is affected because of the Blackmagic benchmark. That's a nonsense comparison because the Blackmagic benchmark tests writing files 1-5GB in size and a swap memory page starts as 16 kilobytes and then is compressed before paging out. I don't know one way or the other but we could just as easily assume that the RAID controller for the dual flash system slows down the small transactions relative to the single flash system.
Chrome is bit of am aberrant use case when it comes to swap-- I'd consider it more like caching than swap-- but what can we expect? I use Safari, but when I open a MacRumors tab it seems to be about 500MB. When I switch away from that tab the memory usage gets reduced to something less than 200MB. That probably gets compressed before paging out but let's just assume it doesn't. It'll take a 512GB M2 0.073sec to read that 200MB back from disk and a 256GB M2 0.135sec for a difference of 62ms. It takes you about 100ms to blink.
All of this has been hashed over in the past... That MaxTech nonsense worked everyone into a lather but honestly unless you're trying to create a benchmark to emphasize the difference, it's not life changing.
Last edited:
I might be giving Apple too much credit, but I think they have a better reason (than encouraging us consumers to upgrade) for going with a single 256gb NAND flash SSD chip rather than 128gb+128gb.
Maybe the M2 chip simply doesn't support chips smaller than 256gb. Maybe they got a better deal with a contract that allows their supplier to close down its 128gb fab. It could be a lot of things.
It will be interesting to confirm (or not) that the M2 Mini does match the M2 Air in this regard. We all seem to be presuming it.
With consumer electronics that sells in large quantities saving a buck or two equates to saving millions of dollars. In my experience as an engineer who's worked with design teams for mass market consumer items, saving even a penny or two is a BIG deal because it adds up fast.
Over on Reddit in this linked post above, two individuals contacted Apple Support asking about the M2 *Pro* mini SSD being single or double NAND:
512GB base: 1 NAND
1TB BTO: 1 NAND
(lol…)
This is the M2 Pro, we can “almost certainly” assume the regular M2 mini getting the same treatment if not worse.
512GB base: 1 NAND
1TB BTO: 1 NAND
(lol…)
This is the M2 Pro, we can “almost certainly” assume the regular M2 mini getting the same treatment if not worse.
Oh man, if true the global meltdown over this is going to be epic. Buy Alphabet stock because the YouTubes are going to be turbo charged for a few months. 🍿🤭
You linked a video for the M2 Macbook Pro, not the M2 Macbook Air. You seem to be reaching for data to match your opinion rather than accepting that the base M1 Air has a better SSD than the base M2 Air.
But I'll humor you. Let's suppose that benchmarks don't matter because the truth is that not all benchmarks do matter. If the M2 processor is x% better but ALL reviewers have stated that there is negligible difference between the base M2 Air and the base M1 Air, then what is the limiting factor that is preventing the overall package not being x% better? It's the SSD. There is no other major component that would produce this result. The M2 chip is an improvement and the base M2 Air SSD is a regression - exactly as The Verge puts it. The net result being a minor improvement (until you start using large files and then the SSD of the base M2 Air struggles).
This is why the following all suggest to skip the base M2 Air:
MKBHD
Washington Post
PC World
Macrumors
The Verge
Last edited:
I agree 100%. I worked on Wall Street in research for about 3 years. People don't understand how much pressure Wall Street puts on companies to achieve QoQ and YoY growth. Wall Street also seems to be hell bent on Apple maintaining a 40% gross margin. I think all of this is highly unreasonable and unfair. Short-term goal seeking is counterproductive for long-term sustainability, especially in tech and science in general. You don't walk into a science job and produce some profound result at the end of a particular week. Research is not the same as building a financial model.
Tech companies are currently laying off thousands. As an example, Amazon is cutting jobs at Alexa but at the same time saying that they're committed to it in the long run? It doesn't make any sense for long-term growth. The only explanation is to satisfy the investment community.
I think the same thing is happening at Apple.
Is the M2 MBP SSD configured differently than the M2 MBA or are you looking for distinctions without a difference?
If that's what passes for logic in this discussion, then there's really no point in pursuing this further. You have an answer you're seeking and keep trying to create garden paths to that answer. That's fine. The last thing I want to do is make you uncomfortable with your choices.
I won't watch a bunch of YouTube videos to prove this one way or the other, but the PCWorld review quotes the bogus MaxTech results so their recommendation isn't worth anything, and the MacRumors recommendation doesn't say skip the base M2 Air, it says that their results show the base M2 configs are faster but if you're worried about it get the larger storage.
The Verge didn't do anything more than add their echo to the chamber of everyone else parroting those garbage MaxTech results and then repeat the Blackmagic test for 1 and 5GB file copies and somehow extrapolate that result to swapping compressed 16kB pages and copying files (APFS doesn't write any data when you copy on a volume, and as I said before you'd only see it with external copies if you have a very expensive TB3 drive). After all that they still need to concede:
The Verge: "That said, will these particular differences impact you? People the Air is marketed to will likely not see a life-changing contrast between the 256GB and 512GB models when it comes to everyday performance. I ran two 4K YouTube videos over 25 open Chrome tabs for 30 minutes on both machines without either needing to dip into swap memory. Boot time was also pretty identical — I turned the two devices on side by side a number of times. And I didn’t see much of a difference when it came to opening any of the apps I normally use, including Chrome, Safari, Messages, Photos, Activity Monitor, Slack, Music, etc."
Absolutely this. So many people think geekomania forums like this are real life. Besides that the real world performance impact is minimal, it's not a forced upgrade pricing strategy if hardly anyone knows that you have to buy up to the next level to get more bandwidth.
Also I thought it was believed that there may have been some supply chain issues that led to this as this wasn't an issue in previous laptops.
I would think the Air and Pro have different SSDs, even if the configuration is the same but I could be wrong. But I would would not overlay a Pro's results to the Air's.
No one doubted that web performance was an issue until swap was used, even me.
And you linked a different Verge article by Monica Chin written in July 15 2022. I've linked the updated review by Dan Seifert written in Oct 22 2022. That article does not arrive at the recommendation in your article at all. However, your article does provide an important data point: 25 Chrome tabs + 2 4K video streams and no swap usage is impressive, I must admit. Unfortunately, it's the only useful data point Monica does provide besides the benchmarks.
Consider this line:
"And I didn’t see much of a difference when it came to opening any of the apps I normally use, including Chrome, Safari, Messages, Photos, Activity Monitor, Slack, Music, etc."
Is she just opening these apps and not using them? What use is that? She basically listed a bunch of apps and did absolutely nothing to test their functionality. Useless.
And consider her final words:
"My real-world comparisons have found that M2 machines are visibly better for graphics-heavy use cases (such as running games) but that their performance differences are not hugely impactful in other tasks (photo and audio editing, internet work, etc.) that a casual user might do."
How is it that we don't get an improvement in photo and audio editing, internet browsing and other "casual" tasks? The only explanation is that even though the processor is improved the SSD holds it back.
And you are wrong - in the Verge article I linked, Dan is not referencing MaxTech's results:
"In my benchmark testing..."
Last edited: