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
I checked the Apple refurb site last week and used base Mac Studio was $1799...same price as a new one from the Apple Education Store.
Last edited:
That's what I went for too -- M2 Pro 16/1TB. I didn't get the 10GbE even though it's a pretty good price for that option. I read 10GbE gets pretty hot and not sure if I would upgrade entire home network to 10GbE before it's time to replace the Mini in the future. If I do go 10GbE in the next 2-3 years I'd probably get an external unit that will hopefully be cheaper than current prices, plus it keeps the heat outside the Mini.
Last edited:
I'm not in the market for a Mini here, at least not yet, so I can watch this with a certain level of detachment, but I do find it entertaining to watch people keep upselling themselves. The original question was "one NAND or two", now people are freaking out that some of them have 2 instead of 4 and are convincing themselves that they can't buy a 256GB unit because it has one chip they need 1TB so they get 4.
Anyway, it seems to me that the bandwidth to capacity ratios are about right. If you actually need more bandwidth, you likely also need more storage. If you need more storage, you can benefit from more bandwidth. If you're worried about swap then you're in the wrong conversation-- buy more RAM.
Anyway, it seems to me that the bandwidth to capacity ratios are about right. If you actually need more bandwidth, you likely also need more storage. If you need more storage, you can benefit from more bandwidth. If you're worried about swap then you're in the wrong conversation-- buy more RAM.
The tiered pricing with varying bandwidth/capacity ratios don't bother me — but the lack of transparency does. If Apple would have published the 1500MB/s vs 6000MB/s SSD numbers before we pre-ordered, the choice would have been easy. Instead we have have to learn about it from independent reviewers.
I dunno. There will always be some piece of information people will have wanted... The information was available on the day of release. I understand people are excited and wanting to get their new shiny as soon as possible, but this is a risk of pre-ordering.
For a lot of tasks the new Minis outperform last years Studios, regardless of whether the SSD is 3GB/sec or 5GB/sec.
I just saw a new video comparing the M2 Pro in various computers -- all have the 512gb SSD. The M2 MBA in the chart has the slow 256gb SSD. The 512gb SSD in the M2 Pro MBPs is a lot faster than the one in the M2 Pro Mini. The M1 Mini 512gb SSD was probably about the same as this M2 Pro Mini, but I do not recall. I assume the M2 Mini and M2 Pro Mini have the same 512gb SSD.
Here is the video with some other benchmarks:
Here is the video with some other benchmarks:
In the real world benchmarks, did we see differences in performance that could be attributed to disk i/o?
I believe the answer to be no. It looks to me like M2 Pro (mini) and M2 Pro (14” and 16”) differences were almost nil, once he controlled for CPU/GPU setup. The big variable was disk, BlackMagic showed disk i/o wasn’t as good in the mini, and that led to benchmarks that were….exactly the same on all three.
This reiterates to me that the concern around disk i/o that I see endlessly repeated in the Macosphere is a bit overdone.
Just how did anyone edit video with an SSD that was ‘only’ 500MB/s, anyway?
I think the labels got switched. Other sources have reported the 10c M2 Pro getting 12K in Geekbench.
A few months ago this guy was telling people the slow M2 MBA 256gb SSD was much ado about nothing. Now he has reversed his opinion after using the M2 MBA 256gb SSD for awhile.In the real world benchmarks, did we see differences in performance that could be attributed to disk i/o?
I believe the answer to be no. It looks to me like M2 Pro (mini) and M2 Pro (14” and 16”) differences were almost nil, once he controlled for CPU/GPU setup. The big variable was disk, BlackMagic showed disk i/o wasn’t as good in the mini, and that led to benchmarks that were….exactly the same on all three.
This reiterates to me that the concern around disk i/o that I see endlessly repeated in the Macosphere is a bit overdone.
Just how did anyone edit video with an SSD that was ‘only’ 500MB/s, anyway?
The M2 MacBook Air: I was WRONG!
Remember when I vigorously defended the base model M2 MacBook Air earlier this year? Well, it turns out I didn’t dig deep enough.
Almost everything you just said is so wrong I wouldn't even know where to start.
My recollection of what he said:
I had two different OSs on there; I really need to re-test everything at the same time.
Until that happens, it’s interesting guesswork, but that’s all it is. The point of testing is to eliminate variables until you can point to what happened or what is wrong.
I’d also like to see his results duplicated in other’s results, as right now that’s not what’s happening. So far we have vague “It was slower”, but (at least I didn’t see) no benchmarks or further details.
I’m still waiting. But I’m sure you’ve seen the numerous results where all the M2 Pro machines (of same core/GPU amounts) performed essentially exactly the same - with numbers and benchmarks. That’s in post #134.
While we’re asking for the sky: I’d also really like to see an Ars-Technica or Tom’s Hardware review of this phenomenon. The YouTube guys do this for clicks; while we can argue AT/TH does the same, at least they put some serious reasoning and analysis behind it, using a methodology that stands up to casual (++) analysis.
I have not seen a single post for the M2 non-Pro with 512GB speed nor the 1TB speed. Your statement above I think is just a guess what the speed is. It could be 1500MB/s for all we know right now until someone tests and posts it.
Last edited:
Can't speak for the non-pro non-base... but for the Mac Mini M2 Pro...
The Mac Mini M2 Pro 512 Configuration has two NAND chips, rather than one NAND chip
MacWorld Benchmarked the M2 Pro w/ 1TB @ ~5000 MB/s read / ~6000 MB/s write
https://www.macworld.com/article/1480091/mac-mini-m2-pro-review.html
MacRumors Benchmarked the M2 Pro w/ 512GB @ ~2700 MB/s read/write
Last edited:
The 14” he got is not the standard base 14” SKU, by BTO upgrading the SoC to the full M2 Pro, he may have “avoided” being supplied with just 2NANDs and got 4NAND config instead. (<- this is all speculation on my part).
Something that is not clear to me is. For example:
The M2 Pro 512 has 2 Nands and gives 3Gb/s.
The M2 Pro 1T has 4 Nands and gives 6 Gb/s
Apart from speed, is there any other improvement, do they write and read on each chip in parallel in both cases?
thanks
The M2 Pro 512 has 2 Nands and gives 3Gb/s.
The M2 Pro 1T has 4 Nands and gives 6 Gb/s
Apart from speed, is there any other improvement, do they write and read on each chip in parallel in both cases?
thanks
Writing and reading in parallel is exactly how the double "rate" is achieved. You also get more evenly distributed data blocks and perhaps heat. The drawback is it consumers more power per NAND, so using less NANDs ironically improves battery life, though by what margin is hard to gauge.
