I think it's pretty good!! I wouldn't complain getting the 512gb or 1tb in terms of speeds
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.
M1 Pro SSD speeds
- Thread starter BarryBigtime
- Start date
- Sort by reaction score
but setting the test to 2gb gets me 4-5k rsults.and using 5gb test file,i get 3-4k lolI think it's pretty good!! I wouldn't complain getting the 512gb or 1tb in terms of speeds
did you notice how "worse" reading speeds are compared to write ?I think it's pretty good!! I wouldn't complain getting the 512gb or 1tb in terms of speeds
Yeah sometimes it's like that. For example on 4k tests usually writes are faster than reads. Depends i guess on the SSDs too.
j
and somehow whilst blackmagic speeds are similar,i do get muuuch lower resultss using amourphous disk test...remember on my 14 i had 7000 read and 5200 write ? as seen in previous answer
well on the 16 i get around 4800 using amorphous disk test...meaningful or meaningless?
just got a 16inch (although i might just get back to a mba m1 tbf with you haha)
and somehow whilst blackmagic speeds are similar,i do get muuuch lower resultss using amourphous disk test...remember on my 14 i had 7000 read and 5200 write ? as seen in previous answer
well on the 16 i get around 4800 using amorphous disk test...meaningful or meaningless?
yep,14 and 16 base (512)
both score the same in black magic (6000-5000) however the 16 is MUCH lower in amorphous
not even ,I have even less storage used in new 16 than in previous 14
also I tried two 14 both scoring the same so that's weird
actually from max tech teardown the 16 has even more SSD chips ...and also scores higher /faster so idk
@white7561 sorry to tag you if youre not interested but ig you ought to know ....
somehow the default settings were set at 1 (i assume it represents the amount of times it writes right?
like at the left of the amount of gb to write there is a number. last time with the 14 i had set it to 5 whilst this time around it was at 1 for some reason.after setting it to 5 i also get about the same results (within margin of error )
somehow the default settings were set at 1 (i assume it represents the amount of times it writes right?
like at the left of the amount of gb to write there is a number. last time with the 14 i had set it to 5 whilst this time around it was at 1 for some reason.after setting it to 5 i also get about the same results (within margin of error )
Yep. That's the thing. Afaik it's 5 by default. But yeah. The test file was too little probably that's why the performance is lower than expected. Usually increasing the amount of times testing and increasing the test size increases the accuracy
I just got the base 14" and my results are in line with what others are reporting.
What I find odd is that the Random 4K QD1 writes are so slow. Like 30-35 MB/s slow. That seems to be the range most people are getting. And since it's unlikely we all got defective drives that must be the actual performance range of the drives Apple put in.
For reference Lacie's mid range SSD from 2 years ago gets 40-50 MB/s at Random 4K QD1 writes... see https://www.legitreviews.com/lacie-rugged-ssd-review-1tb-portable-drive-tested_222505/3
(Btw for nearly all consumer and prosumer workloads the performance at high QD numbers do not matter since only highly loaded server workloads could saturates 10+ QD. QD1 performance is what you'll experience 99% of the time.)
What I find odd is that the Random 4K QD1 writes are so slow. Like 30-35 MB/s slow. That seems to be the range most people are getting. And since it's unlikely we all got defective drives that must be the actual performance range of the drives Apple put in.
For reference Lacie's mid range SSD from 2 years ago gets 40-50 MB/s at Random 4K QD1 writes... see https://www.legitreviews.com/lacie-rugged-ssd-review-1tb-portable-drive-tested_222505/3
(Btw for nearly all consumer and prosumer workloads the performance at high QD numbers do not matter since only highly loaded server workloads could saturates 10+ QD. QD1 performance is what you'll experience 99% of the time.)
Here's something for thought, it's a whole thread on very technical stuff
For an unsupported OS, which all Linux variants are on M1 chips, I can see loads of hidden gotchas slowing down real world speeds.
But for the latest update of macOS? The only likely reasons why it would presist would be due to a fundamental design choice they made with the M1 chip or with the SSD, or some other control chip (T2?).
It may be due to some obscure gotcha in the M1 chip, but that doesn't seem to make much sense either since writing a lot of small files is not exactly some niche or obscure edge case, so the engineers designing them would have realized.
Upon further reflection I'm thinking it's more likely that Apple chose some novel implementation in the SSD setup that saved on costs and/or power usage, that had the side effect of halving random 4K writes performance.
I think I figured it out. SSDs nowadays all have some intermediate level of storage, usually a much faster and much smaller amount of buffer.
The Lacie drive in question has a sizeable amount of this, so for transfers less than 100 GB it simply is faster than Apple's SSDs that write directly without a buffer. But for larger sizes, pretty much all external SSDs, except for the really high end ones, hit a wall and slow down to like 5 MB/s because then the buffer is full and writes have to go directly to main storage, which is way slower.
So Apple is likely using much faster chips but without an intermediate buffer on their SSDs. Which makes it both costlier and more energy efficient, along with being space saving.
For example, if your transferring 500 GB of small files the Lacie drive would be 5x slower, but for only 5 GB it would be 2x faster.
Last edited:
Larger SSDs, I believe, have greater longevity.
Given that you can never swap out the SSD, it makes sense to go large…
serverfault.com
Given that you can never swap out the SSD, it makes sense to go large…
Does bigger capacity SSD have longer life due to wear leveling?
I have been told that you can get a longer lifespan of an SSD if you buy a bigger capacity SSD. The reasoning goes that newer SSDs have wear leveling and thus should sustain the same amount of writ...
