Disk speed testing on Apple Silicon: AmorphousDiskMark, Blackmagic, etc. [MERGED]

[mecloud]

Test results for a second sample of the M3 Max 16” MBP with the 16/40 core CPU/GPU, and the base 48gb RAM/1tb SSD.

 

[ori69]

Test results for a second sample of the M3 Max 16” MBP with the 16/40 core CPU/GPU, and the base 48gb RAM/1tb SSD.

Logic benchmark please

Logic Pro M1 (Pro & Max) Benchmarks - Music-Prod.com

Here is the Logic Pro X Benchmark project file to use. Just download it, run it and see how many tracks you can get to play.

music-prod.com

 

[unpretentious]

M3 Max 16" 16/40 Core CPU/GPU 128GB RAM 2TB SSD

 

[jdb8167]

Everyone loves a one number benchmark result but Blackmagic Disk Speed Test is meant to measure sequential speeds needed for editing high end video not for general performance. AmorphousDiskMark is meant to measure both sequential and random access speed. It is more comprehensive and will give a better picture of what different SSD configurations affect performance.

AmorphousDiskMark

Edit: My 1 TB M2 MacBook Air results:

View attachment 2148111

I just purchased a 2 TB OWC Envoy Pro SX Thunderbolt SSD. I figured that I would compare to my internal 1 TB M2 MacBook Air SSD.

1 TB M2 MacBook Air. 40% used, 17 months old.

2 TB OWC Envoy Pro SX Thunderbolt, 1% used, new.

Overall the random access results are superior which is good though I expect that as the drive is filled up, those results will be reduced. Overall, the OWC SSD seems like a pretty decent TB SSD for $250 with an absolutely bullet-proof enclosure.

 

[splifingate]

Mac Studio M2 Max 2TB:

SkHynix P31 Gold 2TB in an Acasis TBU405:

2TB Inland Premium in a Mac Pro 5,1 (what I've primarily been using for the past few years):

I foresee Good Times ahead

 

[3Rock]

16" MacBook Pro Max 48 Gb / 1 Tb

 

[jdb8167]

From the new 1 TB M3 MacBook Air:

 

[spnc]

From the new 1 TB M3 MacBook Air:

View attachment 2358396

Thank you.

 

[spnc]

16Gb 13-inch, MBP M1, late-2020

 

[macsoundsolutions]

Testing out a M2 Ultra 1T base model and my external user account drive with my Beelink Thunderbolt 5 enclosure with WDSN850X is wining in all RND4k tests. It also wins in the top two when connected to the M4 Pro Mac mini cause the Studio is only Thunderbolt 4 hence the slower speeds on the top two tests compared with the Mac Studio.
But I was expecting the Studio to win on the random 4k tests but it did not.

 

[johnprogrammer]

My Mac mini M4, 32GB RAM, 1 TB internal SSD.

 

[maflynn]

Here's my M4 Max Studio, both my internal and external, surprisingly my external drive is faster then my internal in some of the tests.

 

[theorist9]

Here's my M4 Max Studio, both my internal and external, surprisingly my external drive is faster then my internal in some of the tests.
View attachment 2537041

Random R/W's, which are supposed to be more important than sequential R/W's for real-world responsiveness, seem disappointingly low on AS vs. the values for my 2019 Intel iMac with a 2 TB WD SN850 (see screenshot). I wonder if Amorphous isn't properly assessing AS.

Amorphous uses 4kb transfer sizes for its random measurements. IIRC, x86 has a 4kb page size while AS has 16kb, so perhaps there's a disadvantage when assessing 4kb transfer sizes for AS. I investigated this a while ago, and wasn't able to get a clear answer.