1TB M1 Max.
Seems like it gets almost half the speed than the 4TB one.
Just so many people gonna think their 512GB M1’s going to have 7.3 gb/s lol
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
Seems like it gets almost half the speed than the 4TB one.
Just so many people gonna think their 512GB M1’s going to have 7.3 gb/s lol
Oh I didn’t know that.
So as bigger the SSD gets, the slower the speed will be.
Hmm, interesting.
It got almost 2GB/s more speed now, 5Gb(before) vs 1Gb (now)
Attachments
Here's the little text on Apple's website :
Testing conducted by Apple in September 2021 using preproduction 16-inch and 14-inch MacBook Pro systems, all configured with Apple M1 Max, 10-core CPU, 64GB of RAM, and 8TB SSD. Tested with FIO 3.27, 1024KB request size, 150GB test file, and IO depth=8. Performance tests are conducted using specific computer systems and reflect the approximate performance of MacBook Pro.
Anyways, here's my test result for 16" 1TB =
Alternatively, everyone needs to remember that SSD’s need 25-30% of their space free to keep the speed up, due to how they store data. Filling it to 90% will make it slower than a platter driven HDD. So a 1tb drive is really only a 700 GB drive, and a 512GB drive is really only a 358GB drive.
Um, no. Black Magic Disk Speed is good for film editors - it's great to know which film formats will work in editing - but it's absolutely terrible for every other aspect of disk speed testing. That read & write speed that it shows, how was it obtained? How many threads? What file size? What queue depth? It doesn't say.
The 4 main corners of SSD speed testing are sequential read/write, and 4K random read / write. Black Magic doesn't test the 2 corners of random r/w at all, which arguably is far more important for the day to day feel of 'speediness' of the computer for most normal users.
There are many other aspects of SSD testing: speed while almost full (important if trying to work out how big a SSD you need); cache size (important for film editors who need to move huge amounts of film footage); IOPS at low & high queue depth (mainly useful for server SSDs); etc that Black Magic doesn't even seek to approach.
I'm considering whether I need to upgrade to the 1tb version with 512GB.
In many articles it is stated that the difference in performance between m1 pro 512gbssd and 1tb ssd is about 30%
Currently my ssd usage is 60%, which doesn't seem like a big difference in performance.
apple made a pro monster and max monster XD
Attachments
I think the performance difference won't be an issue for many tbh. I think it's better to buy the capacity you need and not for speed unless you want to
I also wanted to exchange my macbook pro 16 512gb to a 1to model but in the end I have 390Gb of free space on the 512gb and I have a sandisk extreme V2 1050mo / s 1tb ssd, I think we must choose according to these needs, if I would have gone to the 1tb model the faster writing speed does nothing more for the video / photo export nor for the speed of navigation and the reading speed and almost identical.
I also exported video and photos on my ssd sandiksk extreme v2 and timed, the times are identical the internal ssd will not go faster than my sandisk so the 1tb will not be either.
Because AmorphousDiskMark tests every different things on sequential tests. On QD1 and 8. While on black magic it's only doing one kind of test in sequential
Understanding.
I just mean,which of the two is the closest to apple's testing ?
Cuz amorphous' results are very flattering .i mean the 512gb SSD reaching 6800-7000mbp/s is rather surprising isn't it ?
It's me again.mind sharing/telling what results you get using blackmagic ? I'd be very interested to compare the two and see how much "worse" blackmagic's results will be
Thanks !
It is more accurate if you're trying to compare it to Apple's way of testing. Look at my post here
Look at how it says IO Depth 8. It's generally the same as Queue depth 8 AFAIK. That's why it's more in line of what Apple is testing. If you want to be more precise. U can follow how Apple test their stuff. Like what they said above using FIO 3.27 with those same configurations. I didn't test using black magic tho since its not really what I need to test the SSD speeds. Here is a quote from this website
Q: How about Blackmagic Disk Speed Test? https://apps.apple.com/app/blackmagic-d ... d425264550
A: Blackmagic Disk Speed Test only runs sequential 5400 KiB (0x546000 bytes = 5529600 bytes = 5400 x 1024 bytes) block read/write tests. If you are only interested in a particular scenario (the data transfer rate your storage device/controller/interface can achieve from a single-threaded sequential read/write), there is nothing wrong with Blackmagic Disk Speed Test. With NVMe SSD devices, a single-threaded sequential read/write typically cannot achieve the best performance of the device. This is one of the reasons why CDM and ADM have the multi-threaded sequential read/write measurements.
Thanks !
But do I read it wrong or apple actually tested with 150gb file?? That's enormous !
Most people here seems to have tested with either 1 or 5gb .i tested with 1gb myself .
Yeah. Just to make sure it's consistent I guess. Although I don't think it'll be a good idea for us since. Well 1 the endurance we are wasting and also since SSDs has an SLC cache. And the lower the capacity usually the lower the size. So it'll be so fast until a point where it'll slow down a bit. Until it can recover . And since Apple is using 8TB SSD to test. 150gb won't ever hit the cache limit so yeah.
I think Max Tech has tried to check the SLC cache limit on the new MBPs but I forgot on which videos.. tbh I think just using dd command to write a 0 file should be able to let us know how much SLC cache there is.
Remember though that SLC cache size depends on the capacity that you bought (higher capacity like 8tb will have much more SLC cache limit than 512gb) and also it depends on current free space
You know,i don't think 150gb writing and reading can do anything to our SSD lol
So you're saying the 512/1tb might reach the SLC cache limit with a 150gb test file ?
What I meant is that already from 1 to 5gbp ,the difference was enormous (a guy in this thread got 4500 with 5gb,and 6700 with 1gb test file )
Do you think the 512gb option is a good deal or do you really need the 1tb option?
150gb itself won't do anything to our SSD. But it will add overtime to reduce our endurance..
As for the 1gb to 5gb file sizes. Idk. Mine isn't that different AFAIK.
IMO pick the storage that you need or want. Don't choose it because of the speed. It's not that significant tbh. I got the 1tb since i want it and I think I'm gonna use more than 512gb later on. Also since it's non upgradeable and even if I can use the sdcard slot or my nvme enclosure it's not gonna be the same as it being internal right? . The speed and the convenience too
I only use 120gb maximum the rest of my files that I no longer use are on my external drive. I think I made the right choice, but it was more for the writing speed that I asked myself the question