Im running the GM on a MacBook Pro 15" Early 2015 and everything seems fine. Upgraded my drive and all. Anybody have any speed test numbers for our new filesystem? What did we just do that for? ;-)
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.
APFS Benchmarks?
- Thread starter RumorConsumer
- Start date
- Sort by reaction score
Wow thank you! I dont actually have a baseline to compare it to, as I said I already upgraded my drive without running a benchmark. Anybody else?
Right so are there specific cases where we can observe the benefits? What do I have to do to have APFS show me what it’s got?
I would except I dont have a baseline to compare to. The results are meaningless without a baseline.
Geekbench might be an indicator and then you can compare your scores to someone else with a similar configured system.
Let us provide you your baseline then? Just do a test and post the results...
Cheers
Ok when somebody with an Early 2015 2.5ghz MacBook Pro 15" writes on this thread Ill do it. No baseline provides no meaningful data, similarly comparing non-like models also confounds the data.
I think you're really missing the point using a forum here! Why claim something first before you're willing to provide us some info's to answer your Q?
Opting out...
Respectfully I think you dont understand how benchmarks work. If there is no comparison or no controlled baseline it doesnt help anybody if I run a benchmark. Its also not useful to compare between models or even between specific units because we are using different hardware. What we need is somebody with a machine who can run the test before upgrading and then after. That will tell us something interesting and useful and thats why I started this thread.
"blackmagic speed test" only tells half of the truth (at best). a proper disk-benchmark needs to do measurements using different block-sizes. like QuickBench ftom Intech.
found this post a few weeks back ->
https://malcont.net/2017/07/apfs-and-hfsplus-benchmarks-on-2017-macbook-pro-with-macos-high-sierra/
[doublepost=1505980452][/doublepost]my own findings... APFS being a tad slower across the board. interestingly, APFS was a little bit faster when I measured an HDD instead of an SSD.
found this post a few weeks back ->
https://malcont.net/2017/07/apfs-and-hfsplus-benchmarks-on-2017-macbook-pro-with-macos-high-sierra/
[doublepost=1505980452][/doublepost]my own findings... APFS being a tad slower across the board. interestingly, APFS was a little bit faster when I measured an HDD instead of an SSD.
Last edited:
I haven't heard many claims that APFS is faster by Apple or users. I believe the advantages are a more efficient file system for SSD devices and a means of writing data or moving data in a way that will also extend the life of an SSD. In addition, security, data integrity and stability have has been mentioned.
I am no expert, but I have not noticed any increase in speed.
I am no expert, but I have not noticed any increase in speed.
I very much doubt that APFS is going to be faster than HFS+ in standard benchmarks. I’d actually expect it to be slower. It’s strength is not throughout but latency and laziness. It achieves its speed by avoiding to do stuff. In virtually every scenario, APFS will feel much snappier, except maybe when you are copying karge amounts of data over very fast networks.
this thread is about the new filesystem APFS, not about CPU benchmarks.
[doublepost=1506027947][/doublepost]
actually, latency on APFS seems to be higher by about 10% according to the article I posted a link to earlier.
So the higher benchmark scores are not a result of APFS?
Also, there was talk earlier in the thread about comparing Geekbench scores, that's why I posted the video, as it has already been done. Cheers.
So the higher benchmark scores are not a result of APFS?
seriously? geekbench scores have nothing to do with the new filesystem. this youtuber "adrian" has no clue what he's doing and also he talks sh*t. geekbench scores under macOS Sierra are the same as they are when running on High Sierra. this guy runs benchmarks (CPU/SSD) while screenrecording at the same time. as I said earlier: blackmagic speed test tells you only a fraction of the truth. in the end of the video, this "adrian" says "let's see what APFS does to NTFS formatted volumes, see if we can write to an NTFS volume?".
oh, dear! I don't know where to begin with...a) he only uses blackmagic speed test. sure, I read/write 2.0 GB files all day long... really?
b) he duplicates files in Finder.app and claims that they indeed are being duplicated on APFS. sure...
c) during the benchmarking (CPU/SSD), he's screen recording the whole thing. of course, this is not affecting the results...
d) exporting from iMovie: my guess is that High Sierra somehow uses new intel QuickSync tech that wasn't used under Sierra. as I wrote earlier: geekbench scores (CPU) are the same on Sierra/High Sierra.
e) while in Sierra, the MacBook runs on battery. during High Sierra, the MacBook is connected to the charger.
my advice: stay away from this YT channel. at least the video you linked to is nonsense from the beginning to the end.
Yowza.seriously? geekbench scores have nothing to do with the new filesystem. this youtuber "adrian" has no clue what he's doing and also he talks sh*t. geekbench scores under macOS Sierra are the same as they are when running on High Sierra. this guy runs benchmarks (CPU/SSD) while screenrecording at the same time. as I said earlier: blackmagic speed test tells you only a fraction of the truth. in the end of the video, this "adrian" says "let's see what APFS does to NTFS formatted volumes, see if we can write to an NTFS volume?".
a) he only uses blackmagic speed test. sure, I read/write 2.0 GB files all day long... really?
b) he duplicates files in Finder.app and claims that they indeed are being duplicated on APFS. sure...
c) during the benchmarking (CPU/SSD), he's screen recording the whole thing. of course, this is not affecting the results...
d) exporting from iMovie: my guess is that High Sierra somehow uses new intel QuickSync tech that wasn't used under Sierra. as I wrote earlier: geekbench scores (CPU) are the same on Sierra/High Sierra.
e) while in Sierra, the MacBook runs on battery. during High Sierra, the MacBook is connected to the charger.
my advice: stay away from this YT channel. at least the video you linked to is nonsense from the beginning to the end.
seriously? geekbench scores have nothing to do with the new filesystem. this youtuber "adrian" has no clue what he's doing and also he talks sh*t. geekbench scores under macOS Sierra are the same as they are when running on High Sierra. this guy runs benchmarks (CPU/SSD) while screenrecording at the same time. as I said earlier: blackmagic speed test tells you only a fraction of the truth. in the end of the video, this "adrian" says "let's see what APFS does to NTFS formatted volumes, see if we can write to an NTFS volume?".
a) he only uses blackmagic speed test. sure, I read/write 2.0 GB files all day long... really?
b) he duplicates files in Finder.app and claims that they indeed are being duplicated on APFS. sure...
c) during the benchmarking (CPU/SSD), he's screen recording the whole thing. of course, this is not affecting the results...
d) exporting from iMovie: my guess is that High Sierra somehow uses new intel QuickSync tech that wasn't used under Sierra. as I wrote earlier: geekbench scores (CPU) are the same on Sierra/High Sierra.
e) while in Sierra, the MacBook runs on battery. during High Sierra, the MacBook is connected to the charger.
my advice: stay away from this YT channel. at least the video you linked to is nonsense from the beginning to the end.
WOW. Okay,
a) I said nothing about read/write speeds, although there seems to be an improvement here as well. So how do you explain it? I guess Apple built in some top secret technology in High Sierra and it's not really a result of APFS.
b) He literally did the absolute same thing that Craig Federighi did in his demo at WWDC. Why are you trying to disprove everything?
c) He is screen recording both times (Sierra and High Sierra) so don't you think the results should be affected the same amount?
d) Okay, now I understand that you clearly have an answer for everything so there's really no point in arguing with you.
e) That I give you, hadn't noticed.
Also, I'm not trying to start a big argument, I just find some of your points ridiculous even if there isn't an improvement in scores as you say. The only thing that could affect scores in this particular video is that when the machine is running on High Sierra it seems to be plugged in. Otherwise, I'd say it is a pretty fair comparison.
Last edited:
you got that results becault quickbench is outdated and not optimised for fast (pci-e) ssd. the test file size is so small that the whole test finishes in 1-2 second (on my MBA 13" 2015 , it has 512GB 4x ssd that does 1.5GB/s r/w) , so it could not represend real speed.