M1 disappointing...
Forum rules
An Activity Log is required for support requests. Please read How-to get an activity log? for details on how and why this should be provided.
An Activity Log is required for support requests. Please read How-to get an activity log? for details on how and why this should be provided.
M1 disappointing...
Hi, my gf just bought a m1 air, and i was more excited than she about this. I'm planning to shrink my whole collection of mkvs, however it takes ages on my imac 2019 (i5 3ghz) and i was pretty sure with an m1 i can almost half the required time. i mean, those tons of benchmarks speaks clear:
MacBook Air (Late 2020) vs iMac (27-inch Retina Early 2019) - Geekbench Browser
Ive tested a mkv fragment and result its very disappointing. m1 its 25% slower than the i5... ive checked hanbrake's log in order to see if there was some different settings and everything its the same apart this:
x265 [info]: frame threads / pool features : 2 / wpp(17 rows) on i5
x265 [info]: frame threads / pool features : 3 / wpp(17 rows) on m1
does anyone have idea what it means? every in-app setting was of course the same (exported/imported settings just to be sure)
MacBook Air (Late 2020) vs iMac (27-inch Retina Early 2019) - Geekbench Browser
Ive tested a mkv fragment and result its very disappointing. m1 its 25% slower than the i5... ive checked hanbrake's log in order to see if there was some different settings and everything its the same apart this:
x265 [info]: frame threads / pool features : 2 / wpp(17 rows) on i5
x265 [info]: frame threads / pool features : 3 / wpp(17 rows) on m1
does anyone have idea what it means? every in-app setting was of course the same (exported/imported settings just to be sure)
Re: M1 disappointing...
A software decoder/encoder requires a lot of finely tuned and hand written assembly simd code to be fast, and x265 hasn't got as many for arm as for Intel/AMD yet. Furthermore Intel has got a bunch of better simd instruction to accelerate x265, so I don't know how much x265 can still be optimized.
Anyway, a HandBrake snapshot should be a bit faster, try it: https://github.com/HandBrake/HandBrake-snapshots
And then your MacBook Air will probably slow down the cpu a bit due to overheating when running such a long task, it has got no fan.
Anyway, a HandBrake snapshot should be a bit faster, try it: https://github.com/HandBrake/HandBrake-snapshots
And then your MacBook Air will probably slow down the cpu a bit due to overheating when running such a long task, it has got no fan.
Re: M1 disappointing...
1) but do u know what means the string ive posted?Ritsuka wrote: ↑Wed May 25, 2022 5:07 am A software decoder/encoder requires a lot of finely tuned and hand written assembly simd code to be fast, and x265 hasn't got as many for arm as for Intel/AMD yet. Furthermore Intel has got a bunch of better simd instruction to accelerate x265, so I don't know how much x265 can still be optimized.
Anyway, a HandBrake snapshot should be a bit faster, try it: https://github.com/HandBrake/HandBrake-snapshots
And then your MacBook Air will probably slow down the cpu a bit due to overheating when running such a long task, it has got no fan.
2) what is this snapshot?
3) well, the speed difference was noticeable immediately. ive clicked "start" at the same time on both "cooled" systems and i5 was faster immediately, before even its fans activated
-
- Veteran User
- Posts: 4859
- Joined: Wed May 04, 2011 11:06 pm
Re: M1 disappointing...
Could you please post your HB logs, instructions can be found here:
https://handbrake.fr/docs/en/latest/hel ... y-log.html
https://handbrake.fr/docs/en/latest/hel ... y-log.html
Re: M1 disappointing...
https://forums.macrumors.com/threads/m1 ... t-31123665rollin_eng wrote: ↑Wed May 25, 2022 1:54 pm Could you please post your HB logs, instructions can be found here:
https://handbrake.fr/docs/en/latest/hel ... y-log.html
Re: M1 disappointing...
1) It's how many threads x265 uses, your M1 has got 8 cores (4 performance and 4 efficiency) so it will use more threads.
2) A snapshot is a version of HandBrake build from the latest code, so it contains all the latest changes and update to HandBrake, you can download it from the page I linked
3) Then it just means it's not optimised.
2) A snapshot is a version of HandBrake build from the latest code, so it contains all the latest changes and update to HandBrake, you can download it from the page I linked
3) Then it just means it's not optimised.
Re: M1 disappointing...
1) so more threads means it better right? 3(m1) vs 2(i5): m1 wins?Ritsuka wrote: ↑Wed May 25, 2022 3:09 pm 1) It's how many threads x265 uses, your M1 has got 8 cores (4 performance and 4 efficiency) so it will use more threads.
2) A snapshot is a version of HandBrake build from the latest code, so it contains all the latest changes and update to HandBrake, you can download it from the page I linked
3) Then it just means it's not optimised.
2) so its a "beta"? do these changes will be implented in future official updates? and how to install it?
Re: M1 disappointing...
1) No, it only means a different number is needed for the best performance on a particular cpu.
2) Yes these changes will be in the future releases. There are instructions on that page.
2) Yes these changes will be in the future releases. There are instructions on that page.
Re: M1 disappointing...
Your logs on the MR post show everything is running fine and nothing out of the ordinary going on.
As such, this is just M1 performance. Geek bench is not a reliable indicator of day to day performance across all applications. While the nightly might be a little faster I suspect it won't be able to overtake Intel on this one.
I'd maybe sure your connected to power on the off chance they are doing some funky power management on the air.
As such, this is just M1 performance. Geek bench is not a reliable indicator of day to day performance across all applications. While the nightly might be a little faster I suspect it won't be able to overtake Intel on this one.
I'd maybe sure your connected to power on the off chance they are doing some funky power management on the air.
Re: M1 disappointing...
what is "nightly"?s55 wrote: ↑Wed May 25, 2022 4:46 pm Your logs on the MR post show everything is running fine and nothing out of the ordinary going on.
As such, this is just M1 performance. Geek bench is not a reliable indicator of day to day performance across all applications. While the nightly might be a little faster I suspect it won't be able to overtake Intel on this one.
I'd maybe sure your connected to power on the off chance they are doing some funky power management on the air.
-
- Veteran User
- Posts: 4859
- Joined: Wed May 04, 2011 11:06 pm
Re: M1 disappointing...
Nightly = Snapshot
Re: M1 disappointing...
ok thanks to all for the answers. hope next HB versions will be more optimized for apple silicon
-
- Posts: 1
- Joined: Thu Jun 09, 2022 8:46 am
Re: M1 disappointing...
Interesting topic as I'm planning to buy a Macbook M2 and expected to have some substantial gains with x264/265 encoding.
Apple has included H264/265 hardware modules in its M1/2 chips so I was expecting those chips to match GPU hardware encoding. Should I understand it's not the case? If not is it because HandBrake does not make use of those units? (yet?)
Apple has included H264/265 hardware modules in its M1/2 chips so I was expecting those chips to match GPU hardware encoding. Should I understand it's not the case? If not is it because HandBrake does not make use of those units? (yet?)
Re: M1 disappointing...
x264/265 are software encoders, so a different GPU or different HW encoder won't matter at all.
HandBrake expose the hardware encoders, if you want to use them you can just select them in the encoders list, if not use the software encoders, they work quite well on the M1. The issue in this topic is that it lacks any factual data, so there is nothing to talk about.
If you want to compare some actual data, there are some reproducible benchmarks you can use to compare M1 to your computer on viewforum.php?f=9
HandBrake expose the hardware encoders, if you want to use them you can just select them in the encoders list, if not use the software encoders, they work quite well on the M1. The issue in this topic is that it lacks any factual data, so there is nothing to talk about.
If you want to compare some actual data, there are some reproducible benchmarks you can use to compare M1 to your computer on viewforum.php?f=9