macOS 10.13 High Sierra vs. Ubuntu Linux Performance

[Cape]

Right. They don't care about performance at all. Aside from forking KHTML and turning it into a world class web browser, developing Clang and many other open source projects, designing the fastest mobile SoC etc. They don't care about performance at all.

You're free to prefer any OS you'd like, but if you need to invent bullshit in order to rationalize your decision, you have a problem.

CLANG is still overall inferior to GCC. And anyway, they are funding (funding != developing) CLANG because GCC has a more freedom respecting license, nothing to do with perfs.

Safari suck balls and it's not fast at all.

Also they didn't design any SoC, they just went shopping in Shenzeng with a bag full of capital investment money and started selling super-over-priced HW to IT illiterate yuppies thanks to an excellent marketing campaign (again, investment munny).

Apple is not good at engineering, but Excell in marketing. Also dumb-dumb ppl with lots of dling-dling!

[carewolf]

Wow, do you really think he meant the old Mac OS or did you just want to shoot his mouth off?!

Mac OS is old. The new one is called macOS.

[freedom]

In audio benchmarks macOS would totally dominate Ubuntu. This is the last big technical hurdle to overcome before Linux can become a mainstream OS. I wish Phoronix did audio benchmarks and thus pushed Linux forward in this area.

[duby229]

In audio benchmarks macOS would totally dominate Ubuntu. This is the last big technical hurdle to overcome before Linux can become a mainstream OS. I wish Phoronix did audio benchmarks and thus pushed Linux forward in this area.

I so, so very much agree. Somebody very public needs to highlight that.

[freedom]

A few thoughts on possible audio benchmarks

I think latency is a keyword because it destroys the experience just like low FPS destroys the experience. VoIP is a good stress-test for audio. Anything that makes the audio system switch between sources and thus creates potential for measurable and audible latency.

I think a good test would be to feed the computer sound at intervals (through a simulated microphone) and also let the computer playback sound at intervals. Like VoIP essentially. This sound exchange is recorded and measured against the original transmission. You’’ll find that the computer can’t keep up with the original transmission. There will always be latency/lag.

If the latency is big enough the sounds will start mixing. In a VoIP conversation this causes confusion and people start speaking at the same time.

I don’t know of any audio tests, but I imagine it wouldn’t be too hard to measure the computer output vs the original transmission. Of course it would take someone to program this test. I imagine it already exists and that every (smartphone) audio engineer use it while designing/tweaking smartphone hardware and software.

Think VoIP but in this test sound sources are switched in milliseconds.

Then there are other more traditional tests that could be performed (although less important):

1. Fidelity of sound output (is there internal remixing, is it accurate?)

2. Volume gain/loss/noise level (comparing several OSes on same hardware)

3. Is the audio system consistent or does the audio system fluctuate in latency, speed etc.?

4. Can the audio system stay in sync while performing demanding tasks like video editing?

5. Various sound distortions like crackling etc.

All I really want is that Ubuntu and other Linux distros get closer to the out-of-the-box audio experience that macOS and Windows offer. Right now Ubuntu is the black sheep and best avoided if audio performance is a concern.

[carewolf]

A few thoughts on possible audio benchmarks

I think latency is a keyword because it destroys the experience just like low FPS destroys the experience. VoIP is a good stress-test for audio. Anything that makes the audio system switch between sources and thus creates potential for measurable and audible latency.

I think a good test would be to feed the computer sound at intervals (through a simulated microphone) and also let the computer playback sound at intervals. Like VoIP essentially. This sound exchange is recorded and measured against the original transmission. You’’ll find that the computer can’t keep up with the original transmission. There will always be latency/lag.

If the latency is big enough the sounds will start mixing. In a VoIP conversation this causes confusion and people start speaking at the same time.

I don’t know of any audio tests, but I imagine it wouldn’t be too hard to measure the computer output vs the original transmission. Of course it would take someone to program this test. I imagine it already exists and that every (smartphone) audio engineer use it while designing/tweaking smartphone hardware and software.

Think VoIP but in this test sound sources are switched in milliseconds.

Then there are other more traditional tests that could be performed (although less important):

1. Fidelity of sound output (is there internal remixing, is it accurate?)

2. Volume gain/loss/noise level (comparing several OSes on same hardware)

3. Is the audio system consistent or does the audio system fluctuate in latency, speed etc.?

4. Can the audio system stay in sync while performing demanding tasks like video editing?

5. Various sound distortions like crackling etc.

All I really want is that Ubuntu and other Linux distros get closer to the out-of-the-box audio experience that macOS and Windows offer. Right now Ubuntu is the black sheep and best avoided if audio performance is a concern.

You probably wouldn't find any significant benifit of macOS under such VoIP conditions. You would do better with something like MIDI response timing. Linux can do things like that efficiently but it doesn't do any of it out of the box, so macOS or Windows would win any such competition. And it is only timing that is worth testing, the rest is either hardware or the specific software processing the audio, there are no distortions or fidelity involved anywere in what an OS does with sound.

[Apopas]

In audio benchmarks macOS would totally dominate Ubuntu.

Where do you base this?

[freedom]

Where do you base this?

Practical experience, but practical experience is not enough and that's why I would like to see tests performed.

[carewolf]

Where do you base this?

Windows and macOS gives real-time priority to sound processes to lower latency and in old days with slower hardware to avoid skipping due to empty buffers. By default Linux does not give any preferential treatment to audio.

[Guest]

You are very ignorant. It was close if anything. This is quite literally a free improvement going to apfs. Migrate filesystem, catch up to the performant Linux.

It seems you are ignorant here. Ext4 with fsync vs unsafe mac os file system. This makes PostgreSQL, SQLite results meaningless.