HQplayer 4 and Mac-mini M1

[Miska]

8 hours ago, pis99 said:

Yes, 7ec-sin_L does work in M1 as I listen to mostly original DSD files. If you want to do 44.1 to 7ec-sin-L, you can achieve this with another upsampling of 44.1 to 64 DSD(eg. Roon) before feeding HQplayer. 

 

sinc-L (or any of the other oversampling filters) is not used for DSD sources...

 

Doing any upsampling in Roon before HQPlayer would make HQPlayer's upsampling moot. Especially if you upsample to DSD in Roon you would have notable performance degradation.

 

[pis99]

16 hours ago, Miska said:

 

sinc-L (or any of the other oversampling filters) is not used for DSD sources...

 

Doing any upsampling in Roon before HQPlayer would make HQPlayer's upsampling moot. Especially if you upsample to DSD in Roon you would have notable performance degradation.

 

Thank you Miska, Yes, I did seem to remember that you mentioned using Roon upsampling before HQplayer is not a wise choice. I am also clear now that oversampling filters are not effective for DSD sources.  Very much appreciated!

[Miska]

There are two rate conversion options and number of integrators for DSD-to-DSD conversions in "DSD Sources" dialog. Load-wise these are not much different.

 

[MikePid]

I’ve joined the M1 HQPlayer club 😀.  I picked up a refurbished 8GB from Apple.

 

I’ve set up the M1 exclusively for HQP, with Roon on my 2012 rMBP, and ultraRendu NAA.  It was easy to set up, and I’ll post my setup instructions when I get a chance.  Basically, I turned off all the good Apple stuff, so that HQP is the only thing running on it (plus Activity Monitor for now so I can watch CPU load).  I set it up with a separate Apple ID so I could configure it for auto-login and screen sharing without worrying about security in case someone steals it.

 

I transferred the HQP license, and now have it set to DSD256, with ASDM5EC, and ext3 for 1x.  Wow!  Jussi hit it out of the park with this on the M1.  75% CPU with no stutters.  90% on the performance cores, and 60% on the efficiency cores.  The M1 is barely warm.  The sound is so realistic it is scary!  Thanks Jussi!  For nX, I needed to set it to ext2 to get no stutters.
 

I first started with my max settings on the MacBook Pro i7 (DSD256/5V2 256+fs/ext2) and CPU usage was about 25% on main cores, and zero on efficiency cores.  On my laptop, fans were running high (using TGpro) to keep temps just under 80C. No fans kicking in (that I can hear) on the M1 with 5EC and ext3.  7EC gaps every 10s with ext3 though.  So when I get a chance I’ll compare 5EC/ext3 with 7EC/ext2 to see which I prefer.  What a luxury to have this choice!

 

For me, this was a no-brainer upgrade!

 

[BrownMagic]

On 2/7/2021 at 9:14 PM, giordy60 said:

a question
is there any user who with a mac mini M1 (+ NAA) and a dac holo audio or denefrips has tried the upsampling in pcm 1.411-1.536 mhz?
thank you

Hello Did you manage to get 1.536 using HQP on Mac Mini and NAA? Which filters work? Will the Sync MX filter work?

[giordy60]

I don't have the M1,    🤷‍♂️......I was asking for a friend.

[AudioDoctor]

It works in MacOS Monterey as well.

 

 

[Bushikai]

Hello MIkePid, What version of HQplayer are you using? I don't have poly- sinc-ext3 in my preferences.

 

Thank you.

[BrownMagic]

On 1/26/2021 at 8:20 PM, pis99 said:

Make sure you have NAA to to work with OSX.

What NAA works?

[pis99]

20 hours ago, BrownMagic said:

What NAA works?

I have this one that is suggested by Miska:

https://up-shop.org/up-gateway-atom-x5-z8350-w-4g-memory-32g-emmc-board-w-vesa-plate.html

 

I also have a sonore microrendu that works as NAA for Hqplayer too.

 

Hope this help.

 

[BrownMagic]

Just now, pis99 said:

also have a sonore microrendu that works as NAA for Hqplayer too.

Thanks. Does the Ultra Rendu support the Sinc MX filter? Which is 1.5m. 

[AudioDoctor]

1 hour ago, BrownMagic said:

Thanks. Does the Ultra Rendu support the Sinc MX filter? Which is 1.5m. 

 

Why wouldn't it?

[BrownMagic]

3 minutes ago, AudioDoctor said:

Why wouldn't it?

Because I was told that only intel based NAA can do the 1.5M filters. 

[AudioDoctor]

19 minutes ago, BrownMagic said:

Because I was told that only intel based NAA can do the 1.5M filters. 

 

The NAA doesn't do any filtering or processing, it is a Network Audio Appliance, NAA, and all it does is accept a data stream of music from the machine that is running HQPlayer where the processing is done, and output it via USB or something to the DAC. A Raspberry Pi can do it.

[BrownMagic]

4 minutes ago, AudioDoctor said:

The NAA doesn't do any filtering or processing, it is a Network Audio Appliance, NAA, and all it does is accept a data stream of music from the machine that is running HQPlayer where the processing is done, and output it via USB or something to the DAC. A Raspberry Pi can do it.

 

Thanks. That was my understanding too but I believe some NAA can only handle certain data stream. Maybe @miska

can confirm. Thanks

[AudioDoctor]

4 minutes ago, BrownMagic said:

 

Thanks. That was my understanding too but I believe some NAA can only handle certain data stream. Maybe @miska

can confirm. Thanks

 

If a raspberry Pi can accept anything up to DSD 512 and PCM 768 then I am sure you'll be okay. I am listening to HQP right now and using the sinc-mx filter at 768 and it works just fine. The filter you use has no bearing on the ability of the NAA to output a stream.

[BrownMagic]

1 minute ago, AudioDoctor said:

The filter you use has no bearing on the ability of the NAA to output a stream.

Thanks. I think what I wanted to know was what NAA can do 1.536mhz Sinc-Mx for PCM. I asked around and I was told RPI cannot support this neither can a M1 mac hence the question around Intel based NAA

[AudioDoctor]

1 minute ago, BrownMagic said:

Thanks. I think what I wanted to know was what NAA can do 1.536mhz Sinc-Mx for PCM. I asked around and I was told RPI cannot support this neither can a M1 mac hence the question around Intel based NAA

 

Well I dont think an M1 MAC can be an NAA at any sample rate. I know Miska has a link to what he suggests at his website, have you been there?

 

https://www.onlogic.com/eu-en/computers/industrial/fanless/cl100-series/

 

and 

 

https://up-shop.org/up-gateway-atom-x5-z8350-w-4g-memory-32g-emmc-board-w-vesa-plate.html

[BrownMagic]

2 minutes ago, AudioDoctor said:

Well I dont think an M1 MAC can be an NAA at any sample rate. I know Miska has a link to what he suggests at his website, have you been there?

Ah yes. Both are mini pcs which I know work. I have an Ultra Rendu with me now and I wanted to know it can do the same thing what the mini pc that miska recommended can do. Because as of now I do not have HQP installed so I am not able to test. Cheers Doc!

[AudioDoctor]

38 minutes ago, BrownMagic said:

Ah yes. Both are mini pcs which I know work. I have an Ultra Rendu with me now and I wanted to know it can do the same thing what the mini pc that miska recommended can do. Because as of now I do not have HQP installed so I am not able to test. Cheers Doc!

 

A Rendu can not. I have a Sonore unit and if I remember correctly they only go up to DSD 512

[camott]

Can someone do me favor and see if the new poly-sinc-gauss-xla can run at 1.536mhz PCM on the M1 Mac Mini? Thanks in advance!

[camott]

On 7/15/2021 at 3:44 PM, camott said:

Can someone do me favor and see if the new poly-sinc-gauss-xla can run at 1.536mhz PCM on the M1 Mac Mini? Thanks in advance!

 

So I ran some HQPlayer performance tests on my i9-9980HK MacBook Pro 16 using the null output. 24/192 → 1536 with gauss-xla could not quite do 1x processing speed with no multicore DSP (2 cores). With multicore DSP set it used 4 cores and was able to process almost 2x speed. Given that the M1 single core is about 25% faster than my 9th gen Intel i9, it should be able to do gauss-xla at 1x speed a bit under 50% utilization on the 4 high-power cores. Maybe even possibly without multicore DSP but the two cores would likely be pegged. Next step is to order one and actually try it out.

[Miska]

8 hours ago, camott said:

So I ran some HQPlayer performance tests on my i9-9980HK MacBook Pro 16 using the null output. 24/192 → 1536 with gauss-xla could not quite do 1x processing speed with no multicore DSP (2 cores). With multicore DSP set it used 4 cores and was able to process almost 2x speed.

 

Multicore DSP should be set to grayed (auto) which usually gives best overall performance in playback cases.

 

8 hours ago, camott said:

Given that the M1 single core is about 25% faster than my 9th gen Intel i9

 

Where does this figure come from? At least it is not the case comparing my M1 Mac Mini to my iMac with i9-9900K, where single core of 9900K is about 2x faster than single performance core on M1, under HQPlayer loads.

 

[camott]

8 hours ago, Miska said:

 

Multicore DSP should be set to grayed (auto) which usually gives best overall performance in playback cases.

 

 

Where does this figure come from? At least it is not the case comparing my M1 Mac Mini to my iMac with i9-9900K, where single core of 9900K is about 2x faster than single performance core on M1, under HQPlayer loads.

 

 

So I am unclear on auto multicore vs forced multicore ...

 

Below are the performance numbers I got in each of 3 modes for a 24/192 file -> 1.536M using poly-sinc-gauss-xla + LNS15 output to the null device. The play time of the 24/192 file is 4:55. Again, cpu was i9-9980HK in MacBook Pro 16.

 

no multicore - took 5:42 to process 4:55 file with two cores fully maxed. Obviously not fast enough.

 

97.9% CPU

98.8% CPU

 

multicore forced (checked) - took 3:11 to process file with 4 cores around 90% utilized and 2 partially utilized. sample:

 

9.0%

9.1%

86.8%

87.5%

87.5%

88.6%

 

auto multicore (grey) - took 2:38 to process file with 6 cores around 80% utilized and 2 partially. 

 

14.7%

14.9%

80.1%

80.4%

80.4%

80.5%

81.5%

82.0%

 

 

So, auto multicore used roughly 50% more CPU for a pretty small increase in performance. I got very similar results with 16/44.1k -> 1.411M. (Note that I have tried these tests multiple times and the results are consistent). What I don't understand is why auto multicore uses 2 additional cores than forced multicore.

 

Regarding performance of M1 core vs i9 9th gen - this was based purely on various benchmarks I have seen comparing M1 vs i9-9980HK, be it cinebench or geek bench or whatever. They all suggest that the M1 is slightly faster in single core tests. But they also give similar results for your i9-9900K. So not entirely sure why your real world tests with HQPlayer suggest otherwise - I suppose those other benchmarks are testing more than pure cpu which is what HQPlayer is all about (vs memory bus etc)? One thing I will note is that your i9-9900K has a baseline frequency of 3.6Ghz whereas my i9-9980HK for above tests has a base frequency of 2.4Ghz.

 

Can your M1 do 1.411/1.536M PCM with guass-xla??

[Miska]

9 minutes ago, camott said:

What I don't understand is why auto multicore uses 2 additional cores than forced multicore.

 

Auto is "smart mode" while forced is "dumb mode". In most cases, auto should give best results. But the dumb mode is there just in case.

 

9 minutes ago, camott said:

Regarding performance of M1 core vs i9 9th gen - this was based purely on various benchmarks I have seen comparing M1 vs i9-9980HK, be it cinebench or geek bench or whatever. They all suggest that the M1 is slightly faster in single core tests. But they also give similar results for your i9-9900K. So not entirely sure why your real world tests with HQPlayer suggest otherwise - I suppose those other benchmarks are testing more than pure cpu which is what HQPlayer is all about (vs memory bus etc)? One thing I will note is that your i9-9900K has a baseline frequency of 3.6Ghz whereas my i9-9980HK for above tests has a base frequency of 2.4Ghz.

 

Those benchmarks are not applicable to HQPlayer workloads. Actually I have not seen any synthetic benchmark that would be directly applicable to HQPlayer. Even more so when you mix in realtime processing and modulator loads like ASDM7EC. That is why I added kind of benchmarking functionality to HQPlayer with null outputs.

 

15 minutes ago, camott said:

Can your M1 do 192 -> 1536 PCM with guass-xla??

 

My only 1.5M capable DAC is Spring 2 which is at different location than my M1 Mac Mini. If Spring 3 will still have that 1.5M capability, I will test with it once I get one (hopefully within a month or so).