HQ Player

[bibo01]

3 hours ago, SwissBear said:

Whaoo Jussi ! Congratulations ! The Mac Mini M1 is able to upsample DSD 128 to DSD 256 with ASDM7EC and apply convolution.

And switching from one song to the other is so quick and smooth :-) :-)

I can't believe it !

Sincere gratitude for making this happen so quickly.

 

I suppose you are talking about the new amd64 macOS version, aren't you?!

[Mahler and Bach on Computer]

3 hours ago, SwissBear said:

Whaoo Jussi ! Congratulations ! The Mac Mini M1 is able to upsample DSD 128 to DSD 256 with ASDM7EC and apply convolution.

And switching from one song to the other is so quick and smooth :-) :-)

I can't believe it !

Sincere gratitude for making this happen so quickly.

 

Just saw the link of “arm64” on the web site.  I just placed the order of Mac Mini M1 this afternoon.  

[jamesg11]

On 12/6/2020 at 3:37 AM, Mahler and Bach on Computer said:

I recently updated the OS of a 7-year old MacBook Pro (I believed is 4-core CPU) to the latest version of Catalina.  Under the previous versions of Catalina, the highest level I could go is DSD128.  Now I can run DSD256 easily without any breaks.  The updating of HQP4D from 4.7 to 4.8 is another improvement on performance.  The percentage of idle CPU increased from 18% to 36%.  

Got round to updating my MBPro (2013, 2.4GHz Quad-Core Intel i7, 16GB) to 10.15.7 ... but no, still can only do 128 without constant issues.

[Sevenfeet]

1 hour ago, Mahler and Bach on Computer said:

Just saw the link of “arm64” on the web site.  I just placed the order of Mac Mini M1 this afternoon.  

 

Wow that was fast. Didn't Jussi just get his Mac Mini a few days ago?

[jamesg11]

1 hour ago, Mahler and Bach on Computer said:

Just saw the link of “arm64” on the web site.  I just placed the order of Mac Mini M1 this afternoon.  

Hmm ... So this means it can definitely run dsd256 at asdm7ec with say poly-sinc-ext2 ...?

Or ...?

[AudioDoctor]

Are you telling me there HQPlayer for the M1 based Macs is already out and working well?

wow...

[Mahler and Bach on Computer]

59 minutes ago, jamesg11 said:

Got round to updating my MBPro (2013, 2.4GHz Quad-Core Intel i7, 16GB) to 10.15.7 ... but no, still can only do 128 without constant issues.

 

59 minutes ago, jamesg11 said:

Got round to updating my MBPro (2013, 2.4GHz Quad-Core Intel i7, 16GB) to 10.15.7 ... but no, still can only do 128 without constant issues.

I only used ASDM7, not ASDM7EC, to do DSD256.  Mine MBP is i5 2.?GHz, dual-core, with 8GB ram.  I tried EC, not work at all.  Before all the updates, I was only able to do DSD128.  

[ted_b]

6 hours ago, SwissBear said:

Whaoo Jussi ! Congratulations ! The Mac Mini M1 is able to upsample DSD 128 to DSD 256 with ASDM7EC and apply convolution.

And switching from one song to the other is so quick and smooth :-) :-)

I can't believe it !

Sincere gratitude for making this happen so quickly.

 

 

Are you saying a new $699 Mac Mini M1 (Best Buy, etc, more $$ for larger memory and storage) can do DSD256 with ASDM7EC?    Have you tried 64 to 256, and how much is your convolver (which I may not need) affecting performance.....IOW, can you run a simple test for me and do 64 to 256, no convolution?  :)  I only do DSD to DSD and PCM to PCM with my current dacs, so I am VERY interested in this M1 idea as a server (and I purchased a NUC 7i7DNHE for an NAA)

[Mahler and Bach on Computer]

48 minutes ago, jamesg11 said:

Hmm ... So this means it can definitely run dsd256 at asdm7ec with say poly-sinc-ext2 ...?

Or ...?

My brief experience with Mac Mini i7 6 core (2018), with 16GB ram was that it was able to run ASDM7EC at DSD256 for HQP4D.  I recalled the % CPU was somewhere around 210%, without fan being activated. 

[Mahler and Bach on Computer]

I did try to load matrix under this demanding condition, and it did not work.  However, I believe some of the current effort by @Geoffrey Armstrong can creatively solve this problem.

[Uni]

On 11/19/2020 at 5:21 PM, Miska said:

 

Same thing on all platforms. Loopback interfaces may work, but have the limitation that input rate needs to be changed manually. On Linux automatic rate switching can work through digital inputs without separate input NAA, using RME ADI-2 (Pro) or UAC2.

 

 

Did your Mac mini arrive?  How's the performance of HQPlayer 4.8 desktop on it?  Can it run DSD upsampling with EC modulator, up to 256, or even 512?  Will it kick off the fans noise?

[SwissBear]

Hi everyone,

 

To answer a few questions:

- I mentioned upsampling D128 to D256 was flawless as this was the most demanding operation in the emulation mode

- upsampling from DSD 64 to DSD 256 with ASDM7EC is flawless too...

- my convolution is very demanding (524 ktaps impulses at 192 kHz) as I have to use a very steep filter to prevent modes in my room

- consumption of resources is shown here:

 

 

 

 

 

So congratulations and thanks again to Jussi :-)

[jamesg11]

R

21 minutes ago, SwissBear said:

Hi everyone,

 

To answer a few questions:

- I mentioned upsampling D128 to D256 was flawless as this was the most demanding operation in the emulation mode

- upsampling from DSD 64 to DSD 256 with ASDM7EC is flawless too...

- my convolution is very demanding (524 ktaps impulses at 192 kHz) as I have to use a very steep filter to prevent modes in my room

- consumption of resources is shown here:

 

 

 

 

 

So congratulations and thanks again to Jussi :-)

Really promising then!

Waiting for Jussi to get involved in this teasing out of where we’ve got to here ... before ordering!

[k6davis]

1 hour ago, SwissBear said:

To answer a few questions:

- I mentioned upsampling D128 to D256 was flawless as this was the most demanding operation in the emulation mode

- upsampling from DSD 64 to DSD 256 with ASDM7EC is flawless too...

- my convolution is very demanding (524 ktaps impulses at 192 kHz) as I have to use a very steep filter to prevent modes in my room

 

Well, this is certainly exciting. Thanks for your detailed post.

So, @SwissBear, it sounds like you're saying that, with the newly optimized HQP, your M1 can upsample all commonly available file formats to DSD256 ASDM7EC with poly-sync-ext2? With demanding convolution applied? Is that correct?

 

I wonder if it's also capable of upsampling PCM 48x rates (48k, 96k and 192k) to "48k DSD" DSD256 EC . That's the most demanding load for my server.

 

Anyway, based on these initial impressions, the base Apple Silicon may be as HQP performant as the higher end Intel/AMD chips. But at much lower cost and with potentially greater efficiency and less cooling needed. 

 

I can't wait to learn more about the M1's HQP performance! Especially since it has been confirmed that even more powerful Apple Silicon machines are going to be released in 2021.

[Miska]

3 hours ago, Uni said:

Did your Mac mini arrive?  How's the performance of HQPlayer 4.8 desktop on it?  Can it run DSD upsampling with EC modulator, up to 256, or even 512?  Will it kick off the fans noise?

 

Yes, it arrived. At least for me, with native version, upsampling with EC modulators to DSD256 doesn't work (from RedBook, with poly-sinc-ext2). Output to a NAA. With regular modulators it does work.

 

Fan is running, but it is still fairly quiet, so could be used in a listening space.

 

So the performance is roughly what I expected, similar to equivalent U-series Intel CPUs. Over time Apple will likely come up with bigger and more powerful versions to match the bigger Intel CPUs in performance.

 

[Uni]

6 hours ago, SwissBear said:

Hi everyone,

 

To answer a few questions:

- I mentioned upsampling D128 to D256 was flawless as this was the most demanding operation in the emulation mode

- upsampling from DSD 64 to DSD 256 with ASDM7EC is flawless too...

- my convolution is very demanding (524 ktaps impulses at 192 kHz) as I have to use a very steep filter to prevent modes in my room

- consumption of resources is shown here:

 

 

 

 

 

So congratulations and thanks again to Jussi :-)

 

4 hours ago, Miska said:

 

Yes, it arrived. At least for me, with native version, upsampling with EC modulators to DSD256 doesn't work (from RedBook, with poly-sinc-ext2). Output to a NAA. With regular modulators it does work.

 

Fan is running, but it is still fairly quiet, so could be used in a listening space.

 

So the performance is roughly what I expected, similar to equivalent U-series Intel CPUs. Over time Apple will likely come up with bigger and more powerful versions to match the bigger Intel CPUs in performance.

 

 

You two's findings are different, isn't it? 

[ted_b]

4 minutes ago, Uni said:

 

 

You two's findings are different, isn't it? 

Well, Swissbear is not reporting what PCM (namely, redbook, in Jussi's example) to DSD256 is doing, only DSD to DSD.  I personally don't care, cuz I stay within formats, but others may care a lot!

[Uni]

31 minutes ago, ted_b said:

Well, Swissbear is not reporting what PCM (namely, redbook, in Jussi's example) to DSD256 is doing, only DSD to DSD.  I personally don't care, cuz I stay within formats, but others may care a lot!

 

Thanks.  I want to stay within formats too.  But the sound quality improvement from HQPlayer upsampling makes this impossible!

[ted_b]

5 minutes ago, Uni said:

 

Thanks.  I want to stay within formats too.  But the sound quality improvement from HQPlayer upsampling makes this impossible!

Oh...don't get me wrong..I upsample everything!  I take all PCM to 352/384 and all DSD to 512 (but once my room is done I will update to EC modulators and do EC at 256).

[shadowlight]

On 12/7/2020 at 10:20 PM, bibo01 said:

In Audirvana how do you send to HQPlayer Desktop without embedded? By rerouting signal?! 

I thought Audirvana also supported HQP API calls and that was the reason for my comment.  If that is not the case apology for the misinformation and the only option is HQP embedded.

[jamesg11]

9 hours ago, Miska said:

Over time Apple will likely come up with bigger and more powerful versions to match the bigger Intel CPUs in performance.

Yep, ... wonder how long that be then ... so, for the present, back to a new Intel build plan.

[Miska]

5 hours ago, ted_b said:

Oh...don't get me wrong..I upsample everything!  I take all PCM to 352/384 and all DSD to 512 (but once my room is done I will update to EC modulators and do EC at 256).

 

...which only makes sense with DACs that actually have two native converters. Which is pretty much just Holo Audio and Denafrips...

 

[ted_b]

9 minutes ago, Miska said:

 

...which only makes sense with DACs that actually have two native converters. Which is pretty much just Holo Audio and Denafrips...

 

Yes, thanks, my Holo LOVES PCM as PCM and DSD as DSD.  And my computer muscles love it, too, as there is less strain when doing that kind of upsampling. 

 

Although I also do a lot of multichannel, my "don't convert" rules still apply, and upsampling is more like 2x at most.  But I worry that there are no real good multichannel benchmarks or experiences to know the cpu needs, especially for me cuz my system is not really mch functional right now, so I can't even say what my i7-6700k measures at.  And those measurements would be with HQP 3 and therefore no EC work.  I was planning on waiting to buy HQP 4 when my room is done.  :(  Thinking out loud, maybe I should move that up and and see what happens.  My i7 server and Holo  is in my RAAL headphone system right now.  I assume multichannel upsampling loads on OS (Windows 10 right now) would be legitimate to capture, even if I don't have my multichannel exaSound unboxed to connect to it?  Would HQP run if it doesn't see a 6 channel dac?

 

Thx

[Sevenfeet]

3 minutes ago, ted_b said:

Yes, thanks, my Holo LOVES PCM as PCM and DSD as DSD.  And my computer muscles love it, too, as there is less strain when doing that kind of upsampling. 

 

Although I also do a lot of multichannel, my "don't convert" rules still apply, and upsampling is more like 2x at most.  But I worry that there are no real good multichannel benchmarks or experiences to know the cpu needs, especially for me cuz my system is not really mch functional right now, so I can't even say what my i7-6700k measures at.  And those measurements would be with HQP 3 and therefore no EC work.  I was planning on waiting to buy HQP 4 when my room is done.  :(  Thinking out loud, maybe I should move that up and and see what happens.  My i7 server and Holo  is in my RAAL headphone system right now.  I assume multichannel upsampling loads on OS (Windows 10 right now) would be legitimate to capture, even if I don't have my multichannel exaSound unboxed to connect to it?  Would HQP run if it doesn't see a 6 channel dac?

 

Thx

 

Take the plunge. HQP3 and HQP4 are different apps anyway so you can A/B compare them at your leisure.

[k6davis]

18 hours ago, Miska said:

Yes, it arrived. At least for me, with native version, upsampling with EC modulators to DSD256 doesn't work (from RedBook, with poly-sinc-ext2). Output to a NAA. With regular modulators it does work.

 

Fan is running, but it is still fairly quiet, so could be used in a listening space.

 

So the performance is roughly what I expected, similar to equivalent U-series Intel CPUs. Over time Apple will likely come up with bigger and more powerful versions to match the bigger Intel CPUs in performance.

 

Hmmm... I was hoping for better results. 

 

Quoting from Tom's Hardware: 

 

Quote

 

When the M1 does get to run natively, though, it seems to pack some serious power. Engadget reports that the M1 MacBook Air had Geekbench 5 results of 1,619/6,292. That’s well above their results for the 2020 i7 MacBook Air, which were 1,130/3,053. Meanwhile, the Tiger Lake Dell XPS 13 9310 scored in 1,496/5,254 on our own Geekbench 5.0 benchmarks, while the ThinkPad X1 Carbon Extreme Gen 3 with an Intel Core i7-10850H chip scored 1,221/6,116.

 

The M1’s single-core score also beats the 27-inch 2020 Core i9 iMac’s single-core score, which only hit 1,246. It loses out to the iMac’s 9,046 multi-core score, but that officially gives the M1 higher single-core test results out of any Intel Macs, even desktops.

 

 

I don't know much about how all of this works, but I had hoped that these strong single core performance results may be good news for us. Also, away from the test bench, in typical Apple use cases like video editing, the M1 is reportedly outperforming Intel based Macs, as these metrics would suggest.