Regardless of who has the superior algorithm, heavy lifting tasks can be performed much more efficiently and with much higher precision on chips and FPGAs. This is a well known fact....

[completeluxury]

in acourate when you make convolutions it makes one for each sample rate.. 44.1, 48, etc up to 192 max.

 

at the moment for me those convolutions then get loaded into acourate convolver and then it will use the convolution for the sample rate it detects

[Blizzard]

in acourate when you make convolutions it makes one for each sample rate.. 44.1, 48, etc up to 192 max.

 

at the moment for me those convolutions then get loaded into acourate convolver and then it will use the convolution for the sample rate it detects

 

By convolutions do you mean IR.wav files for each channel?

 

I thought earlier you could import the VST convolution plugins like you can on Jriver but I was wrong.

[completeluxury]

honestly im probably just using it wrong who knows. i havent had a chance to actually listen to my system much since i set it up!

 

and i dont mean the ir.wav files.. it makes a different format. .csv or something i dont have access to the control pc to check what the actual format is

[firedog]

If you have a computer that will run Miska's stuff at 1% CPU, I'd like to know what's in it.

 

I have a CAPS Pipeline and online conversions to DSD 128 use something like 12-15% CPU; that's a pretty powerful computer, I don't think in a home audio environment you are going to get anything much more powerful. I think you'd have to go to a real commercial type of server with multiple CPU's, and I don't think you will get conversion down to 1% CPU.

I assume Miska has an idea about what is possible in this area, and could chime in.

[Blizzard]

honestly im probably just using it wrong who knows. i havent had a chance to actually listen to my system much since i set it up!

 

and i dont mean the ir.wav files.. it makes a different format. .csv or something i dont have access to the control pc to check what the actual format is

 

On the HQplayer webpage it says Acourate can generate the IR.wav files needed. So you'll just have to figure out how to generate the files in Acourate for each channel and import them into the HQplayer convolution engine. From there looks like HQ player will apply the DSP to any format.

[Blizzard]

I have a CAPS Pipeline and online conversions to DSD 128 use something like 12-15% CPU; that's a pretty powerful computer, I don't think in a home audio environment you are going to get anything much more powerful. I think you'd have to go to a real commercial type of server with multiple CPU's, and I don't think you will get conversion down to 1% CPU.

I assume Miska has an idea about what is possible in this area, and could chime in.

 

Yeah but going from 12-15% to 20-40% could make a big difference in sound. If I want to pull off 20 channels of conversion to DSD 512, and apply crossover DSP/room correction, it's not gonna happen running the Algorithms/DSP in a Windows environment. Will probably also have to back it down to DSD 256 for 20 channel mode. 512 can be for 8 channel (stereo).

[Blizzard]

I think I'm going to need a dedicated powerful Intel quad core CPU for each 4 channel DSD 512 endpoint. I can run 1 channel per core. Then I can probably pull it off no problem.

 

Not 1% CPU load but can probably get it down pretty low using a realtime operating system designed to efficiently process DSP.

[Blizzard]

I just want to say after further listening to HQplayer, even with my compromised low powered setup, the sound I'm experiencing is leaps and bounds better than any other media player software I've experienced. And that's on both Windows and OSX.

 

The offline converted files played back on the Mirus SD card are leaps and bounds better than the NUC/NAA combo. But this is completely a hardware issue, not a software issue.

 

Jussi's contribution to audio with his state of the art algorithm's will go into the high end audio history books as a milestone in the reproduction of digital audio.

[dapla]

<edit>

Jussi's contribution to audio with his state of the art algorithm's will go into the high end audio history books as a milestone in the reproduction of digital audio.

 

 

As a relatively (and appropriately) "quiet" reader of this site, who can still wear several pertinent hats in these matters, this is certainly my take.

 

I am actually going to unplanned lengths (inc. $$), hoping to be able to properly utilize Jussi's efforts. And I have finally heard a system other than my own where I was satisfactorily (for me) able to compare appropriate playback software.

[Blizzard]

As a relatively (and appropriately) "quiet" reader of this site, who can still wear several pertinent hats in these matters, this is certainly my take.

 

I am actually going to unplanned lengths (inc. $$), hoping to be able to properly utilize Jussi's efforts. And I have finally heard a system other than my own where I was satisfactorily (for me) able to compare appropriate playback software.

 

"Quiet" is probably a wise choice on this forum. But at least if somebody wants to argue about any matters pertaining to the issues discussed in this thread, I can send them a link to this thread now. I think many misconceptions were set straight.

 

For me anyways.