TOPIC:

I understand that now Dirac is generating filter files to be loaded into the SHD.
Is this a known/open format?

Reason for my questions is:
It seems the Dirac section in the SHD is always on and it makes the latency quite high (to high for some HT systems).
If I could load a 'bypass' FIR filter with only the first coeficient 1.0 and the rest zero, this would result in lower latency of the whole processing chain of the SHD.

Who knows if this possible?

I don't believe that's possible. The way I understood the explanation from devteam on this is, you have an inherent 25mS (minimum) latency in the SHD no matter how you program it.

But, you can certainly load your own FIR coefficients to the appropriate part of the software architecture. They just have to be formatted suitably.

Dave.

dreite wrote:

I don't believe that's possible. The way I understood the explanation from devteam on this is, you have an inherent 25mS (minimum) latency in the SHD no matter how you program it.

But, you can certainly load your own FIR coefficients to the appropriate part of the software architecture. They just have to be formatted suitably.

Dave.

The peak of the impulse response correction generated by the Dirac software is somewhere after the first sample, so by moving a 1.0 to the first location you gain that amount of time shift in the latency.

But its good news I can load my own Fir coefficients, any one has an example filter file that I could look at?

Look in the owners manual. Section 10.4.

Dave.

dreite wrote:

Look in the owners manual. Section 10.4.

Dave.

10.4CUSTOM BIQUAD PROGRAMMING

Thats what you mean?

I have read the whole manual before I posted this question.
And nowhere is the filter file format described....

mvs0 wrote:

dreite wrote:

Look in the owners manual. Section 10.4.

Dave.

10.4CUSTOM BIQUAD PROGRAMMING

Thats what you mean?

I have read the whole manual before I posted this question.
And nowhere is the filter file format described....

You must not be reading closely enough. There's an example file format in Section 10.4.2 and also a link to an excellent spreadsheet utility you could use to create them. There are numerous other 3rd party utilities that can create appropriate biquad coefficients. Also, REW (and some other testing programs) can spit out files correctly formatted.

There is also this in the miniDSP Applications section:
www.minidsp.com/applications/advanced-to...d-biquad-programming

Dave.

dreite wrote:

mvs0 wrote:

dreite wrote:

Look in the owners manual. Section 10.4.

Dave.

10.4CUSTOM BIQUAD PROGRAMMING

Thats what you mean?

I have read the whole manual before I posted this question.
And nowhere is the filter file format described....

You must not be reading closely enough. There's an example file format in Section 10.4.2 and also a link to an excellent spreadsheet utility you could use to create them. There are numerous other 3rd party utilities that can create appropriate biquad coefficients. Also, REW (and some other testing programs) can spit out files correctly formatted.

There is also this in the miniDSP Applications section:
www.minidsp.com/applications/advanced-to...d-biquad-programming

Dave.

Hi Dave, read my starting post more closely. I am not talking about biquad programming.
I am talking about the FIR filters that are loaded by the Dirac software.

You said you wanted to load a FIR "bypass" filter.....so as to reduce latency. (Which it won't BTW.)
If bypassing the FIR capability, then you'd have no other option but to use the biquad capability. Yes?

Am I going to fast for you?

Good golly.

Dave.

dreite wrote:

You said you wanted to load a FIR "bypass" filter.....so as to reduce latency. (Which it won't BTW.)
If bypassing the FIR capability, then you'd have no other option but to use the biquad capability. Yes?

Am I going to fast for you?

Good golly.

Dave.

That was rather patronising.

The OP wants to know hiw to enter their own FIR filter coefficients, which is not what you've described.

No, he said he wanted to load a "bypass FIR filter."
At this point, I'm not sure he knows what kind of filter he wants to implement.

Regardless, I'm out. Carry on.

Dave.

Ultrasonic wrote:

dreite wrote:

You said you wanted to load a FIR "bypass" filter.....so as to reduce latency. (Which it won't BTW.)
If bypassing the FIR capability, then you'd have no other option but to use the biquad capability. Yes?

Am I going to fast for you?

Good golly.

Dave.

That was rather patronising.

The OP wants to know hiw to enter their own FIR filter coefficients, which is not what you've described.

Perhaps... but sometimes our typing doesn't reflect what we really mean. Consider dreite wrote to Dave "read my starting post more closely" which could also be considered a bit rude considering Dave's trying to be helpful!

Ok! all misunderstandings seem to be solved now

And now we can go back to the original question:

Create my own FIR filter in Dirac format..
Can anyone send me a copy if his Dirac generated file (for SHD)?
Maybe I can look into it and find out the format.

mvs0 wrote:

Can anyone send me a copy if his Dirac generated file (for SHD)?
Maybe I can look into it and find out the format.

I don't know if the filter files are stored locally on the PC in a form that is accessible, and even if they were I wouldn't know how to send an edited version to an SHD. I've never seen anything to suggest that what you want to do is possible by the way but I am open to being proved wrong.

If you don't have an SHD your questions do rather beg the question of why you are interested in this? In that if you don't want to use Dirac it might be that a different product is more suited to you? Although having said that I did buy an SHD without really wanting Dirac...

Ultrasonic wrote:

mvs0 wrote:

Can anyone send me a copy if his Dirac generated file (for SHD)?
Maybe I can look into it and find out the format.

I don't know if the filter files are stored locally on the PC in a form that is accessible, and even if they were I wouldn't know how to send an edited version to an SHD. I've never seen anything to suggest that what you want to do is possible by the way but I am open to being proved wrong.

This is what I found in the manual:
"The default location for project files is MyDocuments\MiniDSP\Projects(Windows) or Documents/MiniDSP/Projects(Mac)."

Assuming you have a SHD, can you check whats in there?

Would be great if these files would be simple readable text files with FIR coefficients

mvs0 wrote:

his is what I found in the manual:
"The default location for project files is MyDocuments\MiniDSP\Projects(Windows) or Documents/MiniDSP/Projects(Mac)."

Assuming you have a SHD, can you check whats in there?

Would be great if these files would be simple readable text files with FIR coefficients

I think what is saved in the projects is the measurements rather than the filters but I'll have a look when I get chance later. Even if the filters were there I'm not sure how you'd be able to send one to an SHD independent of the Dirac software.

What was the answer to my question of why you are considering an SHD in the first place?