TOPIC:

Hi Ultrasonic,
There's a way you can do the measurements with and without dirac/PEQ without the mic I.e. using the SHD as input and output device. Do you remember we came across this/discussed that method in another post... Just cant find it at the moment.

 

asx77 wrote:

Hi Ultrasonic,
There's a way you can do the measurements with and without dirac/PEQ without the mic I.e. using the SHD as input and output device. Do you remember we came across this/discussed that method in another post... Just cant find it at the moment.


 

Set the SHD as your input and output device in REW. First run a sweep at a volume level of something like -30 dB with no processing, the response should be completely flat. Then engage Dirac and other EQ you have applied and run the same sweep. By comparing the two you can see the level of net boost you have applied.

Michael

mdsimon2 wrote:

asx77 wrote:

Hi Ultrasonic,
There's a way you can do the measurements with and without dirac/PEQ without the mic I.e. using the SHD as input and output device. Do you remember we came across this/discussed that method in another post... Just cant find it at the moment.






 

Set the SHD as your input and output device in REW. First run a sweep at a volume level of something like -30 dB with no processing, the response should be completely flat. Then engage Dirac and other EQ you have applied and run the same sweep. By comparing the two you can see the level of net boost you have applied.

Michael

In theory I thiink you only need to do the sweep with Dirac and the other EQ engaged, since as you say the result with nothing selected should be completely flat. Frustratingly I couldn't get this to reliably work for me which is why I don't normally think of it. If someone can get it to work then it's definitely the best option though.

I think I got this to work a couple of times but then it stopped working and I couldn't figure out why.

@Ultrasonic
Hmm that's interesting - I didn't realise it was a bit hit and miss with you. When we all first discussed this I was using a ddrc24 and it worked really well. When I got an SHD I tried it again and I just couldn't get REW to see the SHD as BOTH input and output. Is there some thi g with the SHD do you think that stops this working?

 

I agree that you shouldn't need to run a baseline but I like to just to make sure I truly defeated all the processing.

I've been using this method for years with a 2X4HD and more recently with a SHD Studio. Really not much to it other than selecting the appropriate "input" channel in REW (which is actually an output channel in the SHD) for the "output" channel you are using in REW (which is actually an input channel in the SHD). REW also doesn't like it if your input / output sample rates are different so I do make sure that I set the sample rate in Audio MIDI Settings (I'm on a Mac) to 96 kHz and run the sweep at 96 kHz.

What sort of issues were you seeing?

Michael

@mdsimon2

Michael,

Did you see my post about a minute before yours? We must have been typi g at the same time. I was using the method with the ddrc24 no problem but have never got it to work with the SHD. Any chance you could post some screenshot of your settings?

Thanks, 
Alex

asx77 wrote:

@Ultrasonic
Hmm that's interesting - I didn't realise it was a bit hit and miss with you. When we all first discussed this I was usi g a ddrc24 and it worked really well. When u got an SHD I tried it again and I just couldn't get REW to see the SHD as BOTH input and output. Is there some thi g with the SHD do you think that stops this working?
 

It was my SHD where I struggled but I'll be honest I can't remember exactly how now. I might give it another try tomorrow. I've never tried with my 2x4 HD but as it doesn't has Dirac there isn't really any need.

Maybe the SHD Studio is different? Or maybe Mac is different from Windows? You could also use an analog / digital input from your computer instead of the USB input on the SHD.

This is what REW looks like for me.

 

Michael

much appreciated... Looks obvious enough for me to have been doing that correctly but I'll double check against your settings. 

 

@asx77 below is a link to the old thread that you asked about above, which may be of interest to anyone new the subject. It moves on quite a lot from page one, including @mdsimnon2 teaching me how to looke at phase response graphs better  .

www.minidsp.com/forum/dirac-series-suppo...ernal-headroom#48111

I've just given this another go and it's working for me right now  . As before, I am though always getting a sharp ('brick-wall' type) roll-off at 20 kHz, irrespective of sample rate*. For some reason I can't divide the with/without Driac lines to shift them to zero but I just applied a manual offset to do this. Here's an example of what I have currently, but this was a quick first attempt with some new speaker (and existing sub), in a sparsely furnished room following redecoration, and still lacking curtains.

 

*This was caused by a set Windows sample rate of 44.1 kHz (I think it was) on both input and output.

Interesting, that brick wall at 20 kHz strongly suggests that something is resampling upstream of the SHD, my measurements are flat out to 48 kHz (not that I think it matters for these purposes). I am not familiar with how windows treats audio but imagine there is some sort of system wide resampling going on.

Michael

mdsimon2 wrote:

Interesting, that brick wall at 20 kHz strongly suggests that something is resampling upstream of the SHD, my measurements are flat out to 48 kHz (not that I think it matters for these purposes). I am not familiar with how windows treats audio but imagine there is some sort of system wide resampling going on.

Michael

I'd wondered before it it might be something that REW does but if you're seeing differently then clearly not. Just to confirm that you are right to guess that I'm using Windows (10) by the way.

Just had a look and it was a Windows setting causing the sharp roll-off. Now sorted, although it makes no practical difference.

Just to clarify on what you might be seeing here when you are making a loopback measurement and seeing 48kHz sample rate instead of the processing rate of 96k. 
REW will select the WDM driver under Java. That driver is the windows driver. Not the ASIO driver. When you do so, it's basically controlled by Windows on the sample rate so make sure that you select the sample rate within windows control panel. 

If you use ASIO, then you can easily use REW to control the sample rate.