Comparing DiracART and GLM with one subwoofer

[Mort]

This is of some interest to a few of you so I'll make a separate thread and avoid cluttering up other threads. I'm surprised at how similar they are. I have done this test three times and came up with similar results. I'm pretty familiar with REW, dsp and measuring, but I'm no pro. I added the MDAT as well.

Setup: My main monitors are Genelec 8361a,which I use with a 15" 7380a subwoofer and GLM. For GLM I did a fresh calibration of autocal and autophase with a crossover at 85hz. For the DiracArt, I made a new group in GLM, assigned it to analog, switched the monitors to full range and removed any crossover. I then bypassed all calibrations before I ran DiracArt. I ran DiracArt straight except to limit the upper curtain to 300hz which is what GLM does.

 

[levimax]

This is of some interest to a few of you so I'll make a separate thread and avoid cluttering up other threads. I'm surprised at how similar they are. I have done this test three times and came up with similar results. I'm pretty familiar with REW, dsp and measuring, but I'm no pro.

Setup: My main monitors are Genelec 8361a,which I use with a 15" 7380a subwoofer and GLM. For GLM I did a fresh calibration of autocal and autophase with a crossover at 85hz. For the DiracArt, I made a new group in GLM, assigned it to analog, switched the monitors to full range and removed any crossover. I then bypassed all calibrations before I ran DiracArt. I ran DiracArt straight except to limit the upper curtain to 300hz which is what GLM does.

View attachment 508551
View attachment 508554
View attachment 508556

View attachment 508558View attachment 508559
View attachment 508560
View attachment 508561

The last 2 graphs are the same i.e. both GLM. These 2 graphs are the ones I am most interested in.

 

[Mort]

The last 2 graphs are the same i.e. both GLM. These 2 graphs are the ones I am most interested in.

Thanks for that. I updated.

 

[levimax]

Very interesting. Now much difference as you say. Can you hear any difference?

 

[Mort]

Very interesting. Now much difference as you say. Can you hear any difference?

Not really, no. It's hard to go back and forth instantly. I could probably hear that bass bump from Dirac if I listened closely, but it wasn't instantly obvious.

 

[Curvature]

Your room decay is already very low. Do you use a lot of treatment?

 

[Mort]

Your room decay is already very low. Do you use a lot of treatment?

I don't, no just a carpet. It's a small 3mx3m room with fair amount of glass. I'm 1.5m from the monitors, and 2m from the subwoofer.

 

[kawauso]

The frequency responses are similar, but the decay is completely different. Don’t they sound quite different to you? With ART being tight, even if you raise the low-frequency target curve, the bass doesn’t become boomy.

 

[Mort]

The frequency responses are similar, but the decay is completely different. Don’t they sound quite different to you? With ART being tight, even if you raise the low-frequency target curve, the bass doesn’t become boomy.

It certainly wasn't a pronounced difference, but I've only had a bit of time to listen. I'll focus on that next.

 

[kawauso]

It certainly wasn't a pronounced difference, but I've only had a bit of time to listen. I'll focus on that next.

Please try listening with that aspect in mind. It also appears that ART is reducing the standing-wave dips in the 100–200 Hz range. At the same time, this confirms that GLM has a very good default target curve and that it has the advantage of preserving headroom, with little to no reduction caused by boosting.

 

[AudioJester]

I think you allready have a pretty good starting point with GLM.

I agree the 100-200Hz region looks much better with ART.

I found the RCA outputs on my old X3600 do not have the grunt to drive my 8361a compared to direct digital AES/EBU input or xlr output from my Okto DAC.

 

[kawauso]

Also, in the ART thread, it appeared that ART was not very effective after GLM was applied as a pre-processing step. I had expected that the ART results would be nearly the same regardless of whether GLM was used or not. It was a slightly unexpected result.

 

[Mort]

Also, in the ART thread, it appeared that ART was not very effective after GLM was applied as a pre-processing step. I had expected that the ART results would be nearly the same regardless of whether GLM was used or not. It was a slightly unexpected result.

I would want to test that again before confirming. It's easy to miss a button or a step.

If I can, I also want to test with two subwoofers, which should show DiracART's strength relative to GLM.

 

[holbob]

Need measurements before art, and after art as well ideally. Preferably without any boosting as well so we can see how well art works.

 

[Mort]

Need measurements before art, and after art as well ideally. Preferably without any boosting as well so we can see how well art works.

The ART is after, unless I misunderstand you.

 

[Pekka Perämoottori]

Isn't this whole ART system based on providing an supporting acoustic stimulus to the assisted sound channel, from different locations of the room? The more stimulus channels and the more different locations of the room they are placed, the better ART can work.

Did I understand correctly that this system is only two-channel, with one subwoofer? So if that's the case, then ART can't do much, even in theory, with so few playback channels. Non-existent support channels will produce non-existent results.

CAPABLE (really capable and not some typical surplus crap) surround and ceiling speakers would therefore be a prerequisite for good ART results. A acoustic 4 + 4 support group (four in the ceiling, four in the basic surround position), for example. That's when ART would have a chance to start getting actual effect.

Of course, most people run out of money at this point and family living room installers are in big trouble anyway. But if this ART is to make a real difference and achieve those advertised things, that's how you could get them. At least in theory.

 

[Mort]

Need measurements before art

These are raw room and monitor/subwoofer lines behind the orange. No crossover.

 

[Mort]

Isn't this whole ART system based on providing an supporting acoustic stimulus to the assisted sound channel, from different locations of the room? The more stimulus channels and the more different locations of the room they are placed, the better ART can work.

Did I understand correctly that this system is only two-channel, with one subwoofer? So if that's the case, then ART can't do much, even in theory, with so few playback channels. Non-existent support channels will produce non-existent results.

CAPABLE (really capable and not some typical surplus crap) surround and ceiling speakers would therefore be a prerequisite for good ART results. A acoustic 4 + 4 support group (four in the ceiling, four in the basic surround position), for example. That's when ART would have a chance to start getting actual effect.

I think most readers of this thread are aware of that. It's not presented as a failure or success of ART, although indeed ART was slightly better. Definitely not a non-existent result.

 

[Mort]

Dirac and GLM - just 10-200. Dirac looks beautiful.

 

[AudioJester]

Did you measure GLM with dirac ART over the top?
Does beautiful looking result in an audible difference?

Nice one giving it a good trial!