What is Dirac doing?

[Hear Here]

If your suggestion was merely to limit room correction to the lower frequencies where the room dominates and leave excellent performing speakers alone on the higher frequencies where the speaker dominates you'd get more support.

Yes, that's exactly what I'm suggesting - with the upper frequency range sent directly to the speakers' upper drivers without DSP.

 

[Hear Here]

That makes sense, although if it's causing an unwanted dip in an out-of-band frequency range I would call that an issue that needs addressing. It should apply those corrections differently, or be able to correct the dip.

The version of Dirac Live we are discussing cannot correct any peaks or troughs above 500 Hz. There is a premium version that can adjust the entire frequency range, but not the one in the OP's or my amp.

 

[kyuu]

Yes, that's exactly what I'm suggesting - with the upper frequency range sent directly to the speakers' upper drivers without DSP.

Well that's akin to suggesting that the problem is inherent to using DSP. Also, woofers are often sent frequencies well over 500Hz, so your proposed solution wouldn't necessarily avoid the DSP having an effect anyway.

The version of Dirac Live we are discussing cannot correct any peaks or troughs above 500 Hz. There is a premium version that can adjust the entire frequency range, but not the one in the OP's or my amp.

I'm aware. What I'm suggesting is that if, in the course of doing whatever it does, it causes an unwanted change in frequencies outside of the desired band (in this case over 500Hz), then it needs to be able to correct that unwanted change regardless of the licensing or else avoid making that unwanted change.

 

[Hear Here]

Well that's akin to suggesting that the problem is inherent to using DSP.

That's what I believe - why should a "signal processor" not adversely affect the signal?

What I'm suggesting is that if, in the course of doing whatever it does, it causes an unwanted change in frequencies outside of the desired band (in this case over 500Hz), then it needs to be able to correct that unwanted change regardless of the licensing or else avoid making that unwanted change.

Yes, perhaps that's the case, but it's unable to "correct" anything over 500 Hz. Whichever way it's looked at, the sound (to our ears) loses a little of it's "sparkle" and that sparkle is from the upper frequencies that shouldn't be affected by the signal processor, Sadly, the signal cannot avoid the processor, so a slight deterioration cannot be ruled out. This slight loss in top end sparkle (perhaps not a technical description, but I'm sure you recognise it) may well not be audible with mediocre speakers.

 

[theREALdotnet]

My conclusion is that the entire frequency range has to endure this processor, even though DL can only adjust sub-500 Hz.

That is not my experience. I routinely create two sets of filters, one full-range and one with a curtain around my room's Schroeder frequency. They measure and sound very differently. With the full-range filters I get a more accurate frequency response all throughout. With the curtained ones I preserve the characteristic tone of my speakers better, warts and dips and jaggies and all. I have no clear preference for either, the best ones to use seem to depend on the material.

 

[existingbrain]

That is not my experience. I routinely create two sets of filters, one full-range and one with a curtain around my room's Schroeder frequency. They measure and sound very differently. With the full-range filters I get a more accurate frequency response all throughout. With the curtained ones I preserve the characteristic tone of my speakers better, warts and dips and jaggies and all. I have no clear preference for either, the best ones to use seem to depend on the material.

I agree that this should be the expected behavior, and is exactly how the functionality of the curtains is described in the Dirac Live manual. I understand that adjustments to impulse, phase, level, and whatever else might affect FR outside of the curtains, but one should not expect it to change as drastically as the graphs I posted, particularly the second one. The impact to FR is literally greater beyond the curtain compared to within the specified curtain range, and any kind of mechanism that causes greater unintended change relative to desired change is IMO broken.

I did leave out one other detail because I did not want to complicate the story further, but I previously had another NAD C658 unit where I ran Dirac, and when I measured in REW with Dirac on vs. off, I found that Dirac only altered the FR within the curtains, as to be expected. This was in the same room with the same setup, albeit the speaker location was moved about an inch closer together since then (another complicating detail). This is the reason why I am thinking my current unit might have some defect, though I can't fathom what kind of defect would cause such a specific abnormality. I am waiting to hear back from NAD and Dirac, but may try another factory reset and re-run Dirac in the mean time to rule out software issues.

 

[Miniyouuuu]

If Dirac implements a delay in one of the speakers in order to adjust some issue at lower Frequencies it is normal to modify the upper frequency range. Both speakers are not playing same frecuency at the exact moment. Each speaker works as expected but combination of both speakers with that delay has that minor issue.

 

[Geert]

Easily tested by measuring the response of L and R separately with REW.

 

[Paffi]

This tallies rather with my own ear-based results when running the sub 500 Hz version of DL in my NAD M33. The top end loses a little of its sparkle when a DL filter is engaged. I've mentioned this repeatedly on forums, but many claim this is impossible with a digital processor.

My conclusion is that the entire frequency range has to endure this processor, even though DL can only adjust sub-500 Hz.

Made the same experience some years ago - Altough setting the curtain low, it destroyed the phantom imaging and killed the dynamics, hence I stopped using Dirac years ago...

 

[Hear Here]

Made the same experience some years ago - Altough setting the curtain low, it destroyed the phantom imaging and killed the dynamics, hence I stopped using Dirac years ago...

Yes, I agree

If you buy great speakers, set them up well, feed them a nice flat signal and think about how you can avoid the worst of room anomalies, that's the much preferable solution than messing with the signal to compensate for laziness / mediocre kit!

 

[Geert]

If you buy great speakers, set them up well, feed them a nice flat signal and think about how you can avoid the worst of room anomalies, that's the much preferable solution than messing with the signal to compensate for laziness / mediocre kit!

You think people can fix 15dB peaks and dips in the low end of the frequency spectrum by moving speakers and listening position around in the few square inches of wiggle room most people have available for their HiFi setup?

 

[Paffi]

You think people can fix 15dB peaks and dips in the low end of the frequency spectrum by moving speakers and listening position around in the few square inches of wiggle room most people have available for their HiFi setup?

You don't have to fix 15db to a +/-1db tolerance in the low end. Making it half may be fine and that may be possible.

 

[Hear Here]

You think people can fix 15dB peaks and dips in the low end of the frequency spectrum by moving speakers and listening position around in the few square inches of wiggle room most people have available for their HiFi setup?

Without knowing more about the room, the equipment and its placement, I couldn't comment, but possible contributory factors may be poor choice of speaker type as well as make / model, and many other factors. You shouldn't blame the room's features alone for that sort of peak or trough - there must be more to 15 dB than that!

Remember that before DSP was even thought of, most true audio buffs managed to achieve excellent sound from their systems, even though they didn't spend their time with oscilloscopes measuring to see if there was a dip or peak in the frequency response. It is regrettable that it's now all too common to chuck DSP at just about every perceived problem (audially significant or not) and expect it to magically sort things out. Whether the peaks and dips are room, speaker, or even amp generated, the easy solution is DSP.

And how do those analogue fans manage when they wouldn't dream of converting their analogue signal to digital, then DSP it, before converting back to analogue? They manage I suspect by using other means to achieve great sound.

 

[Sal1950]

Even when Dirac is set to modify just a frequency range, Dirac applies volume and delay variations between channels. That affect all the frequency range. Thus you get variations out of the configured frequency range.

I think you nailed it here. AFAIK Dirac still makes timing and level corrections to the full spectrum.

 

[Geert]

You don't have to fix 15db to a +/-1db tolerance in the low end. Making it half may be fine and that may be possible.

Leaving peaks larger than 6dB in the low end is still unacceptable for me. They will colour the sound (resonate) very clearly once you're used to a clean sounding bass. And you probably won't be able to improve peaks with 6dB or more by moving your speakers a few inches. The wave lengths of low frequencies are much to long for that.

 

[Sal1950]

And how do those analogue fans manage when they wouldn't dream of converting their analogue signal to digital, then DSP it, before converting back to analogue? They manage I suspect by using other means to achieve great sound.

One persons opinion of "great" sound is often different than the next.
The science of Home Music Reproduction has come a long way in the last 75 or so years.
And improved vastly.

 

[Geert]

You shouldn't blame the room's features alone for that sort of peak or trough - there must be more to 15 dB than that!

Room modes in a concrete and bricks room can easily cause peaks larger than 10dB. Boundary effects can come on top of that.

Remember that before DSP was even thought of, most true audio buffs managed to achieve excellent sound from their systems, even though they didn't spend their time with oscilloscopes measuring to see if there was a dip or peak in the frequency response.

Do we have evidence of that? I remember listening to serious room issues for more than 30 years. Studio's have been using analog eq's for room correction for more than 40 years.

 

[popej]

The reason why this is weird is because I don't have the full range version of Dirac and am therefore limited to 500hz

I think limited license doesn't exist anymore. Maybe you got upgrade?

 

[Todd68]

Made the same experience some years ago - Altough setting the curtain low, it destroyed the phantom imaging and killed the dynamics, hence I stopped using Dirac years ago...

I owned the minidsp Flex with Dirac for a few months. I tried so many times and the different mic patterns till I was just plain burnt out with it. I never got better or more satisfying sound quality, so I gave up and sold it.

The only thing it helped in my case was objectively better bass. But the overall sound quality was not better just different. I didn’t experience any kind of improvement in imaging etc. like I read about. So now I only do minimal EQ to the subwoofers. The only thing I do is nock down the 37hz room mode I have. Which I did using my sub amps adjustable PEQ.

I more satisfied now and don’t miss the constant measuring, adjusting and listening for any improvement. I guess some people with better results are more skilled and able to tune their speakers, but for me it wasn’t worth it. Also, I mostly own efficient speakers compared to the norm and miniDSP units added audible noise.

 

[Sal1950]

I think limited license doesn't exist anymore. Maybe you got upgrade?

Not according to their website.

Denon AVR-X4800H / AVC-X4800H – Dirac

DIRAC LIVE FOR Denon AVR-X4800H / AVC-X4800H Get your license

www.dirac.com