Digital correction: Why is there resistance?

[rhbblb1]

Suggest you take a look at my system's link and my response graph, w/o DSP.

ACK, I have reviewed your frequency response graph and am convinced that your system would benefit from digital room correction. You have a significant dip between 60 to 100 Hz and a significant bump for 40 to 50 Hz. These are over long and significant areas. In addition, you have reversed the phase of your sub to reduce a bass bump and to extend the lower frequency response. I have done the same thing. It makes your frequency response look better, but it strains the dynamics. If you played your sub in phase with your mains and then used digital room correction, you would be shocked at the improvement. Your system would sound much more dynamic. In addition, digital room correction seems to have even better improvements with dipoles. Did wonders years ago for my Genesis 300's. By the way, digital room correction DOES work in the time domain. That is why it is superior to simple equalization. I have been using digital room correction for almost 20 years. Remember the SigTech. I now use Dirac which is integrated with my music server. I have a dedicated room which is fully acoustically treated. I have no deep nulls or very tall bumps. I do have some dips and small bumps. My room was measured by the chief acoustician from Hyundai. (They wanted ideas to design a room in Korea to study acoustics for their autos). Concerning the frequency response, he stated ( with a thick Korean accent) "almost perfect". Despite that, I am able to improve the sound with digital room correction. Dirac is only a couple of hundred bucks and they have a free trial. I agree that the colorations imparted by the speaker room interaction is the main problem that audiophiles face. I believe that digital room correction actually will get you closer to the original recording, not take you further away.

 

[microstrip]

http://www.computeraudiophile.com/c...-loudspeaker-correction-software-walkthrough/

Here is a lengthy, detailed, and involved article on using one DSP program. In this case Accurate. Down into the article you can see measured unsmoothed room response vs corrected response. It is a big difference. (...)

Esldude,

Do you have a link to a similar page about DIRAC?

 

[jap]

Esldude,

Do you have a link to a similar page about DIRAC?

Try this:

http://www.ultrahighendreview.com/dirac-live-room-correction-software-review/

 

[ack]

Thanks, perhaps you are right, but the ill effects of DSP in the rest of the range, to me, do not justify a fix in a very narrow band in the bass. As far as dynamic suppression goes, I am not hearing that - easy to prove, switch the sub in and out. In fact, it's quite the opposite - with the sub on, when I play the opening of Britten's Symhonia Da Requiem (http://referencerecordings.com/RR_120_DETAIL.asp) I literally jump up - so powerful, taught and clear is that drum hit - and so has every one of my unsuspecting guests. And neither has anyone ever "heard" the sub. Finally, the size and power of the o-Daiko drum (http://www.amazon.com/Heartbeat-Drummers-Audiophile-Reference-Series/dp/B0000009FE) has everyone running [figuratively] scared. Not to mention Mahler climaxes, etc. What's really missing is multiple subs.

 

[dallasjustice]

Maybe Hyundai can put DIRAC into the Equus. I'm in the market for a new car and I'm pretty sure it's going to be a Lexus with all options except the Mark Levinson system. I don't normally care about car audio but I bet DIRAC would sound pretty awesome in the Bentley system. Maybe I could afford the Bentley if I took up manufacturing audiophile power cables or something.

ACK, I have reviewed your frequency response graph and am convinced that your system would benefit from digital room correction. You have a significant dip between 60 to 100 Hz and a significant bump for 40 to 50 Hz. These are over long and significant areas. In addition, you have reversed the phase of your sub to reduce a bass bump and to extend the lower frequency response. I have done the same thing. It makes your frequency response look better, but it strains the dynamics. If you played your sub in phase with your mains and then used digital room correction, you would be shocked at the improvement. Your system would sound much more dynamic. In addition, digital room correction seems to have even better improvements with dipoles. Did wonders years ago for my Genesis 300's. By the way, digital room correction DOES work in the time domain. That is why it is superior to simple equalization. I have been using digital room correction for almost 20 years. Remember the SigTech. I now use Dirac which is integrated with my music server. I have a dedicated room which is fully acoustically treated. I have no deep nulls or very tall bumps. I do have some dips and small bumps. My room was measured by the chief acoustician from Hyundai. (They wanted ideas to design a room in Korea to study acoustics for their autos). Concerning the frequency response, he stated ( with a thick Korean accent) "almost perfect". Despite that, I am able to improve the sound with digital room correction. Dirac is only a couple of hundred bucks and they have a free trial. I agree that the colorations imparted by the speaker room interaction is the main problem that audiophiles face. I believe that digital room correction actually will get you closer to the original recording, not take you further away.

 

[rhbblb1]

Thanks, perhaps you are right, but the ill effects of DSP in the rest of the range, to me, do not justify a fix in a very narrow band in the bass. As far as dynamic suppression goes, I am not hearing that - easy to prove, switch the sub in and out. In fact, it's quite the opposite - with the sub on, when I play the opening of Britten's Symhonia Da Requiem (http://referencerecordings.com/RR_120_DETAIL.asp) I literally jump up - so powerful, taught and clear is that drum hit - and so has every one of my unsuspecting guests. And neither has anyone ever "heard" the sub. Finally, the size and power of the o-Daiko drum (http://www.amazon.com/Heartbeat-Drummers-Audiophile-Reference-Series/dp/B0000009FE) has everyone running [figuratively] scared. Not to mention Mahler climaxes, etc. What's really missing is multiple subs.

ACK,
Forty to 100 Hz is not a narrow band.
Of course the bass is more potent on than off. What I am saying is that "on" in phase, but room corrected, will likely be even better. Remember that the upper frequency will be effect less by room correction.
I forgot to mention in my last post that the biggest improvement in your system would be making the left and right channels equivalent. Your room is not symmetrical. Therefore, the frequency responses of each speaker will not be identical. This will effect imaging and will also cause tonal aberrations. Digital room correction will have major improvements with these problems. These are areas of improvement that is often left out of room correction discussions.

 

[esldude]

http://www.computeraudiophile.com/c...e-alignment-driver-linearization-walkthrough/

Here is another article where Mitcho does crossover and time alignment of his speakers using the DSP software

 

[audioguy]

Maybe Hyundai can put DIRAC into the Equus. I'm in the market for a new car and I'm pretty sure it's going to be a Lexus with all options except the Mark Levinson system. I don't normally care about car audio but I bet DIRAC would sound pretty awesome in the Bentley system. Maybe I could afford the Bentley if I took up manufacturing audiophile power cables or something.

I just purchased a Lexus RX350 (no ML audio system). The car is awesome and the sound system is good enough.

When I was with SigTech we tried (unsuccessfully) tried to get some audio companies to do room/car correction in the vehicles

 

[esldude]

Thanks, perhaps you are right, but the ill effects of DSP in the rest of the range, to me, do not justify a fix in a very narrow band in the bass. As far as dynamic suppression goes, I am not hearing that - easy to prove, switch the sub in and out. In fact, it's quite the opposite - with the sub on, when I play the opening of Britten's Symhonia Da Requiem (http://referencerecordings.com/RR_120_DETAIL.asp) I literally jump up - so powerful, taught and clear is that drum hit - and so has every one of my unsuspecting guests. And neither has anyone ever "heard" the sub. Finally, the size and power of the o-Daiko drum (http://www.amazon.com/Heartbeat-Drummers-Audiophile-Reference-Series/dp/B0000009FE) has everyone running [figuratively] scared. Not to mention Mahler climaxes, etc. What's really missing is multiple subs.

So exactly what are the ill effects in the rest of the range you refer to?

 

[dallasjustice]

I'm leaning toward the new es 350 but the rx 350 would make more sense for me. I respect you for driving a momma car. . At least the ladies are checking your ride out.

I just purchased a Lexus RX350 (no ML audio system). The car is awesome and the sound system is good enough.

When I was with SigTech we tried (unsuccessfully) tried to get some audio companies to do room/car correction in the vehicles

 

[zydeco]

Wow – it’s interesting that this topic has elicited such an enthusiastic response. I don’t have the room, budget or equipment of others but have tinkered with DSP over the past 5-years (see signature). My take is that digital room correction is just another tool in improving room response. It should be used in collaboration with, not at the exclusion of, optimisation of room treatment as well as listener / speaker position. And, of course, getting good results is both time consuming and demanding. The idea that we shouldn’t consider digital room correction on the basis that some use it to avoid the hard yards of room-speaker optimisation or because automated settings are sub-optimal seems inconsistent with the “what’s best” notion of this forum. We don’t, for example, dismiss room treatment just because some folks over-treat their rooms creating a dead environment. My guess is that in the future the best digital playback set-ups will use digital room correction that compliments room treatment and room-speaker optimisation with this correction defined by trained acousticians with the right measurement tools and techniques.


Zydeco

P.S. It also worth noting that whilst most of this discussion has centred on DSP as digital room EQ there are over important possibilities including speaker / driver correction and digital cross-overs.

 

[jsn]

From my past experience using behringer DCX2496 , the DRC was carried out in the way : room measurement firstly then digital EQ it.
The sound quality was reduced significantly, particular, less transparent, less coherent.
.
Then I had bias to all DRC.
.
However, when I have had to face the problem of using a big loudspeakers in a small room, I needed to handle the potential over boom issue and started to investigate the DRC again and found that the new way to apply the DRC via FIR filter using convolution engine in player software is really different from the old way. Bascially, it also needs to carry out the room measurment. But the following steps are different. It involves setting up the target room response curve and parameters , generating the filter, reviewing the resulting curve, readjusting the parameters and applying the filter to the convolution engine to the player (JRiver)
It involves very complicated calculation but we have very powerful PC nowaday so we can work out this kind of DRC.
.
I've tested the Acourate and audiolense and I could not find any negative effect for the sound and,on the other hand, it let me hear more details.
.
Is there disadvantage ? Yes.
The way to use the FIR filter with convolution engine needs long learning curve in comparing with the hardware EQ.
Without the help of support, it would not get a fruitful result. Furthemore, it would not work for DSD file in JRiver.
.
Even I love the result of using the DRC , it would only be last step to go in tuning process. The analogy is like making a bowl of salad. The component matching, speaker placement , room acoustic treatment, power conditioning , cable selection,.. are the ingredient of the salad. You have your most fresh , crispy, ingredient in the right proportion and seasoning in the bowl firstly. The latest DRC is like the salad sauce to bring out the utmost flavor of your precious ingredient at the last step.

I think the resistance coming from the previous bias, not understanding the technology, bad feeling in using the software without good support and negative experience for some overwhelming tuned system.
.
Cheers, jsn.

 

[Don Hills]

... Anytime you change a digital file, it can not be any better than the original.

If I recall your profession correctly, you make digital files better for a living.

 

[JackD201]

DRC to me is the equivalent of a straight razor. Put it in a 10 year old's hands and you can imagine what can happen. Give it to a proficient person however and you can get one heck of a great shave.

 

[audioguy]

I'm leaning toward the new es 350 but the rx 350 would make more sense for me. I respect you for driving a momma car. . At least the ladies are checking your ride out.

Given I had a Chevy Tahoe, I could not drop down to a sedan.

Sorry for OT posts

 

[mojave]

Nulls can not be fixed but dips can.

You bring up a good point that is often overlooked. Several others have already mention in this thread that you can't fix a null as if that statement eliminates any DSP. The problem is that, in real life measurements, very few have true nulls. If you take a measurement at 70 dB and another at 85 dB and the frequency response looks the same, then you have increased the volume over the entire frequency spectrum and don't have a null. For there to be a null, the level has to stay the same even when you increase the volume.

If taking two sweeps at different volumes indicate that you can increase levels over the frequency spectrum, then it also mean that you can boost a dip to smooth the bass response.

The idea that boosting also uses a lot more amp power is also incorrect. It does use more amp power, but only at the frequency and bandwidth (Q) that you are boosting. In relationship to the overall power being used across the frequency spectrum, that increase is actually very small. Furthermore, any cuts you do are reducing power requirements at other frequencies. With a 4 dB boost at 80 hz and a 4 dB cut at 40 Hz, you actually end up using less power since lower frequencies require more power output.

 

[dallasjustice]

I agree with Mojave. The key to excellent DSP application is knowing what to correct and what not to correct. As far as nulls are concerned, my experience with them is that one can usually move the seated position just a few inches to get out of any true front/back wall null. Of course, some folks experience sidewall or other wall nulls. Those are dependent on individual situation. The good news is that there's some really great software available now that can help sort this stuff out such that one doesn't need to know the mathematical differences between minimum phase and non-minimum phase to generate appropriate filters. Also, its important to understand what the magnitude response really means. Just because you have what looks like a null from a single measurement, doesn't mean you will "hear" the null in the same way the measurement is represented on the graph. That's where Jack's point is so right. One needs to know when to hold em' and when to fold em'. If you do, you can make a big improvement without ANY downside. That's the goal and it's totally achievable for anyone that is genuinely interested in this topic.

Michael.

 

[microstrip]

From the Mike Rivers report on the Audio Engineering Society 135th Convention, New York, NY, October 17-20 2013

The panel on Digital Room Correction (and does it really work?) was interesting in the several different methods and degrees of processing used by the systems
presented by the four panelists. The conclusion is that it can work, but that the better the room you start out with, the better it will work. There’s a whole lot of
measurements to be made and compensation applied, and it’s only become practical since inexpensive computers have become so powerful that extensive
digital signal processing can be implemented at fairly low cost, though like any other software-based system, it’s the process and software development that’s
still costly.

I would conclude that results depend on the quality of the room and most of all on process of taking measurements of the room to tailor the DRC.
Should I also conclude that soon we will be debating if blind tests are needed to evaluate the different DRC products available in the market?
BTW, I have now downloaded the DIRAC evaluation version.

 

[JackD201]

I agree with Mojave. The key to excellent DSP application is knowing what to correct and what not to correct. As far as nulls are concerned, my experience with them is that one can usually move the seated position just a few inches to get out of any true front/back wall null. Of course, some folks experience sidewall or other wall nulls. Those are dependent on individual situation. The good news is that there's some really great software available now that can help sort this stuff out such that one doesn't need to know the mathematical differences between minimum phase and non-minimum phase to generate appropriate filters. Also, its important to understand what the magnitude response really means. Just because you have what looks like a null from a single measurement, doesn't mean you will "hear" the null in the same way the measurement is represented on the graph. That's where Jack's point is so right. One needs to know when to hold em' and when to fold em'. If you do, you can make a big improvement without ANY downside. That's the goal and it's totally achievable for anyone that is genuinely interested in this topic.

Michael.

Sadly as far as DRC application goes, I'm like 10 3/4s. In one show however when we drew a concrete box with carpet, we were fortunate enough to have a really good barber next door who saved us. My partner who has zero WAF flexibility relies on DRC with very, very good results as opposed to no DRC.

 

[dallasjustice]

Sadly as far as DRC application goes, I'm like 10 3/4s. In one show however when we drew a concrete box with carpet, we were fortunate enough to have a really good barber next door who saved us. My partner who has zero WAF flexibility relies on DRC with very, very good results as opposed to no DRC.

I think you are saying that one can only benefit from DSP if their room is acoustical defective in some serious way. If that's not what you are saying, I apologize. However, I think some of the more timid rhetoricians in this thread imply that's the case. Yet, most all of the anti-dps'ers have never actually tried DSP.

My experience already having an excellent room tells me otherwise. I know that no matter how awesome one thinks their room is, there's gonna be a coloration imparted that DSP can help to ameliorate, if it's done right.