To Equalize or Not to Equalize: That Is the Question

[audioguy]

Right now, I'm only using the A/D converter section of the TacT. It converts the analog output of my Oppo BDP-83SE (playing SACDs and HDCDs) to 24/96 digital words and I run the digital output of the TacT into my PS Audio Perfect Wave DAC for D/A conversion and balanced analog output to my Bryston 7B-SST amps and JL f112 subs.

I'm confused (that happens a lot). How do you take advantage of the TacT crossover abilities and drive the sub and mains separately if you take your output to a single external PS Audio DAC that drives the subs and mains. Or (more likely) am I misreading what you wrote? Do you not use the analog sub outs of the TacT?

 

[RUR]

...(playing SACDs and HDCDs)...

Now I understand why you're using analog. I can rip 2ch of the SACD to .wav and play them on my PWT, but there's no way I can preserve the last bit when I rip HDCD's and I absolutely believe I'm losing SQ as a result. :-(

 

[tmallin]

Now I understand why you're using analog. I can rip 2ch of the SACD to .wav and play them on my PWT, but there's no way I can preserve the last bit when I rip HDCD's and I absolutely believe I'm losing SQ as a result. :-(

And you'd be right.

 

[tmallin]

I'm confused (that happens a lot). How do you take advantage of the TacT crossover abilities and drive the sub and mains separately if you take your output to a single external PS Audio DAC that drives the subs and mains. Or (more likely) am I misreading what you wrote? Do you not use the analog sub outs of the TacT?

You are not confused. Right now I am not taking advantage of any of the TacT's EQ or crossover functions. Remember, the title of my post was, "To Equalize or Not to Equalize: That Is the Question." Right now, I am choosing not to equalize. As I mentioned in my point #10, the sound I'm getting from the PS Perfect Wave equipment right now (and for many months now) is so very fine I haven't felt the need to (possibly) muck it up by running the PS Audio equipment's output back through the TacT's A/D, EQ, crossover, and D/A functions. I may--and probably will--eventually get around to trying it, but not just now. I try not to make system changes until I'm dissatisfied. I get dissatisfied often enough.

 

[JackD201]

Hi Tom,

As a studio brat, I agree with you totally. Placement really is key. Just like a speaker, tonal balance can be manipulated by placement and should be the first option. What I was pointing out is that in today's world where everything is charged by the hour, it is highly unlikely that the 18 to 24 mics for a typical band be fastidiously placed. Typically they are placed in "good enough" fashion and priority is given to level. Rule of thumb is no eq or anything else while recording but come mix down...............

I personally never liked graphic eqs. The high distortion is in great part a function of contact points of the individual faders later summed at the buss. The higher the number of bands, the higher the distortion from their contacts. Sonically they are lumpy because each has a narrow Q making transitions from band to band quite unpredictable the greater the difference in dB cut or boost. These are the reasons they never gained traction in studios albeit they did gain traction in PA applications especially in fixed installations. Parametric EQs which are the recording and mixing standards the other hand have far fewer contact points and avoid the lumpiness. Phase shift is present in both but is low enough when static. It is only really detectable when the EQ is being ridden manually or by automation during playback.

This is why I theorize EQ got a bad rap. The mass market was exposed to graphics eqs. Even the best ones likes those from Klark Teknik were prone to the problems above. What came in the racks....oh boy. I think a whole lot of people discovered like I did that these domestic systems actually sounded better with the EQs off or better yet totally bypassed. The other side of it is being subjected to systems whose users abused EQ. The notorious flying V setting for max boom and sizzle was and continues to be a problem.

This brings me to another point. Analog EQ or DRC can not be inherently bad if the results say otherwise. There is however a potential for abuse. I do not wish to judge anybody. I believe everybody has a right to determine when he's had enough to drink. We all should know however when that drink is one too many.

There seems to be a black hole sucking up information about Boz and Peter's break up or even just who really did what when they were working together. I've been led to believe that Peter did the bulk of the design work for the original TacT units and that the RP system which came later either built on that or took it in another direction altogether. I got my RP for Global mode. My system and room really does not need any DRC. It is a ground up build. The room I have is rather large however and is meant for entertaining up to more than a dozen people. This puts the couches flush against the rear corners. Global mode allows for near uniform bass response even in these nodes. It comes at the expense of imaging but when the music is for mood setting and not active listening it is a good enough trade off. I'd rather have no imaging than a guest go nauseous in the corners

Jack

 

[tmallin]

Hi Tom,

As a studio brat, I agree with you totally. Placement really is key. Just like a speaker, tonal balance can be manipulated by placement and should be the first option. What I was pointing out is that in today's world where everything is charged by the hour, it is highly unlikely that the 18 to 24 mics for a typical band be fastidiously placed. Typically they are placed in "good enough" fashion and priority is given to level. Rule of thumb is no eq or anything else while recording but come mix down...............

I personally never liked graphic eqs. The high distortion is in great part a function of contact points of the individual faders later summed at the buss. The higher the number of bands, the higher the distortion from their contacts. Sonically they are lumpy because each has a narrow Q making transitions from band to band quite unpredictable the greater the difference in dB cut or boost. These are the reasons they never gained traction in studios albeit they did gain traction in PA applications especially in fixed installations. Parametric EQs which are the recording and mixing standards the other hand have far fewer contact points and avoid the lumpiness. Phase shift is present in both but is low enough when static. It is only really detectable when the EQ is being ridden manually or by automation during playback.

This is why I theorize EQ got a bad rap. The mass market was exposed to graphics eqs. Even the best ones likes those from Klark Teknik were prone to the problems above. What came in the racks....oh boy. I think a whole lot of people discovered like I did that these domestic systems actually sounded better with the EQs off or better yet totally bypassed. The other side of it is being subjected to systems whose users abused EQ. The notorious flying V setting for max boom and sizzle was and continues to be a problem.

This brings me to another point. Analog EQ or DRC can not be inherently bad if the results say otherwise. There is however a potential for abuse. I do not wish to judge anybody. I believe everybody has a right to determine when he's had enough to drink. We all should know however when that drink is one too many.

There seems to be a black hole sucking up information about Boz and Peter's break up or even just who really did what when they were working together. I've been led to believe that Peter did the bulk of the design work for the original TacT units and that the RP system which came later either built on that or took it in another direction altogether. I got my RP for Global mode. My system and room really does not need any DRC. It is a ground up build. The room I have is rather large however and is meant for entertaining up to more than a dozen people. This puts the couches flush against the rear corners. Global mode allows for near uniform bass response even in these nodes. It comes at the expense of imaging but when the music is for mood setting and not active listening it is a good enough trade off. I'd rather have no imaging than a guest go nauseous in the corners

Jack

Hi, Jack. I just looked up the specs on the last two graphic equalizers I owned. The Rane DEQ-60L (digital) is rated at .02% THD + noise. The Audient ASP231 (analog) is rated at .005% distortion at +4 dBu full bandwidth with noise at -92dBu full bandwidth. Those are a far cry from that old Advent I mentioned.

 

[JackD201]

You think the drop might be due to better discreet parts Tom?

 

[tmallin]

You think the drop might be due to better discreet parts Tom?

I don't know, Jack. I don't think ordinary analog bass and treble controls, introduce anything like the 3% or so harmonic distortion the test reports showed for the old Advent EQ unit.

 

[audioguy]

When is flat not exactly flat

I currently use an Audyssey Sub EQ to calibrate my four subs for HT use, It does a GREAT sub of providing as smooth FR (and according to it, "flat") and more importantly, greatly reduced bass ringing.

Below is the before and after curve of my subs with the Audyssey SubEQ remembering that their plots are "close approximations" since they are derived from, in my case, 9 measurement points.

I then decided to run the TacT through the corrected subs. As you can see, while the FR is pretty "straight", flat is not a word that comes to mind. If you think about most of the target curves the TacT provides, they usually have a rise in the lower end which would compensate for what we see here.

I also plan to run REW through the subs as well as the SigTech and some other software to see what it looks like.

So there you have it: One mans "flat" is another's "not exactly"

 

[tmallin]

Audioguy, I don't think you are properly interpreting either graph. The Audyssey graph is undoubtedly the target of its correction, not the measured result.

The TacT curve, if it is in fact a measurement of your left sub, will be incorrect in terms of showing the frequency response because it is clear from the picture that your crossover is enabled and that whatever correction filters you have loaded into preset #1 are engaged. If this is a measurement of the response, it is showing the response as altered by the TacT's EQ and crossover functions. Another strangeness is that despite the crossiver filter being engaged at 60 - 65 Hz, the response shown is rising above that frequency even though you are apparently only measuring the left sub. Something is amiss.

 

[audioguy]

Audioguy, I don't think you are properly interpreting either graph. The Audyssey graph is undoubtedly the target of its correction, not the measured result.

The TacT curve, if it is in fact a measurement of your left sub, will be incorrect in terms of showing the frequency response because it is clear from the picture that your crossover is enabled and that whatever correction filters you have loaded into preset #1 are engaged. If this is a measurement of the response, it is showing the response as altered by the TacT's EQ and crossover functions. Another strangeness is that despite the crossiver filter being engaged at 60 - 65 Hz, the response shown is rising above that frequency even though you are apparently only measuring the left sub. Something is amiss.

There may be something amis but not the chart. I ran TacT as a 2.1 so all four subs were connected to the left channel. I did NOT run the meaasurement through a preset even though it looks like it. I have attached another plot where you can see I ran it in bypass and no crossover was enabled. I measured the subs and moved the measurement into Memory 1.

And yes, Audyssey does have a "flat" target curve which s exactly the reason for my post. Their flat and TacT flat are not equal. The Audyssey measured response is the first plot, the corrected to their "flat" is the second.

 

[amirm]

From what I recall, TacT measurements are always "raw" and not through any of its processing steps. And that is my beef because it will not measure post correction changes. At least this is the case with my older multichannel system/software.

Audioguy, how consistent are the Audyssey measurements from pass to pass? Since it measures through its own filters it seems, what happens if you run it again? Do you get the same results?

 

[audioguy]

With the newer version, you can run the test through an existing filter and see exactly what it looks like. It also has a function called "compute" or something like that which provides the "theoritical" view of the corrected response, which (of course) is almost identical to the target. But when you tkae a measurement thru a filter, it is reallly does represent reality - which looks very disimilar to the theoritcal resposne.

The measurements are very consistent. I have a very quiet and dedicated room so I get virtually the same results if I take 5 measurement or 25. The exception is in the lowest bass where air conditioning or other LF noise can show.

In this case, I ran the output of the TacT through the Audyssey SubEQ so that it what was measured.

 

[tmallin]

From what I recall, TacT measurements are always "raw" and not through any of its processing steps. And that is my beef because it will not measure post correction changes. At least this is the case with my older multichannel system/software.

Audioguy, how consistent are the Audyssey measurements from pass to pass? Since it measures through its own filters it seems, what happens if you run it again? Do you get the same results?

My TacT 2.2XP AAA has always allowed measurements with any EQ curve active and the crossover enabled. That supposedly allows you to measure the results of what you are doing. In fact, you have to be careful that the crossover and preset EQ are not enabled when taking measurements--thus my comments to Audioguy.

The V1.0 software also has a response calculator function which doesn't measure anything and just shows the theoretical sum of the measured raw response of your sytem and the EQ and crossovers you apply. However, you can still run real measurements through the EQ and crossover as before.

I have found that when I measure the response of my system with non-TacT measurement software (either Liberty Audio's Praxis or SynRTA), a method which does not rely on the TacT to measure the results of the EQ and crossovers the TacT is applying and which even uses a different calibrated microphone, I do not measure nearly as flat a response as when the TacT is used to measure its own frequency response corrections.

 

[tmallin]

With the newer version, you can run the test through an existing filter and see exactly what it looks like. It also has a function called "compute" or something like that which provides the "theoritical" view of the corrected response, which (of course) is almost identical to the target. But when you tkae a measurement thru a filter, it is reallly does represent reality - which looks very disimilar to the theoritcal resposne.

The measurements are very consistent. I have a very quiet and dedicated room so I get virtually the same results if I take 5 measurement or 25. The exception is in the lowest bass where air conditioning or other LF noise can show.

In this case, I ran the output of the TacT through the Audyssey SubEQ so that it what was measured.

I always turn off the HVAC, dehumidifier, and other noise emitters before running TacT measurements. I also warn my wife so that she isn't walking around while the measurements are run. She is very cooperative in this effort since it takes about 20 minutes to run 70 pulses through left, right, left sub, and right sub. Operating the HVAC system and footsteps on the floor above the listening room during measurements can cause major changes in the measured response in the bottom two octaves.

Assuming the TacT graph is correct about what Audyssey is doing to the sound of your system, you can see one reason why I don't like this system. Not enough bass. And Audyssey doesn't allow sufficient adjustment to counteract this problem. The response drops fairly smoothly in level by 9 dB between 150 Hz and 10 Hz. You will see this problem better if you expand the vertical scale. If anything, to sound well balanced, the response in this area should be falling just a bit from 10 or 20 Hz up to 150 Hz--it should have the opposite type of slope, in other words.

 

[amirm]

Looks like I am stuck with my old version then! And being in mood to fork over more than the $16K I already did, I will have to live with it it seems.

 

[audioguy]

Assuming the TacT graph is correct about what Audyssey is doing to the sound of your system, you can see one reason why I don't like this system. Not enough bass. And Audyssey doesn't allow sufficient adjustment to counteract this problem. The response drops fairly smoothly in level by 9 dB between 150 Hz and 10 Hz. You will see this problem better if you expand the vertical scale. If anything, to sound well balanced, the response in this area should be falling just a bit from 10 or 20 Hz up to 150 Hz--it should have the opposite type of slope, in other words.

Agreed. After you run the SubEQ, you then run all of the rest of the channels through the Audyssey enable pre-pro. At this point you have teh ability to create a target curve (with Audyssey Pro) for the bass that is more realistic

 

[Steve Williams]

I am surprised Chuck that Audyssey betters the TacT in bass response

 

[audioguy]

I am surprised Chuck that Audyssey betters the TacT in bass response

I don't know that it does. What the Audyssey SubEQ does is allow me to have 4 subs with 2 at different distances, time align the subs and then treat them all as one. I am currently playing with the TacT and 4 subs but doing it a bit differently since, until you get to the TCS system of TacT, it does not have a way to deal with this many discrete subs -- but I have a way to try to fake it out.
I'm trying to figure out how to come up with the money for the TacT TCS MkIII and that will solve this problem once and for all.

 

[amirm]

Ah, I was wondering the same thing as Steve. I have the TCS and kept thinking in that context.