RMAF 2016: DEQx, Redgum Amplifiers, VAF Speakers

[amirm]

DEQx folks always do a good job of doing proper AB tests of the speaker correction/room EQ they do and this show as no exception:

The three LEDs on right, controlled a clever AB test where they could test the speakers being driven passively versus active! First such demo I have seen. Active sounded a lot better with warmer sound with better bass. No telling however that this is all due to active versus passive, or just a different crossover.

 

[amirm]

Little Wing by Stevie Ray Vaughan & Double Trouble. http://shz.am/t345705

Not sure if this is from the same album:

Each equalization step improved the sound: from driver correction to active. Quite a remarkable overall difference.

 

[Keith_W]

No doubt the DEQX would have made a massive difference. But what is really interesting is that Philip Vafiadis (owner / designer of VAF) allowed his speaker to be demonstrated in its standard form (with the passive crossover) vs. active and corrected (i.e. passive XO bypassed) with the DEQX. As far as I know, VAF does not offer the I-93 (the model pictured) sans crossover, with an invitation for you to BYO active XO. So it does not make much marketing sense for him to allow this to happen.

I am a big fan of the VAF I-93, which is a superb speaker. But I sometimes think that it is a little too much speaker for some rooms. The cheaper model (I-66) is my preference.

 

[amirm]

I was surprised too. In past shows they would buy cheap speakers online. I asked if they had gotten the speaker crossover info and presentor thought they had. But they sounded so different that I don't think they had.

 

[Keith_W]

Amir, from what I know of the DEQX guys, they would have designed their own crossover using their own measurements. I mean, what is the point of simply replicating a passive crossover into digital, and taking along with it all the compromises the designer had to make? They would have used steep slopes, corrected for the drivers, and time aligned the speaker, then performed an overall room correction at the show. THAT is why they sound so different. DEQX is a great product if used intelligently.

 

[fas42]

They would have used steep slopes, corrected for the drivers, and time aligned the speaker, then performed an overall room correction at the show. THAT is why they sound so different. DEQX is a great product if used intelligently.

Just to emphasise a point, normal crossovers still cause drivers which shouldn't be handing a certain part of the spectrum to be fed that signal - it's just down quite a few dB in level, depending on how far from the crossover point you are. Steep, or brickwall filters mean that no part of the spectrum is fed to the driver except that which it's designed to reproduce - it's the ideal way to feed speakers; normal crossovers are a poor compromise, really ...

 

[Keith_W]

By the way, what did you think of the Redgums? I still snigger a bit at your less than flattering description of them on WBF a year back.

 

[amirm]

Hey, I didn't start that war! Another dude did and then Marc went crazy. But yes, I did have an hand in it.

As to your question, I was not focused on anything but DEQx. With all of their corrections the room sounded nice.

 

[Cosmik]

Amir, from what I know of the DEQX guys, they would have designed their own crossover using their own measurements. I mean, what is the point of simply replicating a passive crossover into digital, and taking along with it all the compromises the designer had to make? They would have used steep slopes, corrected for the drivers, and time aligned the speaker, then performed an overall room correction at the show. THAT is why they sound so different. DEQX is a great product if used intelligently.

You make it sound almost like an automatic process, as though speaker designers aren't the talented, artistic geniuses they make themselves out to be. The designer tells you they spent a year optimising the passive crossovers but some techno whizkids come along and do it better in a couple of hours. (Sadly, I suspect this is the truth).

One thing speaker designers like to talk about, though, is the smoothness of the off-axis response. A particular set of drivers will only give you an 'optimal' blend of their dispersion characteristics when you use specific crossover slopes and frequencies, so even given the unlimited possibilities of a DEQX, the options may be more limited than they appear. However, if a speaker is a three-way then it's hard to get it too wrong IMO.

 

[Keith_W]

It is not my intention to make it sound like an automatic process. Creating a passive crossover is not easy. The more components you include, the more insertion loss you will encounter (on top of drivers which are already inefficient), and the more phase problems you will have. These problems are not issues for the digital active crossover designer.

One example - suppose you wish to time align your drivers. If you are working with passive crossovers, you insert an all-pass filter. This involves phase rotation, causing more issues at the XO point. In the real world, it may not delay all parts of the frequency by the same amount - there by exacerbating phase issues. And you increase insertion loss. None of these are issues for a digital active XO designer.

I can tell you my own experience with a digital active XO (not DEQX). Even taking into account the learning curve, and the fact that I didn't know what I was doing, I managed to create a crossover which is superior to the passive analog XO that came with my speaker within a few months. Now that I have more of a clue, I suspect I could do it in less than a week if I was working on it full time. Someone who REALLY knows should take only a few days. Or a day.

 

[amirm]

By the way, I know of at least one designer who was more than happy to a) explain the crossover points and b) what to do better in an active crossover. Now this is an out of production product but still, to the extent the designer knows active can do better, it seems plausible that they would help get that version done. After all, it is still their speaker and making it sound the best would be good for them. Tons of people use Room EQ which implicitly changes the speaker response anyway so the concept is not new, or can e stopped.

 

[Cosmik]

It is not my intention to make it sound like an automatic process. Creating a passive crossover is not easy. The more components you include, the more insertion loss you will encounter (on top of drivers which are already inefficient), and the more phase problems you will have. These problems are not issues for the digital active crossover designer.

One example - suppose you wish to time align your drivers. If you are working with passive crossovers, you insert an all-pass filter. This involves phase rotation, causing more issues at the XO point. In the real world, it may not delay all parts of the frequency by the same amount - there by exacerbating phase issues. And you increase insertion loss. None of these are issues for a digital active XO designer.

I can tell you my own experience with a digital active XO (not DEQX). Even taking into account the learning curve, and the fact that I didn't know what I was doing, I managed to create a crossover which is superior to the passive analog XO that came with my speaker within a few months. Now that I have more of a clue, I suspect I could do it in less than a week if I was working on it full time. Someone who REALLY knows should take only a few days. Or a day.

No, I realise you weren't suggesting that a passive crossover can be designed easily - and the active version is, indeed, superior and only takes a couple of hours to create. As you say, a novice can come along and do something better than the person who has spent decades bashing his head against a wall.