Many new receivers for home theater use off microphone calibration such as the Audyssey MultEQ. This allows for full digital correction of the frequency response on all channels including the sub outputs. It corrects delays as well. Using this I have been able to put 4 3" aluminum drivers from TB on a small open baffle in a vertical line and get a flat response from 150hz to 20k with seamless integration with a 12" powered sub woofer. It's a great way to deal with the top end sensitivity of metal drivers. You can also mix all sorts of speakers together such as a center channel that is poorly or perhaps I should say horribly match to my mains. I wouldn't hesitate to mix dipoles with conventional speakers either, without the calibration that would be completely insane.
I started this thread because I want to see what others are doing with this cheap technology. Anyone put a pair of Markaudio Alpair 10.2's on a open baffle and try to force them down to 80hz? I bet that would work very well! You might even be able to do it with a cheap 4" driver like the chr-70.
The idea here IMHO is to build speakers that have horrible response but good dispersion characteristics and just let the correction do it's thing. We need to let the lack of crossover components make up for the damage done by all the extra DSP and cheap amplification. Frankly I can't believe people aren't doing back flips over this calibration stuff.
These amps are in the $300 range like my Denon AVR-591 (discontinued). I bought it for $280 off the internet when it was in production.
For the record, the calibration is not very good at dealing with room modes and reflections, but it is awesome for correcting the output of speakers. That's ironic because it wasn't the intent of the technology.
I started this thread because I want to see what others are doing with this cheap technology. Anyone put a pair of Markaudio Alpair 10.2's on a open baffle and try to force them down to 80hz? I bet that would work very well! You might even be able to do it with a cheap 4" driver like the chr-70.
The idea here IMHO is to build speakers that have horrible response but good dispersion characteristics and just let the correction do it's thing. We need to let the lack of crossover components make up for the damage done by all the extra DSP and cheap amplification. Frankly I can't believe people aren't doing back flips over this calibration stuff.
These amps are in the $300 range like my Denon AVR-591 (discontinued). I bought it for $280 off the internet when it was in production.
For the record, the calibration is not very good at dealing with room modes and reflections, but it is awesome for correcting the output of speakers. That's ironic because it wasn't the intent of the technology.
Hi,
I really don't think it's that simple.
There's a big chance that by using this approach, you will be 'fixing'
problems by applying a band-aid in the wrong place.
If a bump or dip in the frequency-response is caused by, for instance,
resonance(s) or a diffraction-problem, correcting the on-axis response
will not fix the cause of the problem.
The problem may show up elsewhere, f.i. in power-response.
As a random example, i once built a small 2-way bassreflex loudspeaker
that had a ~1 dB bump at 1 Khz.
A close-up measurement of the bassreflexport and a waterfalldiagram
of the system showed the problem came from a big portresonance.
I could have 'fixed' the problem by adjusting the filter to make the bump disappear, but the resonance would still be there.
The fix i settled on was making the port smaller and consequently shorter.
The port-resonance was gone, downside was increased port-noise.
Since the midwoofer i used was already screaming for help when it had to produce the spl that made the portnoise audible,
i think the resulting compromise was okay.
Best regards,
Klaas
I really don't think it's that simple.
There's a big chance that by using this approach, you will be 'fixing'
problems by applying a band-aid in the wrong place.
If a bump or dip in the frequency-response is caused by, for instance,
resonance(s) or a diffraction-problem, correcting the on-axis response
will not fix the cause of the problem.
The problem may show up elsewhere, f.i. in power-response.
As a random example, i once built a small 2-way bassreflex loudspeaker
that had a ~1 dB bump at 1 Khz.
A close-up measurement of the bassreflexport and a waterfalldiagram
of the system showed the problem came from a big portresonance.
I could have 'fixed' the problem by adjusting the filter to make the bump disappear, but the resonance would still be there.
The fix i settled on was making the port smaller and consequently shorter.
The port-resonance was gone, downside was increased port-noise.
Since the midwoofer i used was already screaming for help when it had to produce the spl that made the portnoise audible,
i think the resulting compromise was okay.
Best regards,
Klaas
I took a look on Ebay and there are about (6) AVR-591 for sale at around 200.00. Nice price for what you get. I didn't realize it was capable of equalizing ALL the outputs. Nice feature. Why aren't there more receivers, whether 5.1 or 7.1, with is built in??
I'm heading in that direction for 2 channel. I am getting a DCX2496 and building a active 3 way tower system to start. IB Subwoofer later. The DCX will do the active XO and equalization.
To me, the fullrangers have limited appeal. Limited on the low end for power handling and not really enough guts for large rooms. But nice for small to medium areas. I suspect that applying equalization on the low end to pump up the bass output could be hazardous to their mechanical long term health. But equalization is ideal to deal with the lumps and bumps elsewhere. Not having crossovers screwing things up is a very nice feature.
If most people listen to MP3 crippled sound it means to me that they don't really care about quality sound. That's what is sad. Equalization? Too much complexity for most.
I'm heading in that direction for 2 channel. I am getting a DCX2496 and building a active 3 way tower system to start. IB Subwoofer later. The DCX will do the active XO and equalization.
To me, the fullrangers have limited appeal. Limited on the low end for power handling and not really enough guts for large rooms. But nice for small to medium areas. I suspect that applying equalization on the low end to pump up the bass output could be hazardous to their mechanical long term health. But equalization is ideal to deal with the lumps and bumps elsewhere. Not having crossovers screwing things up is a very nice feature.
If most people listen to MP3 crippled sound it means to me that they don't really care about quality sound. That's what is sad. Equalization? Too much complexity for most.
What you are saying is true but I have found the calibration tends to work wonders with the more exotic speaker designs because they do not have a flat response to begin with. For example, my center channel without calibration is horrible, it would drive you out of the room. However, once the calibration is enabled it is smooth as glass. In this example I am boosting the hell out of the low end on these 3"ers and really lowering the output of the top end especially since these are metal cone drivers. This is why I am using 4 of them, that gives me 80watts RMS power handling to allow these things to do what I'm asking them to do. Normally I could not use a metal driver because they are too sensitive in the high frequency range. Now I am able to take advantage of the resolution of the metal drivers and the fantastic dispersion pattern of the dipole design not to mention the extra wide effect that a vertical array has to offer. Also, dipole speakers of this type have a strange oscillating response if you look at a graph, an eq cannot deal with that or it would be very difficult deal with.
The point I'm trying to make is that we can now disregard frequency response in our speaker designs and focus on other aspects such as dispersion, imaging, coherency and point source sound.
The point I'm trying to make is that we can now disregard frequency response in our speaker designs and focus on other aspects such as dispersion, imaging, coherency and point source sound.
I need to point out that you make a very good point! When I said we can disregard frequency response, we still need to pay attention to things like diffraction and port noise. Those are things that calibration does not deal with well. It also doesn't handle room effects very well so treatments of the room are still a good idea.
Forgot to mention, for your active 3 way, you can still use calibration for free with windows vista and windows 7. Look up room correction. I am not a big fan of this feature because it only works on some sound cards, usually aftermarket cards do not support it. And you will be using the analog output of the computer which are not very good. The Audyssey system is cool because it does the adjustment in the digital domain and then sends the adjusted sound the the DACs just before amplification. That's the way to do it. I would like to see a receiver with the calibration and multiple sub outputs. A 7.3 or something. Then it would be a whole new ballgame for the full range driver.
I don't think it is ironic, because in-room modes and reflections are diffused, while what needs to be corrected, is speaker frequency response. I suppose it is smart enough to find the same bumps of each speaker in 6 positions of microphone to attenuate them only.
Yes, Audissey does great job even with bad speakers, but if to remove all possible mechanical ringings before it does much better job. I have currently line arrays in walls: 8 of 4" mid rangers per side, and 16 1" tweeters per side, 12" woofers, in concrete boxes, a pair of 6" in wall full - rangers with vertical stick of 64 of 1/4" drivers as a tweeter in center channel, and pair of 6" coaxials in the rear wall. Plus concrete horn under floor for below 40 Hz frequencies.
Audissey did a nice job equalizing everything very well.
Indeed the 591 is a cheap product. There is much better stuff out there. I have been quite impressed with more expensive offerings that feature this calibration. Still, the 591 sounds great, the biggest problem it has is that there are no multi-channel inputs and the cooling solution is inadequate. That's why I use a cooling fan on mine. It just sits on top with silicone feet and blows upward. The 591 does have 2 rca outputs to support 7.1 but there is no amplification for them so they work like the sub output. All 8 channels are calibrated though. For $200 on ebay that's an amazing value.
Indeed, ideally the DSP can generate what speaker designers like Linkwitz have been doing manually for the past 30 years or so - and get fairly close by brute force and computerised decision making.
I think you'll find that you won't be able to get away without a crossover system for individual speaker units though, because with those types of speakers the dispersion will inevitably be fairly bad where the drivers overlap, and AVR calibration cannot compensate well for the issues that occur in 3D. I'm sure with almost anything you'll be able to get fairly ample results at a single location by slumping in your seat with the microphone on your head during calibration, but I don't see a conventional multi-way (of dissimilar drivers) being aptly compensated by the calibration alone.
I believe it is not a compromise, it is rather careful discovery of real frequency response that dos not depend on microphone placement. When you move microphone and observe results on spectrum analyzer you can see that some bumps and dips depend on microphone placement, some don't. Which don't have to be equalized. It would be a big mistake to equalize those which existence depends on microphone placement.
Also, it aligns speakers by phase. It is good that such smart systems can be used when reproducing recorded sounds because latency does not matter, and video signal in cinema theater can be delayed as well to match sounds. But unfortunately for sound reinforcement systems such latency is not acceptable.
if you want the RCA outs for all of the channels, i think you need to get a Denon 4-digit model, like the 3311, but even if you go with the "low end" of the 4-digit models (like the 1311) it will still be expensive (even though the 1311 is made at the same factory that makes the 3-digit models) the other option would be to make a "breakout" cable adapter that connects to the ribbon cable between the preamp and power amp. you would need a 10-pin cable with a socket on one end, the connector for the amp board on the other end, and cables going to an RCA jack block from in between. that way, if you want to sell the receiver in it's original configuration, you can remove the breakout cable and plug the preamp directly back into the power amp.
Last edited:
Actually, I bought 591 with an intention to throw out power amps and use tube outputs instead, but found that output amps of this thingy sound surprisingly well. I did not open it still even to see what is inside, and use it as is almost a year. Sometimes I have to switch it off and on when start NetFlix movie because somehow it screws up audio decoder, so it needs to be hard resetted. Also, it would be nice to have D/A video converters inside, because I would love to continue using my good old Sony VPL-PX30 projector with modern video signal sources.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.