Emotiva RMC-1 AV Processor Review

[digicidal]

I disagree. No company would disclose details about how their room EQ algorithms work, but luckily for us that is not necessary, as it is not about how they work but about how WELL they work. The only way to test that is to measure the corrected response and check it against the target that has been set. I don't remeber seeing that been done anywhere..

Other than the older Harman tests, and users posts a page or two back in this thread? But yes, I understand and agree with your real point. Easy enough if we all pitched in... not sure how many will be required to dispel criticisms of flawed testing methodologies however.

 

[markus]

I disagree. No company would disclose details about how their room EQ algorithms work, but luckily for us that is not necessary, as it is not about how they work but about how WELL they work. The only way to test that is to measure the corrected response and check it against the target that has been set. I don't remeber seeing that been done anywhere..

But it doesn't make sense to measure each and every time as the implementation is always the same. Better describe the RCS in a dedicated thread.

There's plenty of measurements out there showing before/after graphs.

What exactly are you interested in and how do you want to measure it? Every room is different so the very same RCS might behave differently depending on the actual acoustical situation.

 

[QMuse]

But it doesn't make sense to measure each and every time as the implementation is always the same. Better describe the RCS in a dedicated thread.

I said that same thing few posts above.

There's plenty of measurements out there showing before/after graphs.

Where?

What exactly are you interested in and how do you want to measure it?

Same as with speakers: you measure the difference in linearity between corrected response and target curve.

Every room is different so the very same RCS might behave differently depending on the actual acoustical situation.

You measure and compare several systems in the same room using the same target curve.

Btw, rooms in living spaces are not that different. They differ in dimensions which will affect the frequencies of room modes, but that is not a fundamental difference.

 

[TimoJ]

What exactly are you interested in and how do you want to measure it? Every room is different so the very same RCS might behave differently depending on the actual acoustical situation.

Maybe use a shielded dampered box and put the microphone inside and just measure what happens to SNR, THD, SINAD etc. when the room correction is enabled. I'd be more interested of those measurements vs. the actual room correction curves.

 

[QMuse]

Maybe use a shielded dampered box and put the microphone inside and just measure what happens to SNR, THD, SINAD etc. when the room correction is enabled. I'd be more interested of those measurements vs. the actual room correction curves.

You are missing the point. When you do room EQ absolutely nothing happens to SNR, THD and SINAD. The only thing that is changed is speaker's frequency and phase response.

 

[markus]

Where?

Check AVS. I've posted such kind of measurements for MultEQ and Dirac Live years ago. Others did too.

Same as with speakers: you measure the difference in linearity between corrected response and target curve.

Not interested in the time domain? Most RCS use multiple mic positions. How do you want to account for that?

Btw, rooms in living spaces are not that different. They differ in dimensions which will affect the frequencies of room modes, but that is not a fundamental difference.

There are fundamental differences. For example between Europe and the US. In the EU you most often see textbook like modal behavior whereas in the US bass can be very good out of the box due to predominant framed wall construction techniques.
Depending on the location of speaker and listener you might see a high amount of non-minimum phase issues that can't be corrected (entirely). At other locations even MultEQ will do a fantastic job.

No offense meant but I think you need to read up on room correction in general. Good primer:
http://diracdocs.com/on_room_correction.pdf
On the first word you get stuck try to find more resources to learn. At least that's what I do it

 

[markus]

You are missing the point. When you do room EQ absolutely nothing happens to SNR, THD and SINAD. The only thing that is changed is speaker's frequency and phase response.

RCS have a significant impact on distortion numbers as you need to make up for the gain loss somewhere. The filters themselves might add noise.

 

[TimoJ]

You are missing the point. When you do room EQ absolutely nothing happens to SNR, THD and SINAD. The only thing that is changed is speaker's frequency and phase response.

No, I'm not. Lots of things happen when room eq is switched on in the signal chain. So measuring just with Direct or Reference modes won't give you the whole "picture".

 

[markus]

Maybe use a shielded dampered box and put the microphone inside and just measure what happens to SNR, THD, SINAD etc. when the room correction is enabled. I'd be more interested of those measurements vs. the actual room correction curves.

Not sure this will generate useful data because using RC can have a significant positive impact on perceived sound quality, more than what a couple of dB in SNR loss might suggest.

I'm not against measuring RC implementations. I think they are necessary to check if there are implementation errors but it's nearly impossible to do for us because RCS vendors don't disclose much about their implementation. I can provide you an educated guess but it won't be enough to do the kind of measurements people here are asking for.

 

[QMuse]

Check AVS. I've posted such kind of measurements for MultEQ and Dirac Live years ago. Others did too.

Do you really consider those measurements a systematic test of various room EQ solutions?

Not interested in the time domain? Most RCS use multiple mic positions. How do you want to account for that?

Time domain correction is done based on a sweep made from a center of the LP. It wouldn't make sense to do vector average of multiple sweeps as a basis form time domain corrections as phase doesn't change as much as frequency response in various points around LP.

Unfrotuantely what none of the room EQ solutions do is phase matching between channels so they don't compensate for LF cancellations due to phase mismatch.

There are fundamental differences. For example between Europe and the US. In the EU you most often see textbook like modal behavior whereas in the US bass can be very good out of the box due to predominant framed wall construction techniques.
Depending on the location of speaker and listener you might see a high amount of non-minimum phase issues that can't be corrected (entirely). At other locations even MultEQ will do a fantastic job.

I agree that in EU average desntiiy of walls, floor and ceiling is higher than in USA, but that really depends on the type of building you live in. Doing room EQ in a small room with concrete walls is definitely a more complex task than in the larger room with walls of less dense material but the idea of the test would be to see how different room EQ systems compare in the same environment (room).

No offense meant but I think you need to read up on room correction in general. Good primer:
http://diracdocs.com/on_room_correction.pdf
On the first word you get stuck try to find more resources to learn. At least that's what I do it

Oh, I have done much much more reading than that. I do my corrections mannualy with rePhase and have yet to see an automated system that dose the job better than my manual corrections. Can you say the same for yourself?

 

[QMuse]

No, I'm not. Lots of things happen when room eq is switched on in the signal chain. So measuring just with Direct or Reference modes won't give you the whole "picture".

Pls trust me with this - if there is a change in SNR, THD or SINAD when you activate room EQ than the product is a complete junk. It simply shouldn't happen, except in the scenario where analog signal input is provided so ADC needs to be done to apply room EQ to it.

 

[TimoJ]

Not sure this will generate useful data because using RC can have a significant positive impact on perceived sound quality, more than what a couple of dB in SNR loss might suggest.

Actually instead of a "box", a direct electric loop connection should be used so that the mic input would pickup the measurement signal. I'm just wondering what horrors (and software bugs) a measurement like that would reveal on some units. I don't recall anyone doing these measurements.

 

[markus]

Do you really consider those measurements a systematic test of various room EQ solutions?

No, but they should show what you've asked.

Time domain correction is done based on a sweep made from a center of the LP. It wouldn't make sense to do vector average of multiple sweeps as a basis form time domain corrections as phase doesn't change as much as frequency response in various points around LP.

Speaker distance is calculated based on a sweep made from a center of the LP. Time domain correction is most powerful when multiple locations are considered.

Unfrotuantely what none of the room EQ solutions do is phase matching between channels so they don't compensate for LF cancellations due to phase mismatch.

DL does and it will do it even better once their BM solution is out.

I agree that in EU average desntiiy of walls, floor and ceiling is higher than in USA, but that really depends on the type of building you live in. Doing room EQ in a small room with concrete walls is definitely a more complex task than in the larger room with walls of less dense material but the idea of the test would be to see how different room EQ systems compare in the same environment (room).

So which conditions will ensure that results of RCS testing would be universally applicable?

Oh, I have done much much more reading than that. I do my corrections mannualy with rePhase and have yet to see an automated system that dose the job better than my manual corrections. Can you say the same for yourself?

Keep on reading I've done extensive testing with Acourate. Once you're past confirmation bias you'll realize there's a difference between a good looking filter and a good sounding one. There is no silver bullet as our understanding of sound perception is still too limited.

Have you read through the papers by Bharitkar and Brännmark? If not shoot me your email.

 

[markus]

Actually instead of a "box", a direct electric loop connection should be used so that the mic input would pickup the measurement signal. I'm just wondering what horrors (and software bugs) a measurement like that would reveal on some units. I don't recall anyone doing these measurements.

I've done plenty of these and I've seen a ton of implementation disasters. Usually I try to get it resolved first by contacting the manufacturer. But if they throw back a "works as designed" then I often discuss them on forums.

 

[QMuse]

No, but they show what you've asked for.

That is not what I asked for. I asked for systematic comparative test of various rom EQ solutions.

No, but they show what you've asked for.
Speaker distance is calculated based on a sweep made from a center of the LP.

Actually speaker's delay is calculated based on a sweep from a center of the LP.

No, but they show what you've asked for.
Time domain correction is ost powerful when multiple locations are considered.

As I explained no need for that as phase response is not changing as much as frequency response is. Dirac definitely doesn't do vector averaging to calculate average sweep for time domain calculation. And juding by the corrected frequency response Dirac also heavilly relies on the sweep made from the center of the LP. Try doing a true spatial meaasurement with RTA moving microphone method and you will see how well Dirac did the job.

So which conditions will ensure that results of RCS testing would be universally applicable?

There is no such thing as a reference room so yny room would do for a comparative test of room EQ solutions. Do you think heavilly treated room or anechoic chamber would be a better place to test room EQ solutions?

Keep on reading I've done extensive testing with Acourate.

Wow! So playing with various target curves with a room EQ system that somebody else developed makes you an expert? How about trying to build filters mannualy and see if you can do it better than Acourate? Or maybe you didn't read enough papers to know how to do that, ha?

 

[markus]

Wow! So playing with various target curves with a room EQ system that somebody else developed makes you an expert? How about trying to build filters mannualy and see if you can do it better than Acourate? Or maybe you didn't read enough papers to know how to do that, ha?

With all due respect but you don't seem to have any experience whatsoever with Acourate. It can do anything you want. It's basically a FIR filter designer.

I thought you wanted to learn more about MultEQ, Dirac Live and room correction in general, that's why I offered you access to said papers.

 

[QMuse]

With all due respect but you don't seem to have any experience whatsoever with Acourate. It can do anything you want. It's basically a FIR filter designer.

Trust me, I'm quite familiar with Acourate. IMO that is the best automated room EQ money can buy. I also think very highly of Dr. Uli Brueggemann who built it. But while I agree with you it can do many things it still doesn't allow flexibility you have when building filters mannually with rePhase.

Btw, main idea behind building filters mannualy was not to save money for not paying Acourate licence but to prove to myself I have learned enough to build filters of my own that in some aspects are better than those built by Acourate for my speakers/room scenario.

I thought you wanted to learn more about MultEQ, Dirac Live and room correction in general, that's why I offered you access to said papers.

As I said, I read pretty much everything I could find on the room EQ topic before starting to build my own filters with rePhase. I also have acccess to AES library and I read pretty much everything there is there related to room EQ, so thank you for your kind offer.

 

[oldsysop]

It's strange . The title says "Emotiva RMC-1 AV Processor Review", but it is very difficult to find information on 43 pages.

 

[Koeitje]

There's a ton of them... but it's rarely advertised. Just look at the amps for a low-level output or "satellite" output. Most of the Axiom subs have them, SVS subs too, Rythmik as well, but not with all amps. In general, most DSP enabled subs and almost all pro-audio subs will (since they're designed to pair with their own powered monitors in many cases).

For Axiom it looks like you need 2 subs to get it working, and its not clear to me if the crossover setting is both high and low pass?

I can't find any high pass settings in the manual of the SVS SB-16 Ultra, didn't bother with the lower models.

For Rythmik the output on the 2xRCA version is limited to 80hz high pass (which is too high), and for the 1XLR version its unclear but that means I still need 2 subwoofers.

 

[Dimifoot]

It's strange . The title says "Emotiva RMC-1 AV Processor Review", but it is very difficult to find information on 43 pages.

Read the first pages. There is enough info for you to lose all interest on the device.