ARC Genesis results

[peng]

Different mic positiong patterns would affect the deep bass results a lot. I have done mutiple calibration runs, in some cases, deep hass setting of 3.25 dB works well, in others, I had to use up to 4.375 dB.

When playing with deep bass and room gain, even increments of .125 dB could make enough difference to the look of the REW measured response curve. So, to me, it would be a hit and miss case if one makes changes without measuring before and other each change. As always, ymmv.

 

[hmt]

Imo the results of ARC I see in the forums are a bit underwhelming. The EQ does not seem to do precise filtering of modes. Audyssey or Dirac do a much better job here. ARC seems to adjust the general balance wrt bass but many modes are untouched.

 

[Beershaun]

I posted both with the original target curve, that is, without any post calibration adjustments, and also posted those with post calibration adjustments that obviously would alter the target curves. I posted them on different threads and several times so I don't know which is which anymore.

The one that shows the without adjustments to with adjustments (that is, deep bass and room gain), is like the one below, that I posted a couple times before but repeat here for convenience.

The black curve is the one after adding deep bass, and reduced room gain, the red one was when I had the Denon AVR, so that's with Audyssey:

As I said many time, I was disappointed that so few people, even Audio Advice and Audioholics didn't emphasize the need for serious users to use REW (or equivalent) and the ARCG app to get proper bass response.

ARCG is, at least in my case, failed to achieve "flat bass" per the default curve, but created major dips/suck outs (noted in my green curve). So they don't even make "flat response", but as you mentioned, "neutered bass" is a bad word. Though they didn't do too bad with the mid to lower mid bass, only screwed up the below 50 Hz, that I consider as "deep bass".

Based on my findings, and evidently yours as well, it is highly possible that many, if not most ARCG users may not know what the missed, i.e. the deep bass layer. Surely they may get great sound quality, depending on their speakers/rooms characteristics, but they could have done much better if they don't blindly trust internet hearsay, and ARCG's marketing and technical articles narratives.



View attachment 286325

Ah, yes agreed on all counts. I suspect many people are just increasing their subwoofer volume manually if they don't like the results.

 

[ban25]

Ah, yes agreed on all counts. I suspect many people are just increasing their subwoofer volume manually if they don't like the results.

BTW this is also a difference I noticed between Audyssey and Dirac. Audyssey tends to want you to turn the sub down during calibration, but if you leave it at that setting, Dirac will be forced to bring the mains up too high during calibration. So you may need to manually raise the gain on your sub if moving from Audyssey to Dirac.

 

[Beershaun]

BTW this is also a difference I noticed between Audyssey and Dirac. Audyssey tends to want you to turn the sub down during calibration, but if you leave it at that setting, Dirac will be forced to bring the mains up too high during calibration. So you may need to manually raise the gain on your sub if moving from Audyssey to Dirac.

Yes. Also with ARC you should use quick measure to set your sub level at -75db before running ARC genesis so it can set the appropriate signal level for the subs

 

[Beershaun]

OK, I re-ran my ARC Genesis calibration and then immediately ran REW using the same included mic that Anthem provides with the AVR. I set the REW levels at 80db as recommended by their video tutorial. I am still seeing the strange double dip in frequency response in REW as my last run so now I'm very confused as to what I am doing wrong with REW. Is it because I don't have a calibration file for my ARC Genesis Mic?
@peng welcome your thoughts and feedback as well as ideas for other views to provide.

So far I think things sound better now than before but it may just be due to the actual recordings I've been listening to and that my seating position is now further away from the front speakers and a little more equidistant between front and rear speakers.

ARC: I adjusted my low bass and room gain settings up a bit to see how that would affect the low end. It still seems like ARC G is trying for a flat curve rather than the raised bass response that their target curves indicate. I have attached my ARCG results files including my target curve adjustments for folks who are interested.

Here are the Images of the ARCG quick measure results for my left and right speakers and subs with ARCG on and off, and then REW measurements with ARCG on and off.

My speakers are:
Monitor Audio CP-IW460x 3-way. Crossover frequencies are 500hz 12db/octave and 3khz 18db/octave. Frequency response is 50hz-60khz -6db in room response.
SVS sb-2000

Summary:
REW seems to show much less difference both in SPL and in flattening of the frequency response than ARCG quick measure does. That said it also repeatably shows these crazy dips in my frequency response that I know are not real. So I don't yet trust my REW measurements and I suspect it has something to do with using the ARCG microphone. I welcome thoughts and suggestions on what may be going on with my frequency response dips from those who are much more versed in using REW. I can't move forward with more in-depth analysis until I sort out what is going on with those huge dips in frequency response and what is causing them in my measurement stack.

Left Speaker and Left Subwoofer comparisons:

This is the left front with ARC on and ARC off. The green is ARC on.

This is the left Subwoofer with ARC G on and off

This is the left front and sub separate snapshots same graph with ARCG on

Here is the REW snapshot of the left front and subwoofer together with ARCG on. I don't know what's going on above 1.5Khz. I suspect it's something to do with the Mic as this happened with both speakers. I welcome feedback on what might be happening? Is this the result of not having a mic calibration file?

Here is a REW snapshot of the left front and subwoofer overlaying ARCG on vs off.

Right Speaker and Right Subwoofer comparisons

Here is the Quickmeasure comparison of the right speaker with ARCG on and off:

Here is Quickmeasure with the Right Subwoofer ARCG on and off:

Here is REW measurement of the Right speaker and subwoofer together with ARCG on:

Here is REW overlay of the right speaker and sub woofer with ARCG on and off.

 

[peng]

OK, I re-ran my ARC Genesis calibration and then immediately ran REW using the same included mic that Anthem provides with the AVR. I set the REW levels at 80db as recommended by their video tutorial. I am still seeing the strange double dip in frequency response in REW as my last run so now I'm very confused as to what I am doing wrong with REW. Is it because I don't have a calibration file for my ARC Genesis Mic?
@peng welcome your thoughts and feedback as well as ideas for other views to provide.

So far I think things sound better now than before but it may just be due to the actual recordings I've been listening to and that my seating position is now further away from the front speakers and a little more equidistant between front and rear speakers.

ARC: I adjusted my low bass and room gain settings up a bit to see how that would affect the low end. It still seems like ARC G is trying for a flat curve rather than the raised bass response that their target curves indicate. I have attached my ARCG results files including my target curve adjustments for folks who are interested.

Here are the Images of the ARCG quick measure results for my left and right speakers and subs with ARCG on and off, and then REW measurements with ARCG on and off.

My speakers are:
Monitor Audio CP-IW460x 3-way. Crossover frequencies are 500hz 12db/octave and 3khz 18db/octave. Frequency response is 50hz-60khz -6db in room response.
SVS sb-2000

Summary:
REW seems to show much less difference both in SPL and in flattening of the frequency response than ARCG quick measure does. That said it also repeatably shows these crazy dips in my frequency response that I know are not real. So I don't yet trust my REW measurements and I suspect it has something to do with using the ARCG microphone. I welcome thoughts and suggestions on what may be going on with my frequency response dips from those who are much more versed in using REW. I can't move forward with more in-depth analysis until I sort out what is going on with those huge dips in frequency response and what is causing them in my measurement stack.

Left Speaker and Left Subwoofer comparisons:

This is the left front with ARC on and ARC off. The green is ARC on.
View attachment 290071

This is the left Subwoofer with ARC G on and off
View attachment 290072

This is the left front and sub separate snapshots same graph with ARCG on
View attachment 290073

Here is the REW snapshot of the left front and subwoofer together with ARCG on. I don't know what's going on above 1.5Khz. I suspect it's something to do with the Mic as this happened with both speakers. I welcome feedback on what might be happening? Is this the result of not having a mic calibration file?
View attachment 290074
Here is a REW snapshot of the left front and subwoofer overlaying ARCG on vs off.
View attachment 290078

Right Speaker and Right Subwoofer comparisons

Here is the Quickmeasure comparison of the right speaker with ARCG on and off:
View attachment 290075

Here is Quickmeasure with the Right Subwoofer ARCG on and off:
View attachment 290076

Here is REW measurement of the Right speaker and subwoofer together with ARCG on:
View attachment 290077

Here is REW overlay of the right speaker and sub woofer with ARCG on and off.
View attachment 290080

I am quite sure the Anthem ARCG mic is specifically calibrated for use with ARCG. So your REW graphs will not represent the response accurately if you use the Anthem mic.

To use REW, you need to use a generic calirated mic such as the Umik-1 or Umik-2 mic that come with a calibration file, and use the file (from Umik).

 

[Beershaun]

I am quite sure the Anthem ARCG mic is specifically calibrated for use with ARCG. So your REW graphs will not represent the response accurately if you use the Anthem mic.

To use REW, you need to use a generic calirated mic such as the Umik-1 or Umik-2 mic that come with a calibration file, and use the file (from Umik).

Thanks. That's what I was thinking but figured it wouldn't cost anything to try it out first. Will have to invest in a umik

 

[peng]

Thanks. That's what I was thinking but figured it wouldn't cost anything to try it out first. Will have to invest in a umik

I took a better look of your REW graphs and I think you should investigate further on that huge dip at just over 2K to just below 3K. Calibration files for either mics in that frequency range could explain a few dB but not >10 dB. That dip is still quite narrow so may not be too audible but it is weird, make me think if there is something wrong with the crossover, or something not right about your mic positions. Are the fronts the Elac DBR62? Do both the left and right have the same huge dip there?

 

[Beershaun]

I took a better look of your REW graphs and I think you should investigate further on that huge dip at just over 2K to just below 3K. Calibration files for either mics in that frequency range could explain a few dB but not >10 dB. That dip is still quite narrow so may not be too audible but it is weird, make me think if there is something wrong with the crossover, or something not right about your mic positions. Are the fronts the Elac DBR62? Do both the left and right have the same huge dip there?

Thanks for the additional feedback.
speakers are:
Monitor Audio CP-IW460x 3-way. Crossover frequencies are 500hz 12db/octave and 3khz 18db/octave. Frequency response is 50hz-60khz -6db in room response.

Yes it is repeatable on both sides.

They sit behind my acoustically transparent projector screen.

If this is real it could match my original issue I started this whole exercise about with lack of mid range energy.
It almost as if the tweeter is crossing over properly but the mid range is starting too early. Or the projector screen itself is creating the problem.

I could try doing a frequency sweep with my phone decibel meter app and seeing what I get with a 2khz test tone vs a 1khz test tone.


PS: I just played a simple set of test tones from 1khz to 4khz through both channels and recorded the spl on a simple spl meter app on my phone and they were all within 4-5db of each other around 65db. So I don't think the dip is real...

 

[peng]

Thanks for the additional feedback.
speakers are:
Monitor Audio CP-IW460x 3-way. Crossover frequencies are 500hz 12db/octave and 3khz 18db/octave. Frequency response is 50hz-60khz -6db in room response.

Yes it is repeatable on both sides.

They sit behind my acoustically transparent projector screen.

If this is real it could match my original issue I started this whole exercise about with lack of mid range energy.
It almost as if the tweeter is crossing over properly but the mid range is starting too early. Or the projector screen itself is creating the problem.

I could try doing a frequency sweep with my phone decibel meter app and seeing what I get with a 2khz test tone vs a 1khz test tone.


PS: I just played a simple set of test tones from 1khz to 4khz through both channels and recorded the spl on a simple spl meter app on my phone and they were all within 4-5db of each other around 65db. So I don't think the dip is real...

I would trust REW more because it is a known quantity, whereas ARCG's FR graphs are based on their proprietary stuff that we don't know much about. It is interesting to see that in your post#63, those REW graphs look quite normal:

So what changes?

It seems it started since your post#70 in which you said "OK, well this is interesting...I calibrated my audio interface to get in the "green zone" for REW so I was getting a higher input level at the mic. I did a run with the interface at 100% (mic input level in Windows 11) and a second run with the interface at 90% (mic input level in windows11). At 100% I was getting -3db headroom in REW (red zone) when running L+R measurement so my audio interface may have been too high. When I ran the second run of L+R with the mic input at 90% I was getting 120db headroom (green zone) from REW when measuring. The first graph below shows the differences comparing using the two different input calibration files."

I never understood what that meant, except I found the level was way too high, really weird stuff and I have wonder if you might have done something that should not have been done. Also, it looked so normal that I would have assumed you were using a Umik-1 or Umik-2 mic with the proper calibration file. Can you reconfirm which mic you were using for that measurements? I can't be the ARCG mic because I remember when I used it with REW, the graphs show obvious difference that those mics were not calibrated the same as the Umik-1 mic and that ARCG must have it calibrated for use with ARCG only.

The huge dips in those graphs are nothing unusual if no smoothing is applies, everyone gets those kind of dips in various frequencies from the low to higher mid range frequencies. Yours look very suspicious only because you apparently did apply 1/24 smoothing and with such smoothing, it seems unusual to have such a dip in the midrange. ARCG's will never show such big dips unless there is something wrong, such as if you have defective crossovers, because they clearly use smoothing, we don't know how much smoothing they applied to those graphs.

Can you attach the REW files so I can look into it further?

Here's the one you posted in post#63, it looked pretty good to me in the 2-3 kHz range.

 

[hmt]

Maybe it is because he measured LR combined and it is crosstalk cancellation?

edit:

OK, he measured only the R speaker. Nevertheless it seems that there is some severe comb filtering going on in his REW measurement.

 

[peng]

OK, he measured only the R speaker. Nevertheless it seems that there is some severe comb filtering going on in his REW measurement.

Or floor, ceiling, wall bounces etc., so again, that's normal except with 1/24 smoothing it should look much smoother than what he's getting as show in post#86, especially if gated. Something is wrong..., but REW is just doing its job.

 

[Beershaun]

Sorry for the delay responding. I appreciate the help and support while I work through learning and using REW and comparing to ARC. And ultimately understanding my actual in room response.

I'll respond to the questions posted point by point in a follow up post.

I did want to mention that I did a simple tone sweep from 1khz to 4khz through both speakers together to see if the dip was real. The results were that there was no huge dip in response. My phone's simple sound spectrum app showed the tones all between 60 and 65db through that range. I'll post specific answers to the questions when I get some time.

 

[peng]

Sorry for the delay responding. I appreciate the help and support while I work through learning and using REW and comparing to ARC. And ultimately understanding my actual in room response.

I'll respond to the questions posted point by point in a follow up post.

I did want to mention that I did a simple tone sweep from 1khz to 4khz through both speakers together to see if the dip was real. The results were that there was no huge dip in response. My phone's simple sound spectrum app showed the tones all between 60 and 65db through that range. I'll post specific answers to the questions when I get some time.

I hear you, but you also missed my point that REW is just doing its job. It shows you those narrow dips because you used 1/24 smoothing whereas your phone app may have very different smoothing, and ARCG certainly use smoothing that could be as much as 1/6. Go back to your REW graph and try using 1/6, or psychological, variable smoothing and see what happens. All these may not explain everything, but likely something, to a large extent.

 

[Beershaun]

I hear you, but you also missed my point that REW is just doing its job. It shows you those narrow dips because you used 1/24 smoothing whereas your phone app may have very different smoothing, and ARCG certainly use smoothing that could be as much as 1/6. Go back to your REW graph and try using 1/6, or psychological, variable smoothing and see what happens. All these may not explain everything, but likely something, to a large extent.

I understand. Let me be more precise.

I have a series of test tone files at specific frequencies, 1000, 1250, 1600, 2000, 2500, 3150, 4000. Each is played for 10 seconds. Each of those tones registered within 60-65db playing through my two front speakers. I was expecting a significant dip in spl for the tones that are in the dip area. So my thinking is I am still doing something wrong with REW. I still owe you the file and additional info. So will send that in a follow up post.

 

[peng]

I understand. Let me be more precise.

I have a series of test tone files at specific frequencies, 1000, 1250, 1600, 2000, 2500, 3150, 4000. Each is played for 10 seconds. Each of those tones registered within 60-65db playing through my two front speakers. I was expecting a significant dip in spl for the tones that are in the dip area. So my thinking is I am still doing something wrong with REW. I still owe you the file and additional info. So will send that in a follow up post.

My same point would still apply though, how did you measure the spl when playing the test tone? Are you sure you captured the lowest dips that the Umik mic and REW captured with no smoothing/averaging? Spl meters typically won't show those narrow dips.

That's is not to say there is nothing wrong about your REW graphs, and you may be right about that.

 

[Beershaun]

Can you attach the REW files so I can look into it further?
Here's the one you posted in post#63, it looked pretty good to me in the 2-3 kHz range.

Here are my in-line responses to your questions. @peng Thanks for your continued help and interest.

I never understood what that meant, except I found the level was way too high, really weird stuff and I have wonder if you might have done something that should not have been done. Also, it looked so normal that I would have assumed you were using a Umik-1 or Umik-2 mic with the proper calibration file. Can you reconfirm which mic you were using for that measurements?

I was using the ARG included mic. Agreed, when I first ran REW everything looked really normal and expected. It all changed after I attempted to create an audio interface calibration file based on the video instructions and then did a test sweep at too high a volume. I tried it in one set of measurements but did not use that calibration file in the measurements posted in post #86 to see if that would fix it. It did not. I am now wondering if I somehow broke the mic or if the calibration file is still being used even though I removed it?

What I did do for the measurements in post #86 is set the recording levels at 80db as instructed in the REW instruction video. To your point, it was much too loud in the prior set of tests. Specifically: I used my phone SPL meter app to measure the REW test tone at the mic listening position adjusting the volume knob on the anthem to meet 80db on my phone spl meter app. Then, I adjusted the audio input on my PC until the Mic input SPL in REW also said 80db.

The huge dips in those graphs are nothing unusual if no smoothing is applies, everyone gets those kind of dips in various frequencies from the low to higher mid range frequencies. Yours look very suspicious only because you apparently did apply 1/24 smoothing and with such smoothing, it seems unusual to have such a dip in the midrange. ARCG's will never show such big dips unless there is something wrong, such as if you have defective crossovers, because they clearly use smoothing, we don't know how much smoothing they applied to those graphs.

Agreed. We don't know what the smoothing is for ARCG. I created a graph with 1/3 smoothing in REW and am still seeing a significant dip though.

Can you attach the REW files so I can look into it further?

I tried, unfortunately the forum software will not let me upload the file with that extension. Trying to PM you the file.

I suppose I should get a umik-1 and try to do it correctly at this point since the results seem strange and likely user error.

 

[ad_fletch]

Ah, yes agreed on all counts. I suspect many people are just increasing their subwoofer volume manually if they don't like the results.

Guilty as charged!

This is a very interesting thread for me. I last ran ARCG 1.5.4 using my MRX 310 just over a year ago, and apart from low sub levels was pretty happy with the results. I verified the low deep bass with REW at the time, but have since sold my UMIK-1 unfortunately, so can't do same now.

I have downloaded the latest ARCG version (1.7.4 apparently) and intend to run it soon when the family is out. I am curious about the auto phase/distance functionality for my AVR, because it lacks the multiple sub outs of subsequent generations I ended up just setting the sub distance to an in-between setting where it sounded in time with the rest of the music, which is obviously not ideal.

Obviously I can try and see, but has anyone else run the newer ARCG on an MRX x10 series unit and obtained good dual sub results?

 

[Sideshowdad]

Hey all, just ran arc for the first time today on my AVM70.

I noticed it's set my sub to 30 feet? Is that normal?

I did have a message during setup saying my sub exceeded max target and arc will switch to relative delays, I selected ignore to continue.

Should manually adjust it?

I do have 2 sub but connected as one to sub1 on the AVM.