AVS Forum banner

2681 - 2700 of 2702 Posts

·
Registered
Joined
·
719 Posts
Just noticed in my PDF from ARC genesis it says my sub is at -2db but after uploading the arc genesis data my sub says +1db in the level calibration menu 🤔 that happen to anyone else?
Bet your room gain is around 3 (ARC adds the room gain to the sub in the values shown in the level calibration menu)
 

·
Registered
Joined
·
719 Posts
Thanks, yeah 4.8db room gain. Should I just leave it at +1db in the level calibration then?
Yes. Unless, according to the manual, you turn ARC off - then you need to reduce it by the room gain (otherwise, double bass).
 
  • Like
Reactions: Andy922

·
Registered
Joined
·
298 Posts
Just noticed in my PDF from ARC genesis it says my sub is at -2db but after uploading the arc genesis data my sub says +1db in the level calibration menu 🤔 that happen to anyone else?
The value shown in the menu includes the effects of applying Room Gain. The value in ARC Genesis does not.
 
  • Like
Reactions: Andy922

·
Registered
Joined
·
298 Posts
For those who run ARC non-full-range and find the sound is too bright or dull, the issue may be levels (as set in ARC), as discussed in this post of mine:

As pointed out, you can run ARC only through the bass range (however you define that) if you only want to correct the bass. If you do that, here is a critical point. With correction over a limited frequency range, the "speaker level" settings in ARC are tremendously important in controlling overall tonal balance. That's because speaker levels set the level of the (uncorrected) treble, in contrast to the level of the corrected bass, which is set by the target curve. This moving the speaker levels up or down by 0.5 dB can change the overall balance quite noticeably.
If the target curve doesn't match the average response above the max correction frequency, the level imbalance can make your sound too bright or dull. And even if it sort of matches, a little adjustment might sound better to your ears. The way to fix this is to change the level of that speaker (or speakers) in ARC and run ARC again. ARC does not automatically match levels of corrected and uncorrected segments of the frequency range. It can't . . . the uncorrected response varies with frequency, so exactly where the relative levels should be is a matter of taste.

If the above is already obvious to everyone, my apologies. This has been an oft-overlooked point with DSP systems I've used over a long time, and I hope this post helps some who find bass-only correction disturbs tonal balance.
 

·
Registered
Main system: Preamp STR + A23 + Tannoy Cheviot
Joined
·
58 Posts
For those who run ARC non-full-range and find the sound is too bright or dull, the issue may be levels (as set in ARC), as discussed in this post of mine:

If the target curve doesn't match the average response above the max correction frequency, the level imbalance can make your sound too bright or dull. And even if it sort of matches, a little adjustment might sound better to your ears. The way to fix this is to change the level of that speaker (or speakers) in ARC and run ARC again. ARC does not automatically match levels of corrected and uncorrected segments of the frequency range. It can't . . . the uncorrected response varies with frequency, so exactly where the relative levels should be is a matter of taste.
Dear Mike in NC,
I am sorry but I am not sure to properly understand what you wrote above. Could you explain a bit more?
Thanks,
 

·
Registered
Joined
·
719 Posts
I think he is saying to increase the trim level of the speaker(s) which sound dull. IMO, this likely helps as to me, with ARC ON vs OFF, there is a perceived loudness difference (OFF being louder). On the flip side, even if limiting ARC correction frequency, it is my understanding it still does something to frequencies above this (I recall reading this somewhere and was according to Anthem).
 

·
Registered
Main system: Preamp STR + A23 + Tannoy Cheviot
Joined
·
58 Posts
I think he is saying to increase the trim level of the speaker(s) which sound dull. IMO, th
I think he is saying to increase the trim level of the speaker(s) which sound dull. IMO, this likely helps as to me, with ARC ON vs OFF, there is a perceived loudness difference (OFF being louder). On the flip side, even if limiting ARC correction frequency, it is my understanding it still does something to frequencies above this (I recall reading this somewhere and was according to Anthem).
Increasing the trim level in the "Set Speaker levels" of Genesis ARC? Then in that case, it's to be done after measurement. Is it what is written in the doc: "This screen can be useful, if for example, after running the full range of ARC corrections you are noticing a slight imbalance in your channels which is causing the stereo image to shift to the right. To correct this, you would increase the left channel’s level and upload the corrections again."?
 

·
Registered
Joined
·
719 Posts
I think what you posted is addressing a similar, but different, issue. Change level in ARC & reload to correct channels out of balance (eg one is louder than the other). So, perhaps, it also applies to compensating for overall "dullness" in the highs with ARC On. Guess that makes me wonder, if ARC sets levels to match at 75, why there would be a channel imbalance? Similarly, if boosting selected speaker(s) to correct for high frequency "dullness", wouldn't that "unlevel" match those speakers with other (eg, default is 75 db for all, move fronts up by 1 db, and they'd be slightly louder than sub / surrounds /center, etc.?
 

·
Registered
Main system: Preamp STR + A23 + Tannoy Cheviot
Joined
·
58 Posts
I think what you posted is addressing a similar, but different, issue. Change level in ARC & reload to correct channels out of balance (eg one is louder than the other). So, perhaps, it also applies to compensating for overall "dullness" in the highs with ARC On. Guess that makes me wonder, if ARC sets levels to match at 75, why there would be a channel imbalance? Similarly, if boosting selected speaker(s) to correct for high frequency "dullness", wouldn't that "unlevel" match those speakers with other (eg, default is 75 db for all, move fronts up by 1 db, and they'd be slightly louder than sub / surrounds /center, etc.?
This is kind of "black box" for me. Months ago I performed measurements using my Umik + REW to check ARC (on-off) behavior.
I found out that the correction curves (left & right) were almost exactly the same than the equalization curves from Genesis Curve viewer EXCEPT that my right channel appeared on REW results to have been lowered of 2.5dB on the whole bandwidth... Effectively my right channel was playing lower than the left one with ARC engaged.
However, and that's strange: Using the STR balance I had to apply only +1dB (not +2.5dB) on the right channel to get the sound centered with ARC on.
Really unclear for me. I am till using this one dB correction and not keen to perform another measurement sequence without understanding. I think that Genesis has lowered the level on the whole bandwidth of the right channel to be able to apply the full 12db (+/-6dB) correction. But I cannot see that anywhere in the parameters & PDF report.
 

·
Premium Member
Joined
·
452 Posts
One thing that I have don't like about ARC is that it sets the levels of the various speakers based on the first measurement position. If that 1st position is at a chair, my chair on the left side of the centerline of the room, then the speakers on the right side, front and surrounds, will be boosted to compensate for the greater distance. This means for my wife's chair to the right of the room's centerline, those speakers are "hot". So, I after ARC is done, I symmetrize the trim levels for the left and right side speakers and check levels with an SPL meter. I try and avoid taking the 1st measurement on the room's centerline as nobody sits there and there can be room nodes there. (I learned that the hard way.)
 

·
Registered
Joined
·
719 Posts
This is kind of "black box" for me. Months ago I performed measurements using my Umik + REW to check ARC (on-off) behavior.
I found out that the correction curves (left & right) were almost exactly the same than the equalization curves from Genesis Curve viewer EXCEPT that my right channel appeared on REW results to have been lowered of 2.5dB on the whole bandwidth... Effectively my right channel was playing lower than the left one with ARC engaged.
However, and that's strange: Using the STR balance I had to apply only +1dB (not +2.5dB) on the right channel to get the sound centered with ARC on.
Really unclear for me. I am till using this one dB correction and not keen to perform another measurement sequence without understanding. I think that Genesis has lowered the level on the whole bandwidth of the right channel to be able to apply the full 12db (+/-6dB) correction. But I cannot see that anywhere in the parameters & PDF report.
Interesting. I'm planning on redoing my measurements (looking at adding a 3rd sub for movies only) in the near future. I'll need to remember to remeasure the fronts beforehand to see if I have similar behavior (wouldn't surprise me as on alot of occasions, seems like one channel is lower than the other).
 

·
Registered
Joined
·
298 Posts
Here is an illustration from ARC Genesis. Using the default gain (trim) for each speaker, the "Adjust Targets" screen in one of my files looks like this:
3078589

In the above, if the response is corrected up to 2 kHz only, there will be a rather laid-back sound, because the measurements above 2 kHz lie below the target. So frequencies < 2 kHz will be corrected to the target, and above, response will be less than the target.

If instead the gain of the front channels is increased by 1 dB in the "Adjust Speaker Levels" screen, the same graph will look like this:
3078590


That will lead to a brighter, more open sound, because the measurements above 2 kHz are more or less equal to the target curve.

My point is a simple but subtle one: relative levels of the target curve, compared to measured response in the uncorrected area, determine overall warmth or brightness of the sound. I suspect that this may be why ARC seems to change the sound above the max correction frequency, when the judgement is made by ear. Even if ARC is not doing anything unexpected, balance between bass and treble can change, deciding on the speaker levels set by ARC. (Levels are automatically set, apparently, to lie on average at the target level, but unless the speaker is really flat, you have to look at the graph before doing a partial correction.)

Is that helpful?
 
  • Like
Reactions: madrac

·
Registered
Joined
·
298 Posts
Increasing the trim level in the "Set Speaker levels" of Genesis ARC? Then in that case, it's to be done after measurement. Is it what is written in the doc: "This screen can be useful, if for example, after running the full range of ARC corrections you are noticing a slight imbalance in your channels which is causing the stereo image to shift to the right. To correct this, you would increase the left channel’s level and upload the corrections again."?
Yes, make the adjustment after measurement, after looking at the graphs, but before letting ARC make the corrections.

As to R-L imbalance, IME sometimes an imbalance in the bass (say due to room effects) can cause ARC to set, say, the R speaker hotter than the L. Because the bass is going to be corrected anyway, in that case, one should set ARC to have the same level (trim) for both main speakers before proceeding. If even then there's an audible imbalance with corrections loaded, the measurements may have not been symmetrical, or possibly the speakers are not toed in or tilted the same, or maybe the room is quite different R to L. It is possible for a matched pair of speakers not to match in the treble, but IMO it's uncommon.
 

·
Registered
Joined
·
719 Posts
thanks,Mike. your post with graphs was very helpful.
 
  • Like
Reactions: Mike in NC

·
Registered
Main system: Preamp STR + A23 + Tannoy Cheviot
Joined
·
58 Posts
Thanks Mike and Madrac, I think I am starting to understand... I am not that young and my brain has started to decline :-(
Effectively my right channel has been measured with a gain of roughly 2dB more than the left channel for frequencies above 1Khz. My room is completely asymmetrical. I think that the sound from the left speaker is bouncing on the right side wall+window and that the sound from my left speaker is absorbed by the couch on the left wall... Difficult to be worse than my configuration.

3078757


So Mike, from what you wrote I understand that ARC has readjusted (decreased) the global level of the right channel to match to the target curve.
However and that's the origin of my concerns: This is not visible anywhere.
Should you look at the green curve below, it is centered at 0dB and not at -2dB.

3078766



By using REW, I see a -2.7dB shift (purple curve in the graph below) and the consequence is an unbalanced sound with the left speaker playing louder that the right one. I compensate by using the balance function of the STR by adding only 1dB and not 2.7dB as we could have thought.

3078764



So I understand that the signal level is readjusted by ARC before calculating the correction curve and that explains why this correction curve is centered at 0dB.

My question: Knowing that I cannot remove the windows or the couch, what should I have done? As written in the doc, i.e.: Adjust the speaker level with one more dB as mike shown above and redo the calculation and upload into the STR. No need to re-perform signal acquisition with the microphone since the room has not changed.
FYI, my speakers are Tannoy Cheviot.
Thanks a lot!
 

Attachments

·
Registered
Joined
·
298 Posts
Hi @rorosone -- I am not sure to what frequency you are correcting. I can think of two things you could try within ARC to reduce the channel imbalance.

- Try correcting full-range. ARC and other modern systems are better at that than older systems were. You might like the result, and it should correct the channel imbalance. It also would remove the bump around 7 kHz -- you might like it gone, you might not, or you might be able to reduce it (if you wanted) by adjusting toe-in instead.

- If you want to correct only to 2 kHz (as an example), increase the speaker level of the blue speaker* until it coincides with (overlays) the red curve above 2kHz as much as possible, then let ARC re-calculate the corrections. If that sound proves a little bright, decrease the red curve by 0.5 dB before adjusting the blue one, then let ARC re-calculate. I agree, there is no need to re-measure. Just save the file with a different file name, so it will be easy to go back and forth.

Let us know how that works for you!

* That's done in ARC, in the (3) Adjust ARC Settings > Set Speaker Levels screen.

P.S. I think your brain is fine. But after all, I'm 72 myself. If you were to move away from beautiful Paris, then I would wonder about your mental acuity.
 

·
Registered
Main system: Preamp STR + A23 + Tannoy Cheviot
Joined
·
58 Posts
Thanks Mike :)
Yes Paris is a wonderful city.
I have 3 different profiles with different maximum correction frequencies, 500Hz, 1KHz, 5KHz. The 1KHz gives better results.
The bump after 7KHz doesn't exist, or at least, is not "captured" by my Umik microphone. I think that over 5KHz the microphone is too directive.
I will do what you suggest above and let you know.
Have all a nice weekend
 

·
Registered
Joined
·
23 Posts
Hello,

I have an Anthem AVM 60 processor and I have a 5.1.2 setup. Due to my cinema room not being the biggest, I have to have my couch right up against the back wall, which is why I only have 5 bed layer speakers.

When running ARC genesis, the lowest amount of test positions you can test for using the mic is 5, the problem is, position 4 and 5 appear to be behind the sofa/couch according to the illustration, but with my setup I can't place the mic behind the couch as it's right up against the rear wall.

Would you be able to advise me what would be the best solution? Would of been good if there was an option for just 3 positions which would eliminate having to put the mic behind the seats for position 4 and 5. Does anyone else here have 5 speakers with their seats right up against the rear wall? Shall I just (if I remember correctly) leave the mic in position 3 when asked for position 4 and leave the mic in position 2 when asked for position 5, i.e front left position used again for the rear left position and front right position used again for the rear right position?
 
2681 - 2700 of 2702 Posts
Top