TOPIC:

You ask questions without referring to what document you are talking about. How are we supposed to answer your questions without context?

The document that I am referring to is the only document in this thread.. the one you wrote. And my question was about page 4. I am not clear on the procedure of gain matching subwoofers. My understanding is:

place the mic near field and raise the volume until the SPL and the MLP is at ~75db... So you are using two SPL meters. Or I guess you can measure at the MLP and then when it hits ~75db, measure near field and then record your measurement and amp dials. Then, do the same for sub #2. Correct?

moves wrote:

The document that I am referring to is the only document in this thread.. the one you wrote. And my question was about page 4. I am not clear on the procedure of gain matching subwoofers. My understanding is:

place the mic near field and raise the volume until the SPL and the MLP is at ~75db... So you are using two SPL meters. Or I guess you can measure at the MLP and then when it hits ~75db, measure near field and then record your measurement and amp dials. Then, do the same for sub #2. Correct?

Thanks for clarifying. “Gain Matching” refers to the procedure to ensure multiple subs are outputting at the same level without the influence of where they are positioned in the listening room. To gain match, place the REW microphone in front of the first sub’s driver, oriented horizontally, with the mic tip 1” from the center of the dust cap. Generate a sub test tone to the first sub (I use the AVR speaker level-setting test tones with MV=0). Make sure the test tone is going to only the first sub. Measure the sub output using the REW SPL meter. Move the mic to the next sub, positioned the same as with the first sub. Output the same test tone only to the second sub. Adjust the gain control on the sub until it’s output matches the first sub’s output. Proceed until all of your subs are outputting at the same level. The subs are now gain matched.

The wrinkle in the process is that even though the subs are now gain-matched, the combined level at the MLP with all subs playing may be too high or too low. For example, when running a Dirac Live calibration, the combined sub output level must match the output level of the mains. If the sub output is too high or too low, the gain-matching procedure must be done all over again, with higher or lower gain settings. This is a back-and-forth process until the gain-matched subs are producing the same combined output level at the MLP that is appropriate for the Dirac calibration.

I find that a near-field sub level of ~90dB for each sub will result in close to a 75dB combined output level at the MLP. Different setups may require different sub gain settings. It’s trial-and-error.

10-4 that is much clearer. Thanks for taking the time. Excellent work on the document.

Thanks for clarifying. “Gain Matching” refers to the procedure to ensure multiple subs are outputting at the same level without the influence of where they are positioned in the listening room.

Have you established why that is desirable?

Even assuming you have matched subs, if the goal is to have a certain SPL at the main listening position (MLP), then simply turn off all the subs except one and measure its SPL at the MLP and set it to a lower dBSPL (~70dBSPL). Then turn it off and set the next sub to the same dBSPL. And so on for any additional subs. Then turn them all on and measure the combined output. If it's not what you want, repeat the process. But there's no reason to do a near-field SPL measurement first.

Also, if you're using an AVR for your bass processing, it's going to be able to set the level at least +/- 5dBSPL (and probably much more), so if you're between 70 - 80dbSPL, you're good. You can tweak the level from the AVR.

jburk wrote:

Thanks for clarifying. “Gain Matching” refers to the procedure to ensure multiple subs are outputting at the same level without the influence of where they are positioned in the listening room.

Have you established why that is desirable?

Even assuming you have matched subs, if the goal is to have a certain SPL at the main listening position (MLP), then simply turn off all the subs except one and measure its SPL at the MLP and set it to a lower dBSPL (~70dBSPL). Then turn it off and set the next sub to the same dBSPL. And so on for any additional subs. Then turn them all on and measure the combined output. If it's not what you want, repeat the process. But there's no reason to do a near-field SPL measurement first.

Also, if you're using an AVR for your bass processing, it's going to be able to set the level at least +/- 5dBSPL (and probably much more), so if you're between 70 - 80dbSPL, you're good. You can tweak the level from the AVR.

Was it ever confirmed what approach to follow for best results?

My confusion with in wall subs:

Scenario 1)
Set volume in AVR and do not change it
Turn one sub on in minidsp and get the SPL from the MLP
Turn off the sub and turn a different sub on. Match the SPL reading to the first sub. Proceed for all four subs.
Run room sweeps and add delays
With all subs on, run SPL and adjust miniDSP input to read 75db
EQ the total result to a flat 75 and run Audyssey

Scenario 2)
Set volume in AVR and do not change it
Turn sub one on and with an SPL reading, adjust it in minidsp to 75
Turn off sub one and turn each sub on and off and adjust all to 75db
Turn them all on and run an SPL check. If the reading is greater than 75, lower the miniDSP input gain to 75 (not changing the individual gains). Or, lower each sub below 75, match them all one at a time, and then get a combined reading of 75.
Run room sweeps and add delays
EQ the total result to a flat 75 and run Audyssey

Scenario 3)
Set volume in AVR and do not change it
Turn one sub on in minidsp and get the SPL from the MLP
Turn off the sub and turn a different sub on. Match the SPL reading to the first sub. Proceed for all four subs.
Run room sweeps and add delays
EQ the total result to a flat 75 and run Audyssey.
Audyssey will ask to lower the input gain in minidsp until it hits 75.

Is there a difference in how to approach it all with these three? I assume it matters when to EQ and how it synchs with audyssey.

Thanks

Each of your three approaches involves what is called “level matching”. When you measure a sub’s output by placing the mic in the MLP, the measurement is affected by the sub’s placement in the room. For example, measure two subs to 75dB at the MLP, and then look at the position of the gain setting on the sub (or the output level on the MiniDSP). The output setting is likely to be different, perhaps small, or perhaps significant, depending on the room’s influence on each sub. If you leave the sub’s level-matched, and then gradually output the sub channel output using a test tone, the sub with the highest level setting will “max out” (distort) before the other subs. This means that your sub channel maximum output level is limited by that one sub’s performance, and I find this undesirable.

Try gain-matching the sub’s. Place the tip of the measurement mic immediately in front of a sub’s dust cone and set its level while only this sub is playing a test tone. Repeat the process for all other subs, setting each sub’s output to the same level as the first sub. The level will be quite high, since the mic is so close. In my setup, each sub measures ~90dB nearfield, and this level produces the desired combined sub level at the MLP for input into the room correction calibration (typically 75dB). If not 75dB, repeat the gain-matching process (or adjust the input level on the 2x4) until you reach the desired 75dB level.

Gain-matching avoids the situation in which one sub can be “maxed out”, and allows an overall higher level of output on the bass channel until distortion is reached.

AustinJerry wrote:

Each of your three approaches involves what is called “level matching”. When you measure a sub’s output by placing the mic in the MLP, the measurement is affected by the sub’s placement in the room. For example, measure two subs to 75dB at the MLP, and then look at the position of the gain setting on the sub (or the output level on the MiniDSP). The output setting is likely to be different, perhaps small, or perhaps significant, depending on the room’s influence on each sub. If you leave the sub’s level-matched, and then gradually output the sub channel output using a test tone, the sub with the highest level setting will “max out” (distort) before the other subs. This means that your sub channel maximum output level is limited by that one sub’s performance, and I find this undesirable.

Try gain-matching the sub’s. Place the tip of the measurement mic immediately in front of a sub’s dust cone and set its level while only this sub is playing a test tone. Repeat the process for all other subs, setting each sub’s output to the same level as the first sub. The level will be quite high, since the mic is so close. In my setup, each sub measures ~90dB nearfield, and this level produces the desired combined sub level at the MLP for input into the room correction calibration (typically 75dB). If not 75dB, repeat the gain-matching process (or adjust the input level on the 2x4) until you reach the desired 75dB level.

Gain-matching avoids the situation in which one sub can be “maxed out”, and allows an overall higher level of output on the bass channel until distortion is reached.

Thanks Austin. Your guide was really helpful and detailed. Unfortunately in my scenario, I am limited to Level matching since they are in-wall subs covered up by acoustic fabric. No idea where the subs are at this point to precisely measure for proper gain matching.

Since they are all the same sub, would the ideal scenario then be to use the MLP and either:

Change the output of each sub in minidsp to read 75db on an SPL reading prior to doing any sweeps/delays (which would have a a combined reading greater than 75). At which point, use the input gain to drop the combined total to 75. Then proceed to running sweeps and delays and EQ to level match to 75.

Change the output of each sub in minidsp to read the gain of the furthest sub and then use the input gain with an SPL to drop them to 75db when combined. Then proceed o running sweeps and delays and EQ to level match to 75.

Do not change the output of each sub in minidsp. As they are all the same sub, just proceed with room sweeps and delays and EQ to level match to 75. Then, run an SPL check and reduce the input to 75db.

Do not change the output of each sub in minidsp. As they are all the same sub, just proceed with room sweeps and delays and EQ to level match to 75. Run Audyssey and when asked to lower to 75db, do so on the input level.

Does it matter which approach to take when having to resort to level matching? I would assume that each scenario would provide a different EQ and Audyssey result. Since Audyssey would be matching at reference level rather than the volume we set to in the AVR when running REW, how do we take that into consideration when picking a scenario from the above (or possibly a different one that is better than the ones mentioned)? If we are already lowering the input level to get to 75db at the AVR volume we set, wouldn't that limit the max outpain gain when Audyssey is using reference?

Wish I didnt cover up the subs prior to calibrating (using your awesome guide). The headache....haha

Well, I think placing the mic in front of the fabric would accomplish the gain-matching "close enough". Any other approach will run the risk of "maxing out" one of the subs, IMO.

AustinJerry wrote:

Well, I think placing the mic in front of the fabric would accomplish the gain-matching "close enough". Any other approach will run the risk of "maxing out" one of the subs, IMO.

Got it! So even for those, like me, with the entire room in acoustic fabric, you recommend just placing the mic where we think the subwoofer is (in my case, i placed them roughly in the center of each wall), and use your gain matching approach of furthest sub and matching each output gain to that sub. And then if the MLP is greater than 75, repeat the process over again by lowering the output of the furthest sub and matching them all to the new setting - until we hit 75? My concern is ballpark measurements trying to guess where the sub is in the wall, is going to be way off - and whether that outweighs the cons of adjusting each sub to 75 right at MLP (which may max out subs to your point).

One last point (sorry two) for clarification:

The guide mentions :
Open the Target Settings drop-down and click the “Set Target Level” link, which fills in the
“Target Level” box with the maximum level that was reached during the measurement sweep.

Would we not want to level match to a straight line at 75?
I assume that we need the AVR volume at 0 when doing all of this so that it is consistent with the gain matching test tones through the AVR (which uses 0)?

And lastly, I realized my friend posted this in another thread and you replied there, but he was not clear. In REW (not MiniDSP), under Preferences, there is a Check Levels button that many guides say we should click on before starting so that the sweeps are in the parameters that REW wants. Your guide leaves this part out since it does not cover REW in this one. But do we turn our MiniDSP outputs on that are connected to the subs, and then click 'Check Levels', or do we turn everything off? Want to make sure the sweeps are done accurately.

Can't thank you enough. Excited to have at it with another attempt this week!

Sorry, I may have missed in an earlier post the fact that your subwoofers are hidden behind a fabric wall, and you no longer can precisely locate where they are. In this case, placing the mic immediately in front of the sub to gain-match them is probably not feasible. Just curious, why do you have fabric covering the entire wall, and why did you not measure where the subs are placed before installing the fabric?

The alternative would be to level-match the sub’s.

- Place the mic in the MLP.
- Play the AVR level-setting test tone for the sub channel through one sub. (Remember that the AVR master volume should be at max, or zero, when playing the test tone.)
- Adjust the sub gain so that it measures 75dB.
- Repeat for each of the other subs.
- When finished, turn all subs on and play the same AVR level-setting test tone and measure output level at the MLP.
- If the combined output level is higher or lower than 75dB, repeat the individual sub level-setting with a higher, or lower, level s appropriate.
- Repeat the process until the combined sub level measures 75dB at the MLP.

The “set target level” instruction pertains to using REW to create PEQ filters for the 2x4. How did we suddenly start talking about the REW EQ tool?

As for setting the proper levels for REW sweeps, my REW guide covers this specifically. Have you read the guide?

@AustinJerry - FYI, the link to your guide is currently broken.

Fixed.

I can’t believe I didn’t stumble across this excellent guide earlier.

Gonna see how my old miniDSP plays with Anthem Room Correction,Manichaeism is sort of notorious for not measuring speaker distance.....and may not let me adjust subwoofer trim easily without iterating.