Two Subs...Gain Matching vs Level Matching - Page 2

I tried gain matching my dual MFW-15s (one has a V1 amp and the other has a V2) last night. I've ran them level matched for over two years.

Does it really matter if the subs are in the middle of the room? I would think what would matter is that they are both in exactly in the same place when adjusting the gain?

Here is the method I used:

1) Placed two pieces of intersecting masking tape on the floor at one corner of each sub so I would know their exact "home" positions.
2) Shut one of the subs off.
3) With an SPL meter, made note of the db level of the other sub.
4) Unhooked both subs and moved the second sub into the exact same position where the first sub was.
5) Turned on only the second sub.
6) With an SPL meter, set the gain knob of the second sub to match that of the first sub.
7) Placed the subs back to their home positions, made the needed connections, and turned them back on.

What ended up happening was that I ended up simply turning the gain down on the second sub. Apparently, it had to work harder in the level matching scenario because I had to run its gain higher.

I have an Onkyo 805 receiver and didn't have time to re-run Audyssey, so I tested with it both on and off.

I run my subs about 5db hot. I tested some of my favorite scenes with my volume level at ref -5. So essentially the subs were at reference since I run them 5db hot.

I tried WOTW pods emerging, FOTP crash scene, The Haunting pipes scene, the famous Pulse scene (only went ref -7 with that one), etc.

Wow, I can't believe such a small tweak made a difference. As was mentioned, with level matching, the one sub had to work harder. This used to make it cry out during some of those scenes while the other sub was fine. Now the subs handled those scenes fine. I didn't hear any flapping or bottoming out.

I had never even heard of gain matching until I saw this thread. I had always thought that level matching was the only way to go. But really this method makes more sense (in most situations) because it lets each sub work equally as hard.

I'll have to spend a little more time with it, but at this point I'm sold.

Chad - You have Audyssey MultEQ XT or are you running Pro? Anyway, Audyssey doesn't like anymore than -3 to +3 adjustment on the receiver for the sub channel. You negate what Audyssey does when you go lower or higher the then 3 db. Also, never adjust the gain on the sub once Audyssey has been run. At this point, rerun Audyssey and see how it really sounds.

Can't you just be consistent...in other words....if you set the spl meter 4" from the sub, half-way between the middle of the driver and the middle of the port (if ported, like mine) then do the same for the other sub? Why do you move the sub to the middle of the floor?
Not that I'm arguing with it...just want to know why you should do it this way. Doesn't it mess stuff up once you move one sub to say a triple boundry position, then move the other sub to maybe a double boundry position?

When you gain-match, you want to measure the raw output of the subwoofer, with as little impact from the room as possible. That is why it is recommended to move the sub to the middle of the room... it reduces the levels of the reflections from the room boundaries.

Once you've gain-matched them, and then replaced the subs into their in-room positions, they will still be driving the same amount of energy into the room, they will still have the same headroom, the same point where they go into compression and distortion, etc.

They may measure differently at the (arbitrary) primary listening position due to the room reinforcement from the boundary reflections, but they will be driving the same electro-mechanical energy into the space. That is the whole point of gain-matching.

Craig

Audyssey XT. Like you said, Audyssey needs to be re-run, but I didn't have time to do that, so I did most testing last night with it turned off.

I have the gain knobs on the subs set such that when they are 5db hot my receiver reflects a "0" setting on the sub channel trim. So I have quite a bit of of adjustment range if I want to play around with the levels. I know I probably shouldn't run it hot like that, but I do it anyway.

For me, the biggest advantage of gain matching is that one of the subs is no longer working harder. If I end up with other benefits that will just be icing on the cake.

BTW, when you say -3/+3 do you mean as far as deviation from the level that Audyssey sets the sub level after running the calibration? Or do you mean that there is something wrong with running the actual setting less than or greater to -3/+3 as far as deviation from "0"? It's been quite awhile since I've run Audyssey, but I have always tweaked my channel trims with an SPL meter after running it. Maybe that's the wrong thing to do, but I've always done it that way. I have a very early 805 and there has been talk that the mic included with the early 805s wasn't as good as the later ones. Some folks have been buying new versions of the exact same mic and are finding that they are having to tweak their channel trims less after running Audyssey. I probably ought to order a new mic and see how it works out.

That is why the middle of the room is where you should move the subs to place them away from any boundry positions. Even at 4" you would get inconsistent readings from the boundries. This way they are both gained matched without any interaction with the room. I guess if you wanted to be really anal about it you would do it outside. But we can't count on the weather when a client wants their HT rooms setup.

So now that they are gained matched you return them to their respectful double or triple boundry corners or wherever you determined where they sound the best. This is when you run the Auto calibration tool in your AVR (Audyssey, YPAO, etc). This will then adjust for the summing of the subs and their position in the room.

Edit: ^^ I guess Craig types faster then me. lol

Craig:

Would you agree that as long as the subs are placed in the exact same place when setting the gain, it doesn't really matter where that position is as long as it is the exact same position for each sub? Like with my MFW-15s they are the exact same sub and whether placed in a corner or the center of the room, they are still going to be affected by the room to the same degree when in the exact same position.

I would agree... if you can be certain that the room effect is the same for both subs, that is proably "good enough". I put tape on the floor around the edges of the 1st sub to ensure I get the 2nd sub as close to the same position as possible.

Having said that, *I* would still move them to the middle of the room. And I still use the tape. But that's just me being my usual obsessive/compulsive self.

Craig

The only subject I ever flunked was typing. If I type faster than you... well, at least great minds think alike!

Craig

My typing was so bad I got a D for skill and an F for attitude. Really, true story on High School grade card.

On a side note I was really disappointed I missed out on meeting you when you where in my neck of the woods for CEDIA. Maybe in the future we can try again.

I resemble that remark! I got an F for both. Only subject I ever flunked. I'm currently typing with 2 fingers!

Of course, when I went to high school, the typewriters were manual. You had to *bang* on the keys to get them to work... and there was no spell-checker or grammar-checker. You were graded on the words-per-minute minus the number of errors. I had ERRORS.


I was disappointed we missed that opportunity too. I just had so much going on in such a short time, as I'm sure you did. CEDIA is such a huge show, so many vendors, so many demo's, I can't even remember everything I saw or heard. I can honestly say I've never been to a bigger venue, or a more overwhelming "expo" than CEDIA.

I will be in the Atlanta area next Feb. for a meeting for my "real" job. I will see if I can get some free time to catch up with you. I'll PM you ahead of time.

Craig

Sounds like fun. Be sure and do that. I thought AVS was our real job.

can someone tell me how to properly set the gain on both the sub and amp? what does gain do?

Thank you both Craig (yet again) and htgeek.....I get it. I wish I didn't.
The thought of moving those 2 monsters yet again....well, let's just say it's an adventure.

But that is what this hobby is all about. Doing everything possible to get every last once of performance out of any speaker combination. If you want more help PM me and we can talk on the phone. In your room (60 year old house) have you considered Auralex Gramma riser or similar sub isolation riser?

I've read through most of the posts, but just want to concretely understand it all. A little confusing for half-wit like me.

1. Connect one sub [footnote 1] to the receiver and place it in the middle of the room, stick spl meter 4" away from the center of the woofer.
2. Use receiver test tone [footnote 2] to measure the spl to 75dB, adjust the gain on the amp accordingly to reach this 75dB.
3. Disconnect the first sub, repeat step1 and step2.
4. Place the subs to their home destination.
5. Run auto calibration tools such as Audyssey.

[footnote 1]: How to place mic if the sub in question is Submersive or JL F212.
[footnote 2]: The gain on the receiver must be at [0], corrrect?

I shouldn't admit this, but I level matched earlier. I'm gain matching tonight (just got it in). You know who i'm talking to.

Thanks....appreciate the offer.
I do have pad (eD) for my 350...I need to get one for my 450. That is another scary thought....getting a pad under an A7-450.

I place my SPL 2 feet away, on a tripod about 2 feet off the floor. This will work with all types of subs whether ported, sealed or dual opposed subs. NOTE: With dual opposed subs it would be setup where the sub has the drivers not facing the SPL meter.


What I do is run Audyssey or YPAO first without any sub on. This will set the 5 or 7 speakers to reference [0] first. Then place sub in middle of the room, turn on the pink noise, select the left channel (because this is usually the first speaker to play the pink noise), now turn up the master volume on the AVR until the left speaker is outputting 75dB, turn on the sub and switch the pink nosie so it is outputing the Sub channel, in the AVR, and adjust your sub volume knob until it reads 75dB.


correctamundo for steps 3 - 5

Hopefully Craig will be by today and add anything I might have missed or correct me where necessary.

For my Submersives, I place the SPL meter 1 ft. away from the middle of the "front" of the sub, (the front being the side with no driver.) For the F212, I would probably go for the point between the drivers.


That's were I start, but see further detail below.


Interesting technique. I don't bother with the speaker levels at the point where I'm gain-matching the subs. Do you feel there is benefit in that? If so, I'll try it. I should be receiving my Integra DHC 80.2 on Monday. I'll need to completely redo my setup to integrate Audyssey MultEQ XT32, so that will be a good time to experiment.

There is one thing I do differently than you guys and, that is I set the subs higher than 75 dB. When you measure in the nearfield, the measurement will be higher than when the sub is further away. (Outdoors, you lose 6 dB for every doubling of distance. Indoors, it's more like 3 dB per doubling of distance due to room reinforcement.) If your 1st measurement is at 1 ft., then at 2 ft. you'll be down 3 dB, at 4 ft. down 6 dB, 8 ft. down 12 dB., etc. So, if you set them at 75 dB in the nearfield, they'll be down significantly when they're deployed in the room and measured at the listening position. To get them back to Reference Level, Audyssey would need to boost the trim level. This opens the possibility to overdrive the sub amp's input and cause distortion.

I have found that setting them to 85 dB nearfield, (1 ft.), is a good starting point. Then, when they're placed back in their in-room locations, each individual sub is below 75 dB. However, once combined, they are back to close to 75 dB. I then run Audyssey and it calibrates the combined output of the subs to the speakers at Reference Level. I usually end up with a slightly negative trim level, (-3 to -5), which I believe is better than a positive trim level, (boost).

Anyway, I'll play around with HTgeek's technique next week when I get my 80.2. I'll report back with the results.

Craig

I do believe there is a benefit. First I will now know that all of the speakers are leveled-match to the room and are set at reference level. My technique is probably left over from doing it old school before we had these new-fangle techno thingys.

Now granted the Auto Cal will set the speakers as large or full band because it will not detect a sub but again this is really only done to have all speakers set to [0] reference, and I believe, it gives Audyssey and YPAO a head start when it comes to properly intergrating subwoofers into the mix, which they can have trouble doing sometimes. If Audyssey has worked properly and you play the pink noise out of the left speaker, then when you turn up the master volume, so that it reads 75dB from the left speaker, then the MV should be at 0.

Craig John:
Now I use 75dB as an example only because that is the standard level the experts use. It is not an exact science. I actually use 80dB on all of my calibration and intergration jobs only because it is easier to read the analog SPL meter when the needle is straight up and down sitting on 0 on the gauge and in a sense helps to compensate for the mic being 2 feet away, just like you use 85 at 1 foot.

I hope that I have not confused all of you now. I have done this so many times that I usually do not even think about how I do it. All Craig an I are stating is we find it beneficial to gain-match the subs volume control knob prior to placing the subs in their final resting place within your room.

Glad to hear that GAIN matching seems to yield the best results for many.

I too have dual subs, but here are my concerns with gain matching:

(1) I like the idea of ensuring not to bottom out one sub quicker than the other when gain matching. OTOH, I also think it is possible to have situations where one may have a lot of sub-woofer headroom (where both subs are not pushed to their limits), and the subs will never reach their max. A user can also have a sub (like those made by Epik), which almost never bottom out.

(2) My other concern with having each sub having a different SPL output at the LP is how Audyssey would calibrate them if a single LFE out is used with a y-splitter. I would think that maybe more calibration would be applied to the sub with greatest room reinforcement.

I actually have not had the time to try to GAIN match my subs.....so I should retract my earlier post until I can experience it for myself.

I would love to hear peoples feedback on my above comments.

i dont get the whole gain match thing? i level match the twins and i seem to be fine. They are equal distance to MLP. Also Epik towers so i dont think i have any issues.

Thanks Hometheatergeek + Craig John for the detail explanation. I am able to soak in the concepts. I am going to print it for future reference.

Last question. Why is this gain method rather than level-match method? It seems like a level-match since we're using spl meter to determine the output. I was lead to understand that the gain method is just basically dialing the amp gain to certain point, say midpoint 12 O'clock [given identical subs].

If you are using identical subs, and the amps have exactly the same gain structure, (no variation between amps), and the drivers have exactly the same sensitivity, (no variation between drivers), then setting the amp gain to the same point on each sub amp will result in equal SPL's, (when measured using the same mic, in the same position, and the subs placed in the same position in the nearfield.) For most folks with identical subs, simply setting the sub amp's gain control to the same point is close enough to say they gain matched.

Measuring the gain structure is for us O/C folks who want to ensure they account for any variability in the amps and drivers.

Craig

Since your subs are equidistant to the LP, it's likely that gain-matching and level matching would result in very similar settings, especially if the room reinforcement is similar for both subs, (i.e., they're both corner-loaded for example.)

Gain-matching is more important for those of us whose multiple subs are not equidistant to the LP, and/or where the subs get dissimilar room reinforcement. Gain-matching ensures that the electro-mechanical output of each sub is the same, and that they are driving the same energy into the room, no matter where the output is measured, or what room reinforcement they see. This ensures that no one sub hits it's limits, (compression, distortion, limiting, bottoming, etc.), before the other(s).

Also, this is more important in systems where the desired SPL's are closer to the limits of the sub(s). If you're playing your system well below the limits of all your subs, then you should have no issues with either level-matching or gain-matching.

Craig

Two fingers; eyes on the keyboard... honest!

So what are you trying to say Redskin? I am trying to get better. Really I am.