Home Theater Forum and Systems banner

1 - 20 of 48 Posts

·
Registered
Joined
·
41 Posts
Discussion Starter · #1 ·
Hello everyone, I have a 7.1 system with 4 SW one in each corner, with DIY Acoustic treatments on walls and corners. I ran Audissey EQ XT set up in my HT at my LP only and also at different locations, and while the sound is better after running it, REW shows there are some peaks at around 30 and 70 Hz I want to get rid off because these make my room boomy especially in the back row when watching a movie. Does anyone know if nanoAVR or any other miniDSP will do anything to improve what AEQXT has already produced?
Thanks!
 

·
Banned
Joined
·
4,838 Posts
Hello everyone, I have a 7.1 system with 4 SW one in each corner, with DIY Acoustic treatments on walls and corners. I ran Audissey EQ XT set up in my HT at my LP only and also at different locations, and while the sound is better after running it, REW shows there are some peaks at around 30 and 70 Hz I want to get rid off because these make my room boomy especially in the back row when watching a movie. Does anyone know if nanoAVR or any other miniDSP will do anything to improve what AEQXT has already produced?
Thanks!
Some Audyssey users have found it worthwhile to "pre-EQ" one or two of their worst room resonances with a parametric EQ device like those from miniDSP or elsewhere. This makes the job easier for Audyssey and can give a smoother result. Read the Audyssey section of THIS REVIEW for some detail.
 

·
Registered
Joined
·
512 Posts
Hello everyone, I have a 7.1 system with 4 SW one in each corner, with DIY Acoustic treatments on walls and corners. I ran Audissey EQ XT set up in my HT at my LP only and also at different locations, and while the sound is better after running it, REW shows there are some peaks at around 30 and 70 Hz I want to get rid off because these make my room boomy especially in the back row when watching a movie. Does anyone know if nanoAVR or any other miniDSP will do anything to improve what AEQXT has already produced?
Thanks!
I realize there are those who advocate a single position Audyssey cal, but that's not recommended by the developers. If you want to "weight" Audyssey towards a single seat, try doing 8 points clustered in a virtual bubble around that position and give Audyssey something more to chew on. Move the mic around at least a foot, then see what you get. Measurements that differ spatially are important, and more is better.

Any linear FFT based measurement has most of its data points above 1KHz, so to gain enough resolution at low frequencies you need a lot of points through out the band. The problem with a single measurement position is that it doesn't represent the general sound of the system well, even though it's detailed. If you take multiple measurements over a space and simply average them, you actually reduce resolution, not improve it, though the graphs become easier to look at.

Audyssey came up with a way to reduce the amount of collected data while retaining required resolution. The idea of fuzzy clustering is to find common characteristics between measurements, and ignore the radically odd ones, thus improving the overall resolution, particularly at the low end.

More here, and here.

Other EQ systems that involve differing measurement techniques, and even manual adjustment input can perhaps produces flatter curves under specific circumstances, but they may or may not actually sound better that way. There is often a problem trying to verify one technique with another. I've found that using REW to attempt to verify Audyssey isn't reliable, and only useful generally. Fuzzmeasure seems to be much more similar to Audyssey, and seems to correlate well, though it's difficult to get the mic in exactly the same positions in space.
 

·
Registered
Joined
·
512 Posts
Some Audyssey users have found it worthwhile to "pre-EQ" one or two of their worst room resonances with a parametric EQ device like those from miniDSP or elsewhere. This makes the job easier for Audyssey and can give a smoother result. Read the Audyssey section of THIS REVIEW for some detail.
Agreed, this can work. Two cautions: any time EQ requires a gain filter you have to be cautious about headroom in the system. For that reason, Audyssey limits the gain in any filter to 9dB. If it's doing that, and you go in with another device and add a few more dB of gain, you could be into a power amp headroom issue you'll never see on the final FR graph. The other caution is that there are differences in measurement methods, using REW to verify Audyssey isn't always accurate and reliable. To see this, you need the Audyssey Pro software and kit, which presents before/after graphs (though unscaled...grrr!), in which you'll see very different results that you get with REW. It depends on what you trust.

Let me say right here that I don't want to get into another REW/Audyssey debate, so anyone is free to disagree and I'm not going to argue. I believe there's much more going on in Audyssey's method of handling FFT data, and it has a ton of research behind it. REW is a fine tool, and there's nothing wrong with it other than how it's used by the operator. But I've never seen the two agree consistently, sometimes they do, sometimes they don't.

I used to use an external processor with Audyssey to work a bit more on the LFE, but since I now run XT32 with SubEQ, the only LFE processing I now need and use is an excursion limiter. This version of Audyssey nailed it for both subs.
 

·
Banned
Joined
·
4,838 Posts
Agreed... any time EQ requires a gain filter you have to be cautious about headroom in the system. For that reason, Audyssey limits the gain in any filter to 9dB. If it's doing that, and you go in with another device and add a few more dB of gain, you could be into a power amp headroom issue you'll never see on the final FR graph.
Thanks for the additional detail. I should have included that the technique is best used to reduce large low-frequency peaks before running Audyssey.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #6 ·
Some Audyssey users have found it worthwhile to "pre-EQ" one or two of their worst room resonances with a parametric EQ device like those from miniDSP or elsewhere. This makes the job easier for Audyssey and can give a smoother result. Read the Audyssey section of THIS REVIEW for some detail.
Thanks for the response AudiocRaver, an for the link too, i will check it out!
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #7 ·
Gazoink,
Thanks for the detailed response. I tried to reply to your post but I was caught by the spam filter, so I will just reply here, I had never heard about or though about doing the cluster measurements, I will do that and get back with any comments.
Also, the only way I have run REW is via UAC222/RS DSPL connected to the main subwoofers in front of the room (balanced XLR). do you know if I can run REW via the processor without an EQ and measure response from all 4 subs in the room? If so, I would appreciate any guidance on how to connect.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #8 ·
Agreed, this can work. Two cautions: any time EQ requires a gain filter you have to be cautious about headroom in the system. For that reason, Audyssey limits the gain in any filter to 9dB. If it's doing that, and you go in with another device and add a few more dB of gain, you could be into a power amp headroom issue you'll never see on the final FR graph. The other caution is that there are differences in measurement methods, using REW to verify Audyssey isn't always accurate and reliable. To see this, you need the Audyssey Pro software and kit, which presents before/after graphs (though unscaled...grrr!), in which you'll see very different results that you get with REW. It depends on what you trust. Let me say right here that I don't want to get into another REW/Audyssey debate, so anyone is free to disagree and I'm not going to argue. I believe there's much more going on in Audyssey's method of handling FFT data, and it has a ton of research behind it. REW is a fine tool, and there's nothing wrong with it other than how it's used by the operator. But I've never seen the two agree consistently, sometimes they do, sometimes they don't. I used to use an external processor with Audyssey to work a bit more on the LFE, but since I now run XT32 with SubEQ, the only LFE processing I now need and use is an excursion limiter. This version of Audyssey nailed it for both subs.
what I was doing was the other way around, using the filter results from REW and update on Audissey's via manual EQ, unfortunately, the graph only allows to adjust 63hz and above, so my LF peaks will remain, plus now that you cautioned me on the headroom issues, it makes me think I may have gone beyond its limit , because after I adjusted the gain at 63hz by a few db's, room sounded awful.
 

·
Registered
Joined
·
512 Posts
Gazoink,
Thanks for the detailed response. I tried to reply to your post but I was caught by the spam filter, so I will just reply here, I had never heard about or though about doing the cluster measurements, I will do that and get back with any comments.
When you run Audyssey, the OSD tells you to move the mic to between 6 or 8 possible positions. Those are usually supposed to be at ear level in each seat. You can "weight" the calibration to one or two seats by making measurements in that bubble concept. Audyssey needs multiple measurements over a space to work well.
Also, the only way I have run REW is via UAC222/RS DSPL connected to the main subwoofers in front of the room (balanced XLR). do you know if I can run REW via the processor without an EQ and measure response from all 4 subs in the room? If so, I would appreciate any guidance on how to connect.
So you have 4 subs. You have a USC222 feeding two, but not the others?

Calibrating 4 subs is not trivial. Technically, you should get them to level-match at the LP, then calibrate them all with a single EQ, or, preferred, EQ and cal each one separately then integrate them together. It's not a small task!
 

·
Registered
Joined
·
512 Posts
what I was doing was the other way around, using the filter results from REW and update on Audissey's via manual EQ, unfortunately, the graph only allows to adjust 63hz and above, so my LF peaks will remain, plus now that you cautioned me on the headroom issues, it makes me think I may have gone beyond its limit , because after I adjusted the gain at 63hz by a few db's, room sounded awful.
What graph are we talking about?
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #11 ·
What graph are we talking about?
i meant to say the Manual Parametric EQ, not the graph, sorry, I was looking at my REW graph when I wrote that, but I will post my graph as soon as I reach my 5 posts since I just signed up here.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #12 ·
When you run Audyssey, the OSD tells you to move the mic to between 6 or 8 possible positions. Those are usually supposed to be at ear level in each seat. You can "weight" the calibration to one or two seats by making measurements in that bubble concept. Audyssey needs multiple measurements over a space to work well. So you have 4 subs. You have a USC222 feeding two, but not the others? Calibrating 4 subs is not trivial. Technically, you should get them to level-match at the LP, then calibrate them all with a single EQ, or, preferred, EQ and cal each one separately then integrate them together. It's not a small task!
yeah, sounds like I am going to have to start working on that, one by one and then integrate, but it sounds like I will ultimately have to go the nanoAVR route to EQ all four, which actually I wonder if nanoAVr or any other will do multi sub parametric EQ...
 

·
Banned
Joined
·
4,838 Posts
yeah, sounds like I am going to have to start working on that, one by one and then integrate, but it sounds like I will ultimately have to go the nanoAVR route to EQ all four, which actually I wonder if nanoAVr or any other will do multi sub parametric EQ...
I am reviewing the nanoAVR right now. The nanoAVR has 8 in and 8 out, plus a ninth internal channel for the sub/lfe channel processing, and any-to-any routing/mixing capability, but the only way to get signal in or out is via HDMI. You would have to dedicate the nanoAVR to sub processing running on a separate HDMI channel so 4 of its regular outputs could be used for sub signals, plus follow it with a dedicated AVR or preamp with preamp outputs that could feed subs... Kinda messy. The internals are all there, not the i/o you would want.

The 2x4 with the right plugin looks like a better fit.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #14 ·
I am reviewing the nanoAVR right now. The nanoAVR has 8 in and 8 out, plus a ninth internal channel for the sub/lfe channel processing, and any-to-any routing/mixing capability, but the only way to get signal in or out is via HDMI. You would have to dedicate the nanoAVR to sub processing running on a separate HDMI channel so 4 of its regular outputs could be used for sub signals, plus follow it with a dedicated AVR or preamp with preamp outputs that could feed subs... Kinda messy. The internals are all there, not the i/o you would want. The 2x4 with the right plugin looks like a better fit.
thanks for the feedback AudiocRaver, it looks like my best option is as you mentioned above.

Let me tell you a bit about my subwoofer set up :
SW1: HP+ Master plus Slave in front wall, and
SW2: In-wall coax splits at the end of back wall to feed two 10" subs, one on each back corner.
My front subs, have XLR connectors, my rear subs only have RCA (in-wall via coax, split in the back to feed each sub),
AV processor has both XLR and RCA options for SW1 and SW2 outputs.

So my first REW sweeps i connected the audio interface directly to front master SW, which fed the slave, so it only tested front subs. I have not run REW via AVR because i had no idea how to connect so that I test fir all 4 subs, but looks like now I know. I would prefer to test with only balanced signals if it makes a tangible difference. Based on this, which platform option do you recommend? FYI, I have asked this to miniDSP directly, but it will probably take a couple of days to hear back, so do you know if there is a 2x4 miniDSP for balanced/XLR connections?
Thanks again!
 

·
Banned
Joined
·
4,838 Posts
thanks for the feedback AudiocRaver, it looks like my best option is as you mentioned above.

Let me tell you a bit about my subwoofer set up :
SW1: HP+ Master plus Slave in front wall, and
SW2: In-wall coax splits at the end of back wall to feed two 10" subs, one on each back corner.
My front subs, have XLR connectors, my rear subs only have RCA (in-wall via coax, split in the back to feed each sub),
AV processor has both XLR and RCA options for SW1 and SW2 outputs.

So my first REW sweeps i connected the audio interface directly to front master SW, which fed the slave, so it only tested front subs. I have not run REW via AVR because i had no idea how to connect so that I test fir all 4 subs, but looks like now I know. I would prefer to test with only balanced signals if it makes a tangible difference. Based on this, which platform option do you recommend? FYI, I have asked this to miniDSP directly, but it will probably take a couple of days to hear back, so do you know if there is a 2x4 miniDSP for balanced/XLR connections?
Thanks again!
Looks like their Balanced 2x4 with their 4-Way Advanced plugin would do what you need. The balanced connections are via terminal blocks, and there is a sensitivity jumper for proper voltage levels,
 

·
Super Moderator
Joined
·
9,280 Posts
Calibrating 4 subs is not trivial. Technically, you should get them to level-match at the LP, then calibrate them all with a single EQ, or, preferred, EQ and cal each one separately then integrate them together. It's not a small task!
The experiences of numerous Forum members here suggests that the best way to EQ multiple subs is to level match and then equalize them all as a single entity, with a single set of filters. After all, combined output is all you hear from multiple subs, you don’t localize any of them so it doesn’t make any sense to equalize them separately. Typically the results of EQing each sub independently is that you get each one with an ideal curve, but then measuring them all together gets response that’s totally wacked.

What you might do is measure each independently to see of any of them has a gross issue (e.g. one has a huge peak that the others don’t) and correct that separately, but you probably shouldn't go much beyond that with individual EQ.

Regards,
Wayne
 

·
Registered
Joined
·
512 Posts
The experiences of numerous Forum members here suggests that the best way to EQ multiple subs is to level match and then equalize them all as a single entity, with a single set of filters. After all, combined output is all you hear from multiple subs, you don’t localize any of them so it doesn’t make any sense to equalize them separately. Typically the results of EQing each sub independently is that you get each one with an ideal curve, but then measuring them all together gets response that’s totally wacked.
Unless an individual sub's response is wacked already, at which time you need EQ on an individual basis.
What you might do is measure each independently to see of any of them has a gross issue (e.g. one has a huge peak that the others don’t) and correct that separately, but you probably shouldn't go much beyond that with individual EQ.

Regards,
Wayne
Yeah, that's what I meant.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #18 ·
Thank you guys, miniDSP 2x4 on order with 4 way mini plug. Hopefully it will get here soon. I will be posting any results ....or inquiries ...once I start the quest for the subwoofer best integration to my system and have something to share.
 

·
Registered
Joined
·
868 Posts
Looks like their Balanced 2x4 with their 4-Way Advanced plugin would do what you need. The balanced connections are via terminal blocks, and there is a sensitivity jumper for proper voltage levels,
So R2RO directed me here for an answer to my question I posted in another thread a few days ago. I am looking to get the miniDSP 2x4 box to EQ multiple subs (up to four subs). I want to have independent gain, phase (distance) and EQ bands for each sub. It is my understanding that more than one of the plugins will do it, but it sounds like the 4 way advanced will do the job, correct? Can you apply EQ filters to each output individually, and also to the entire set as a single unit? It looks like you can, based on what I read over at their website. I will be doing two subs initially, but would like to be able to expand to four. Thanks.
 

·
Registered
Joined
·
41 Posts
Discussion Starter · #20 ·
So R2RO directed me here for an answer to my question I posted in another thread a few days ago. I am looking to get the miniDSP 2x4 box to EQ multiple subs (up to four subs). I want to have independent gain, phase (distance) and EQ bands for each sub. It is my understanding that more than one of the plugins will do it, but it sounds like the 4 way advanced will do the job, correct? Can you apply EQ filters to each output individually, and also to the entire set as a single unit? It looks like you can, based on what I read over at their website. I will be doing two subs initially, but would like to be able to expand to four. Thanks.
don't know if you have looked at this article from the miniDSP community, but I skimmed through it the other day, not read it in detail yet, but the plug in exploration section may have some info that can be useful to anyone trying to connect multiple subwoofers in addition to the great guidance from HTS community, which in my opinion is priceless...
http://www.minidsp.com/support/community-powered-tutorials/201-dual-sub-integration
 
1 - 20 of 48 Posts
Top