First Reflection Point Question

[nathan_h]

I’ve got a calibrated usb mic I can use with my iPad but there isn’t a program like REW for the iPad. Need a laptop for REW.

 

[Fastfwd]

Well, I did finally buy a MiniDSP mic. It came in today and I fiddled with taking measurements. Mostly just testing 2 channel full range and seeing what differences I would find with the various Audyssey settings. I'm not sure I've taken measurements that are worth examination for much more than my curiosity. I didn't set everything to a 'reference level' volume for instance. I did follow a few guides I watched on Youtube and from over at the Audio Science Forum for basic setup.

Now, my Audyssey should be setup currently to only correct below 800mhz. What initially sort of baffled me was how much different 'Reference' and 'Flat' were at the top end in particular. In just a little closer review it looks like the Dynamic EQ is responsible for a pretty significant bump to 'Flat' at the top end. 'Reference' and the 'Pure Direct' mode have a almost identical top end with Dynamic EQ turned off and that is probably what I expected to see with Audyssey set to only correct below 800mhz.

I don't mind throwing a few of these up if anyone cares to look at what I've taken so far. It doesn't look like anything to brag about to my eyes. There's a gigantic dip at 110mhz that Audyssey has a hard time trying to smooth out.

I have frankly taken up listening in Pure Direct mode much more recently. I finally noticed that Audyssey is converting my SACD disc DSD files to PCM 48mhz? Something like that - it was from an article on here that discusses it. I still use it with almost anything 16/44. It can get a bit bright with some material, but now I see that is probably the Dynamic EQ that is doing it.

I'll continue to research how to do this better. Take multiple samples around the listening position, etc. I was mostly just playing around with the various DSP settings I already use.

I merged Right and Left channels for these. I suppose to diagnose specific room acoustic treatment needs it would require separate R/L measurements and I did take those, but I thought it was interesting to see what was happening with and without Audyssey and the various tweaks within it.

I did do a few with the sub, but I was mostly curious about my full range 2 channel for now. I'm sure this will be handy to see how my sub is performing in my room too.

I measured 10hz to 22khz. Not sure I'm doing it 'right.'

L/R Average with Audyssey Reference and Dynamic EQ on:


L/R Average with Audyssey Flat and Dyanmic EQ on:


L/R Average in Pure Direct mode:


L/R Average with Audyssey Reference and Dynamic EQ OFF:

 

[Fastfwd]

So, I tried to get it close to 75db 'reference level' using the REW meter, pink noise generator, etc. I ran a few more with the mic on a tripod after I found my adapter for it to fit and sit at 90 degrees. Closer to a proper setup. Still no multiple measurements from around the main listening position. I've tinkered with toe in/out since I last ran Audyssey and I might settle on my toe in/out again and run Audyssey again before getting too serious about multiple measurements.

The dip at 110/112 seems a little bothersome. I wondered if I ran the mains full range and used the sub crossed over at 120hz if that would resolve the issue and it didn't do it. So, that must be the room I guess?

My highs really take a disappointing roll off after 14/15khz by about 4/5db to 20khz. I knew my Revels were supposed to roll off slightly, but I thought maybe more like 2db. It could just be that they aren't perfectly on axis. I did have them setup for extreme toe in and now they are back to the more mild toe in to just either side of me. I was questioning if vocals imaging was being hurt by the extreme toe in.

I usually listen in Flat with Dynamic EQ turned on. Depending on material I may add a 5db or 10db bump with Reference Level Offset. I took a few samples with that too, but I didn't create screenshots - mostly the same as these with EQ only.

I screwed around with the range on the display of these measurements. So, they aren't exactly apples to apples from one screenshot to the next.

If these help with any suggestions feel free to let me have them. I think you can do manual adjustments in the Denon for EQ beyond what Audyssey creates. I might look for that and see if I can manually smooth out that dip at 110hz and what else I may possibly be able to do once I get a more definite idea of what's happening.

Left and Right with Audyssey Flat and EQ on


Left and Right Flat EQ off


Left and Right Reference EQ On


Left and Right Reference EQ Off


Left and Right Flat Dynamic EQ with Sub Crossed at 120hz and 80hz


Left/Right Pure Direct at Reference Volume Level

 

[Fastfwd]

Here's what I guess I would call the flattest curve I could squeeze out of the various Audyssey settings. Still have a nasty dip at 110. There's a couple db bump up top, but it got my 20k to just about flat with the rest of the curve. I was a little concerned that I wasn't seeing any available settings getting me there and remaining anywhere near flat up to it.

I took a look at the EQ settings that it arrived at for Reference and Flat and I don't see evidence that it created the 110hz dip or attempted to correct it. At least not effectively. I did run Audyssey again. Twice. I tried using 6 mic positions instead of my usual 3. Still no dice.

Maybe this gets to the point where I should see if I can find more into on creating a custom Audyssey profile. I think I noticed that some people go that route.

Anyway, back to the original question - my left speaker does look to me like it struggles around 70/80hz where the right speaker does not. I originally had plans to put some room treatment behind that left speaker thinking that it just logically looked like it could use some help. There's a bare hard wall directly behind it and then a very hard wood door next to that and a bare wall corner after that. I'm not sure how much one panel would do or if I would even hear a difference frankly. Would it be worth the money, etc.

The 110hz dip bothers me, but as was pointed out in another forum I don't have that same dip in Pure Direct mode. So, apparently Audyssey is dropping the ball somehow. Multiple runs of it give the same result.

I was also a little surprised to figure out that the Denon app on my phone labels Reference Level Offset at having PLUS 5,10, and 15db. I think maybe that is supposed to be MINUS. My +15db measurements produced the 'flattest' curves while the 0db was the most boost. I guess that means those values subtract from the otherwise heavily boosted 'Reference' curve with Dynamic EQ applied.



Here's that flattest curve. I'm guessing that would be helpful to see what Audyssey can and can't do for me. The rest would require room treatment I suppose. This is mains full range. No sub. 2 channel only. The actual settings that got me there were Flat with Dynamic EQ on and Reference Level Offset cut by 15db. I took three measurements of each channel - center and 6" to each side. Merged those for these results. Dynamic EQ off with Flat was a close second flattest, but a slight roll off at 20k.

 

[nathan_h]

Looks interesting.

A few things come to mind:

1. Set Audyssey aside for now and get things as good as possible without Audyssey. Otherwise you are chasing your tail.

2. Yes, run one speaker at a time when measuring to see what's going on. Later, you will run sub(S) + one speaker.

3. It's okay if you want to measure full range for now, but ultimately you'll want to do bass management right, cross over to subs, and tackle room modes, so don't worry about things in the mains that look wonky below 100hz. Setting up the right crossover settings (not just frequency choice, but distance/phase, etc) will let you address that:

and

4. It wasn't clear to me where that dip is from the diagrams since most are not un eq-ed but it sounds like SBIR may be the culprit. Moving the speakers as close as possible to the front wall, placing thick acoustic treatment directly behind them, and re-measuring to confirm that solved it will tell you more.

4. I don't see a layout, diagram or picture of your room. That plus measurements would help other folks advise you on what else might be useful.

 

[Fastfwd]

Looks interesting.

A few things come to mind:

1. Set Audyssey aside for now and get things as good as possible without Audyssey. Otherwise you are chasing your tail.

2. Yes, run one speaker at a time when measuring to see what's going on. Later, you will run sub(S) + one speaker.

3. It's okay if you want to measure full range for now, but ultimately you'll want to do bass management right, cross over to subs, and tackle room modes, so don't worry about things in the mains that look wonky below 100hz. Setting up the right crossover settings (not just frequency choice, but distance/phase, etc) will let you address that:

and

4. It wasn't clear to me where that dip is from the diagrams since most are not un eq-ed but it sounds like SBIR may be the culprit. Moving the speakers as close as possible to the front wall, placing thick acoustic treatment directly behind them, and re-measuring to confirm that solved it will tell you more.

4. I don't see a layout, diagram or picture of your room. That plus measurements would help other folks advise you on what else might be useful.

Thanks for the reply offering to help. I may have to try to get some acoustic treatment to see how that may help and then report back. Originally, I was going to put some treatment behind that left speaker and side wall for the left hoping it would help, but I didn't 'know' there was a problem. Now, I've spent $100 to see evidence of the problem. So, I may have to wait a bit before I shell out more money to try to fix the problem that the money for the mic to get the evidence cost took away from fixing the problem.

 

[ryanosaur]

Thanks for the reply offering to help. I may have to try to get some acoustic treatment to see how that may help and then report back. Originally, I was going to put some treatment behind that left speaker and side wall for the left hoping it would help, but I didn't 'know' there was a problem. Now, I've spent $100 to see evidence of the problem. So, I may have to wait a bit before I shell out more money to try to fix the problem that the money for the mic to get the evidence cost took away from fixing the problem.

Applying room treatments without a measurement mic is usually not a great idea.
I’m not a REW monster by any means, but I consider both my Umik and Omnimic purchases to be worthwhile investments.

 

[nathan_h]

It looks like there is lots of free setup stuff still to explore, now that you have the microphone and have leaned to use REW, before you spend money on anything.

There is one measurement I see that doens't use Audyssey and shows single speaker sweeps (so its one the few measurements we can look at to actually understand the room), and it looks like you have response above 100hz that is +/-3db all the way to above 10khz. That is a great raw result. People spend thousands of dollars trying to achieve that, and you have it already!

So then it comes to how you are setting up your subs, and integrating your subs. That's free stuff again, especially since you mention you already bought a miniDSP right? You can do it without one, too, with a little more effort.

Then, after all that, you can return to using Audssey, which again is a free solution.