Incorporating MiniDSP With Audyssey XT32

[agarwalro]

I think this article by JA is worth reading even for TheWarrior, if only because he quoted Toole and Olive too.

I just want to point out the CSD from the article and waterfall plots in this thread while looking similar are very different animals.

 

[William Lemmerhirt]

So glad you clarified that .

Hey! I don’t want you guys gettin the wrong idea. Lol

 

[agarwalro]

That's what I thought too, so I double checked to be certain. Everything was wired correctly. Red to red, black to black. I didn't expect anything at all. This was very noticeable. Even moreso when I removed the extra cable and put the jumpers back on. This is the first time I tried bi amping them at all. Vocals and instruments were all over the place. Sometimes it sounded like the singer was on both sides of me when he should have been dead center up front.

Weird.

The only way this is happening is the phase on one of those terminals is getting reversed when the jumper is off. Did you check speaker wires too?

Could you recreate this every time the jumper is removed?

Weird indeed.

 

[Pogre]

Well its not the pre-amp, its not the power amp, its not a cable issue, so that leaves some sort of XO mumbo-jumbo as the issue. Bifurcating your towers' internals seems to be the only unknown.

The only way this is happening is the phase on one of those terminals is getting reversed when the jumper is off. Did you check speaker wires too?

Could you recreate this every time the jumper is removed?

Weird indeed.

Nope. Straight up rookie mistake. I had the wires for the top terminals of the right speaker reversed.

 

[Mitchibo]

Nope. Straight up rookie mistake. I had the wires for the top terminals of the right speaker reversed.

Bwaaaahhhh! I knew you'd figure it out. So if this is working OK, where does the mini fit in?

 

[William Lemmerhirt]

Nope. Straight up rookie mistake. I had the wires for the top terminals of the right speaker reversed.

Honest Abe award...

 

[agarwalro]

Nope. Straight up rookie mistake. I had the wires for the top terminals of the right speaker reversed.

Nah! Try this one. Forgetting that power cable on receiver is unplugged. Then, finding the power button unresponsive, come to the conclusion that the damn thing's kaput. Hurl some choice insults at D&M and pack up for the day. The next day, when reaching back to disconnect cables, notice (and remember) you'd unplugged the power.

(Takes bow, waits for applause to stop.)

 

[PENG]

I would just add that on page 4 when talking about the audibility of phase, it can be audible below transition in the form of you can or cannot hear something due to the wavelengths involved. (seating position is either in phase or out at a certain frequency)

This is why I ask phase be included in these graphs since we're talking about subs.

As for your decision on absorption, what frequencies are you trying to absorb? If you crossover at 100 hz, those wavelengths are about 11 feet, going all the way to 56 feet @ 20 hz. So if a large peak is unable to be controlled with placement or DSP, you're gonna need some heavy absorption at those frequencies. Not even a 4in thick panel is going to help - just trying to save you some time/ money!

I was thinking about 115 to 400 Hz range so there shouldn't be a problem, but thanks. They have to be good looking though.

 

[PENG]

Peng, this is not directed at you specifically.

Y'all need to start measuring things one speaker at a time without EQ to get a clear baseline.
Then EQ plots for each channel.
Finally, channel+sub plots.

A 7.1 waterfall is practically useless.

I know, that's why I included the one with Audyssey off. I also did L, R, L+R, C, Sub1, Sub2 as well, I can post them all, but I already know the best looking one appears to be the 7.1 with Audyssey on. You can see why in my case I am quite please with Audyssey. It left something on the table, but if that's all I can have, I am fine with it though that won't stop me from keep trying. Trying different things to improve on something already decent is just as fun.

By the way, thank you for the feedback but at least for the moment I disagree with you that 7.1 plots are useless. To me, 7.1 plots are very useful because it tells me how all the speakers and subs would play together. I don't listen to 7ch stereo but do have quite a few 7.1 movies. You can have a straight line Stereo plot, but your 7Ch plot may not look good.

I plotted the individual speaker and subs only for information and analysis purposes so I may have an idea what to do for manual adjustments.

 

[PENG]

I don't have it that way any more...

I used y splitters to split the 2 l/r signals out from my receiver to 4 different inputs on my amp. Then I hooked 1 channel to each l/r upper and lower binding posts to the corresponding channels after removing the jumpers. Pretty standard stuff.

View attachment 22545
Picture 2 more rca's and 2 more sets of high level speaker wire.

I can tell you one possibility that sounds unbelievable but I have actually experienced with one batch of interconnects. That is, the 3.5 mm jack to RCA were soldered incorrectly so the L and R were reversed. There was no point exchanging it because I have two pairs like that so it must be a bad batch. So I re-soldered them myself. In your case it does not involve jack connectors, but something like that could still happen in the Y or interconnects.

 

[ATLAudio]

Peng, this is not directed at you specifically.

Y'all need to start measuring things one speaker at a time without EQ to get a clear baseline.
Then EQ plots for each channel.
Finally, channel+sub plots.

A 7.1 waterfall is practically useless.

This is why I calibrate the sub alone with REW. After that, I find the best place for the towers, measured alone, no sub. Then I let Audyssey take it from there. After that, that’s when I do some sweeps with LR, and C with subs, fine tuning distances, and tweaking a filter here and there for best results.

 

[PENG]

Peng, this is not directed at you specifically.

Y'all need to start measuring things one speaker at a time without EQ to get a clear baseline.
Then EQ plots for each channel.
Finally, channel+sub plots.

A 7.1 waterfall is practically useless.

This is why I calibrate the sub alone with REW. After that, I find the best place for the towers, measured alone, no sub. Then I let Audyssey take it from there. After that, that’s when I do some sweeps with LR, and C with subs, fine tuning distances, and tweaking a filter here and there for best results.

Yes, but you, and probably agarwalro too, seems to be talking about calibration, I am not. I plotted those graphs after calibration, just to see how all the speakers/subs play together. I thought we are sharing the results there, hoping that someone may be able to recognize potential issues and room to improve, then we could be talking about calibration again. So I guess we are not in disagreement, right? Again, I could post all the other plots for individual speaker and you will see that after calibration, the 7.1 one appears to look the best, in just in response, but waterfalls as well.

 

[PENG]

I couldn't find the ones for the surround and surround back speakers/subs with Audyssey off. They were plotted back in 2014. Below are the L,R,C,Front Sub1, Sub2 that I could find. Black and Red are the FL,FR, green is center, blue and yellow are the subs.

To show the phase, impulse and waterfalls, I would have to post each individually.

 

[agarwalro]

Yes, but you, and probably agarwalro too, seems to be talking about calibration, I am not. I plotted those graphs after calibration, just to see how all the speakers/subs play together.

After optimizing L (for example) and Sub individually you'd measure L+Sub to ensure optimal coupling at crossover.

My issue with the 7.1 is that audio aberrations with one speaker will be masked by another. The only insight I could draw is speakers vs sub(s) level match.

I don't understand what you mean by "how well they play together" and what it would correspond with on the 7.1 waterfall plot.

 

[PENG]

After optimizing L (for example) and Sub individually you'd measure L+Sub to ensure optimal coupling at crossover.

My issue with the 7.1 is that audio aberrations with one speaker will be masked by another. The only insight I could draw is speakers vs sub(s) level match.

I don't understand what you mean by "how well they play together" and what it would correspond with on the 7.1 waterfall plot.

I meant whether there are cancellations and boosts, i.e. the overall frequency response. We all know XOs are not brick walls so the speakers/subs do overlap and interact with each other. I understand your point about the audio aberrations thing that is obvious too, but it is what it is, you can't ignore the effects. I guess we will agree to disagree on this point.

As for waterfalls, i only look for the decay time while trying to find out how else I can make use of them.

 

[PENG]

After optimizing L (for example) and Sub individually you'd measure L+Sub to ensure optimal coupling at crossover.

I tried so many things back pre audyssey so I might have tried that one too. Honestly though, there are so many things to play with by measuring, manual adjusting, with and without REW and minidsp that while continue to play around, I really have to give Audyssey the credit for doing a good enough job so quickly. The after Audyssey calibration REW plots look good, the system sounds good, all for 30 minutes of work. Again, that buys me all the time I need to play outside of Audyssey.

So are you suggesting that the waterfall graphs generated by REW are useless, or somewhat useful, and how?

 

[Pogre]

Nah! Try this one. Forgetting that power cable on receiver is unplugged. Then, finding the power button unresponsive, come to the conclusion that the damn thing's kaput. Hurl some choice insults at D&M and pack up for the day. The next day, when reaching back to disconnect cables, notice (and remember) you'd unplugged the power.

(Takes bow, waits for applause to stop.)

At least I know I wasn't just hearing things!

Power cord. That's a good one. My dog has a way of stepping on my avr remote and Bluetooth keyboard both and finding the mute button almost every time. That caused me some panic once or twice when I couldn't figure out what was wrong!

 

[agarwalro]

Here are the falls for Sub only and 7.1 (5 subs in total) channel, REW default setting.
View attachment 22542

What's your noise floor?
Could you repost the 7.1 with lower limit same as Subs only?

 

[PENG]

What's your noise floor?
Could you repost the 7.1 with lower limit same as Subs only?

I am not sure about my noise floor but from the REW plots I think it is probably around 25 to 30 dB.
Below are the REW plots for different combinations of channels from 10 Hz to 300 Hz, hopefully that's what you meant by "with lower limit same as Subs only".. My subs really don't have much below 15 Hz.

 

[ATLAudio]

I am not sure about my noise floor but from the REW plots I think it is probably around 25 to 30 dB.
Below are the REW plots for different combinations of channels from 10 Hz to 300 Hz, hopefully that's what you meant by "with lower limit same as Subs only".. My subs really don't have much below 15 Hz.

View attachment 22554

That’s very low, usually 40+