AVS › AVS Forum › Audio › Audio theory, Setup and Chat › The "Official" xtz Room Analyzer Thread
New Posts  All Forums:Forum Nav:

The "Official" xtz Room Analyzer Thread - Page 2

post #31 of 291
8/1/10 at 9:30am
Quote:
Originally Posted by fsrenduro View Post

Are you sure you had the setting on the back of the BFD at -10dBv (pushed in for home) instead of +4dBv (pushed out for studio)?

You're right--it was out! Perhaps should try it again.
Reply
Reply

AVS Top Picks

post #32 of 291
8/1/10 at 10:22am
Quote:
Originally Posted by fsrenduro View Post

Are you sure you had the setting on the back of the BFD at -10dBv (pushed in for home) instead of +4dBv (pushed out for studio)?

Thanks much for pointing this out; the BFD now seems to fit in very well.

So now I have a choice (thanks to XTZ which can measure things so efficiently): (1) look at the result with no eq, use the BFD to put in the filters that XTZ suggests, then add Audyssey; (2) replace the BFD with the AntiMode; run it, then run Audyssey. (2) is obviously simpler; one takes whatever the AM gives and doesn't worry about it. (1) is more flexible in allowing one to set whatever filters you want (and see the result with XTZ) over the whole freq range, not just the sub region. Any suggestions from those of you who are more experienced than I? (As noted above, my room is pretty good before any eq, so it seems I don't need any drastic measures.)

One more note: when running XTZ I've simply been holding mike stand in my hand; when doing the 3 position measurement I hold it so the mike is ~ 2 ft to the left of my position, then at it, then ~ 2 ft to the right (very similar to Audyssey's recommendations). Is this too primitive?
Reply
Reply
post #33 of 291
8/1/10 at 6:58pm
I loved the 033 when I had one but with Audyssey I felt it was redundant. My current setup has a BFD doing some pre-work and then Audyssey Pro taking over from there. Great results can be achieved other ways but since the BFD is so cheap and Audyssey does what the 8033 is doing I figured it was the more economical way to go. I'm very happy with the way my system sounds.
Reply
Reply
post #34 of 291
8/4/10 at 5:03pm
Quote:
Originally Posted by millerwill View Post

I'm confused about how to plug the RCA cable into my Onkyo 1007 AVR. I want to check out the response to my full audio system, after running the 8033 AntiMode followed by Audyssey. Do I use RCA cable coming out of the mike, and then the Y-cable and plug it into one of the 1007's stereo inputs (e.g., the VCR input)? Then I don't see any use for the two RCA connections coming out of the base. ?? TIA

Did you get an answer about the issue above?

I did the same thing as well in order to measure the response of my sub + mains. Added the Y-cable and plugged it into my AUX1 inputs. Then set AUX1 to stereo only (otherwise the surrounds would also play). BTW, my AVR is the Onkyo 876. I assume this is the correct method?

Then I could turn on or off MultEQ XT in the AVR and chart the freq response.
Reply
Reply
post #35 of 291
8/4/10 at 5:05pm
Quote:
Originally Posted by millerwill View Post

One more note: when running XTZ I've simply been holding mike stand in my hand; when doing the 3 position measurement I hold it so the mike is ~ 2 ft to the left of my position, then at it, then ~ 2 ft to the right (very similar to Audyssey's recommendations). Is this too primitive?

I would prefer putting the XTZ mic on a stool. More stable than when handheld.
Reply
Reply
post #36 of 291
8/4/10 at 5:09pm
Quote:
Originally Posted by fsrenduro View Post

I loved the 033 when I had one but with Audyssey I felt it was redundant. My current setup has a BFD doing some pre-work and then Audyssey Pro taking over from there. Great results can be achieved other ways but since the BFD is so cheap and Audyssey does what the 8033 is doing I figured it was the more economical way to go. I'm very happy with the way my system sounds.

Did you manage to use the XTZ to graph the AntiMode? Just curious how well it worked to flatten the freq response. I did graph my response with MultEQ XT and it's not nearly as flat as I would like it. Now thinking of adding something like the AntiMode (fully automatic) or the BFD 2496 (manual). The tweaker in me leans towards the 2496 so I have more control over the process.

What's your thoughts?
Reply
Reply
post #37 of 291
8/4/10 at 5:34pm
Quote:
Originally Posted by jchong View Post

I would prefer putting the XTZ mic on a stool. More stable than when handheld.

It's really not hard to hold it quite steady. And even if it did vary a bit, wouldn't this just help in averaging the area covered?
Reply
Reply
post #38 of 291
8/4/10 at 6:06pm
Thread Starter 
Quote:
Originally Posted by millerwill View Post

It's really not hard to hold it quite steady. And even if it did vary a bit, wouldn't this just help in averaging the area covered?

Actually, if it moves it could pick up some "wind noise", (i.e., air rushing over the mic.) I put the mic in the base and set the base on a fluffy pillow at the measurement position. I don't like it directly on the seating because the subs tend to make the seating vibrate a little. The fluffy pillow isolates the base from the seating.

The other thing you could do is get a tripod mic stand/boom with a clip on it. That is actually what Audyssey recommends for their mic.

Craig
Reply
Reply
post #39 of 291
8/4/10 at 6:13pm
Quote:
Originally Posted by craig john View Post

Actually, if it moves it could pick up some "wind noise", (i.e., air rushing over the mic.) I put the mic in the base and set the base on a fluffy pillow at the measurement position. I don't like it directly on the seating because the subs tend to make the seating vibrate a little. The fluffy pillow isolates the base from the seating.

The other thing you could do is get a tripod mic stand/boom with a clip on it. That is actually what Audyssey recommends for their mic.

Craig

Thanks, Craig. I do use a tripod for the Audyssey mike and could certainly do this for the XTZ one. (And I imagine I could use the hand-held procedure with the Audyssey mike!) It just seemed to me that this was not a 'high precision' operation, but I can certainly do it. When I do the XTZ measurement successively, the results are the same +/- 10%, which seemed to me to be within accuracy of the whole process.
Reply
Reply
post #40 of 291
8/4/10 at 6:15pm
Thread Starter 
Quote:
Originally Posted by jchong View Post

Did you manage to use the XTZ to graph the AntiMode? Just curious how well it worked to flatten the freq response. I did graph my response with MultEQ XT and it's not nearly as flat as I would like it. Now thinking of adding something like the AntiMode (fully automatic) or the BFD 2496 (manual). The tweaker in me leans towards the 2496 so I have more control over the process.

What's your thoughts?

Can you post a pic of your response? Did you measure at the point of your 1st Audyssey measurement, or did you do a 3-position measurement and average them?

I have gotten away from using a secondary EQ with Audyssey. I used to have an SMS-1 and I had subs with a built-in single band PEQ. The redundant EQ's caused what I would describe as a loss of articulation. The extra A to D and D to A conversions, plus the additional delays and phase changes were what I attribute it to. I really much prefer the SQ of just Audyssey MultEQ XT by itself over having additional EQ's in the circuit, even if the *graph* looks better.

YMMV, etc.

Craig
Reply
Reply
post #41 of 291
8/5/10 at 5:58am
Quote:
Originally Posted by craig john View Post

I have gotten away from using a secondary EQ with Audyssey. I used to have an SMS-1 and I had subs with a built-in single band PEQ. The redundant EQ's caused what I would describe as a loss of articulation. The extra A to D and D to A conversions, plus the additional delays and phase changes were what I attribute it to.

In relation to the issue of additional delays and phase changes, wouldn't running Audyssey after the secondary EQ account for those matters and correct it?

My graph after running Audyssey still has a lot of dips and peaks. Thinking of getting the BFD 2496 to help.
Reply
Reply
post #42 of 291
8/5/10 at 7:57am
Quote:
Originally Posted by jchong View Post

In relation to the issue of additional delays and phase changes, wouldn't running Audyssey after the secondary EQ account for those matters and correct it?

My graph after running Audyssey still has a lot of dips and peaks. Thinking of getting the BFD 2496 to help.

Yes, Audyssey will deal with delays and phase.
Reply
Reply
post #43 of 291
8/5/10 at 8:48am
Quote:
Originally Posted by millerwill View Post

Yes, Audyssey will deal with delays and phase.

Indeed but, since SMS-1 improvements in the amplitude domain may lead to new problems in the time domain (especially with boosts), Audyssey may be starting a few steps farther back than it might when working alone.
Reply
Reply
post #44 of 291
8/5/10 at 9:05am
How does XTZ compares to Room EQ Wizard? What does it offer in therms of software capabilities?
Reply
Reply
post #45 of 291
8/5/10 at 9:38am
Quote:
Originally Posted by Kal Rubinson View Post

Indeed but, since SMS-1 improvements in the amplitude domain may lead to new problems in the time domain (especially with boosts), Audyssey may be starting a few steps farther back than it might when working alone.

Kal, My present MO is to use the AntiMode 8033 (I've sold the SMS-1) first, and then to run Audyssey. Although I have to say, that after the AM, XTZ shows a freq response that varies less than 3 dB from 16 - 150 Hz, and waterfalls that die out before 100 msec. So there's not much for Audyssey to have to deal with in the sub region.
Reply
Reply
post #46 of 291
8/5/10 at 9:46am
Thread Starter 
Quote:
Originally Posted by Kal Rubinson View Post

Indeed but, since SMS-1 improvements in the amplitude domain may lead to new problems in the time domain (especially with boosts), Audyssey may be starting a few steps farther back than it might when working alone.

Agreed. Plus, Audyssey is a multi-point measurement and EQ system. If the outboard EQ is a single point EQ, the boosts and cuts employed by the outboard EQ may actually be "re-corrected" by Audyssey, due to measurements it found in other locations.

I have used an SMS-1 in conjunction with Audyssey MultEQ XT, and I spent considerable time listening with the two combined. I could make the primary LP virtually flat with the SMS-1 at the primary LP. However, it always made the other positions worse, (or, at least, no better.) With Audyssey alone, I find the bass more "articulate", than with the SMS-1 in the circuit, even though the curve isn't quite as flat.

Quote:


My graph after running Audyssey still has a lot of dips and peaks. Thinking of getting the BFD 2496 to help.

Be sure you are looking at the 3-position curve, not just the primary LP. I would expect the 3-position curve to be flatter than the single position curve.

If you want to play around, try running Audyssey with 8 positions clustered very closely around the primary LP, like within a few inches of each other. Then measure the response with xtz with a single position measurement at the primary LP. I'll bet it will be very flat. Re-run it again the "correct" way, and the flat response at the primary LP will not be as good, but the 3-position measurement will be better. Then, pick your poison.

Craig
Reply
Reply
post #47 of 291
8/5/10 at 11:45am
Quote:
Originally Posted by jchong View Post

Did you manage to use the XTZ to graph the AntiMode? Just curious how well it worked to flatten the freq response. I did graph my response with MultEQ XT and it's not nearly as flat as I would like it. Now thinking of adding something like the AntiMode (fully automatic) or the BFD 2496 (manual). The tweaker in me leans towards the 2496 so I have more control over the process.

What's your thoughts?

I don't have any of the graphs from when I had the 8033. I like tweaking myself so while the Anti-Mode is a great product it wasn't for me. With the XTZ and a BFD you can do a lot. I also have a calibrated ECM8000 and REW to do the more than 3 measurement tweaks. Since the BFD1124 is so cheap you can't really lose by trying one even the 2496 is relatively cheap and easy to resell.

I feel the Anti-Mode is perfect for those that don't have sub correction within their receivers, but if you want to tweak and make your own adjustments BFD's are hard to beat for the price investment.
Reply
Reply
post #48 of 291
8/5/10 at 11:56am
Quote:
Originally Posted by Victor View Post

How does XTZ compares to Room EQ Wizard? What does it offer in therms of software capabilities?

REW is great but you'll need to invest in a calibrated mic, mic preamp and USB input to your computer to get to a level playing field comparison with XTZ.

Just comparing the programs REW does have more flexibility. It allows for more measurement locations and has the ability to download EQ settings to certain EQ's via midi (Behringer stuff), etc.

Both are great. XTZ takes the thinking out of what mic, soundcard and preamp you'll need as well as having to get them all properly intergrated. You could probably build a great measuring system around REW for close to the price of an XTZ if you don't mind the integrating and buying of different pieces.
Reply
Reply
post #49 of 291
8/5/10 at 11:59am
Quote:
Originally Posted by fsrenduro View Post

REW is great but you'll need to invest in a calibrated mic, mic preamp and USB input to your computer to get to a level playing field comparison with XTZ.

Just comparing the programs REW does have more flexibility. It allows for more measurement locations and has the ability to download EQ settings to certain EQ's via midi (Behringer stuff), etc.

Both are great. XTZ takes the thinking out of what mic, soundcard and preamp you'll need as well as having to get them all properly intergrated. You could probably build a great measuring system around REW for close to the price of an XTZ if you don't mind the integrating and buying of different pieces.

I've used a SMS-1, the BFD 1124, and now a AntiMode 8033 (and then run Audyssey to boot). Yes, it was great fun to tweak with the SMS-1 and the BFD, but since I've had the XTZ to look at all the various results, I've never gotten them to give me as good freq and waterfall results as has the 8033. So I have now sold the SMS-1 (and put the BFD back down in storage)
Reply
Reply
post #50 of 291
8/5/10 at 4:41pm
Quote:
Originally Posted by millerwill View Post

Kal, My present MO is to use the AntiMode 8033 (I've sold the SMS-1) first, and then to run Audyssey. Although I have to say, that after the AM, XTZ shows a freq response that varies less than 3 dB from 16 - 150 Hz, and waterfalls that die out before 100 msec. So there's not much for Audyssey to have to deal with in the sub region.

So is 100 msec considered a good waterfall decay time?

Thanks.
Reply
Reply
post #51 of 291
8/5/10 at 5:02pm
Quote:
Originally Posted by moonhawk View Post

So is 100 msec considered a good waterfall decay time?

Thanks.

I believe so, but will defer to the experts.
Reply
Reply
post #52 of 291
8/5/10 at 7:31pm
Thread Starter 
Mark Seaton posted the following over in the Audyssey thread in response to a question about using PLIIx processing while measuring the results of the Audyssey EQ:

Quote:
Originally Posted by Mark Seaton View Post

Hi Jerry,

As others noted, you do want to turn off DEQ if you are measuring, as you want DEQ to start with a more neutral response.

Contrary to what some have posted, you WANT to use the sound modes you will be listening in, as this is how you can see the result of the processing Audyssey has done. My preference is to keep to ProLogic II-Cinema mode (not Music), as this gives you a simple routing of signals that allows you to test the bass management and interactions of all the speakers. In PL2-Cinema mode a signal into one channel, right or left, will route to the respective speaker. If you now send the exact same signal to both channels (Y-split or 2 matched channel outputs) the output will be routed to the center channel.

Of course any measurements taken manually only show you the signal manipulation, which is the same for any measurement location. This corrective action can be in both the time and frequency domain, meaning you would be interested in both a spectral decay plot along with the magnitude response. The main adjustment I find useful after running Audyssey are crossovers and subwoofer distance to get the best integration through crossover. If Audyssey selects a crossover around the one you actually use, you have a better chance of good integration, but in many cases there is significant room for improvement.

The biggest shortcoming of most any of the room correction systems still lies in determining the starting locations you give the correction system to work with. This is where external measurements greatly help, as well as fine tuning crossovers and sometimes channel levels after Audyssey does it's EQ. Hopefully one day Audyssey will add in the functionality of first displaying raw subwoofer response at a few subwoofer and/or listening locations.

I thought this was an excellent idea and thought I would share it with xtz owners. If you want to see the thread it was posted in, click on the little arrow next to Mark Seaton's name.

Craig
Reply
Reply
post #53 of 291
8/5/10 at 7:59pm
Thread Starter 
Quote:
Originally Posted by moonhawk View Post

So is 100 msec considered a good waterfall decay time?

Thanks.

The RT-60 is the usual measurement used to access the spectral decay. The RT-60 is the time it takes for reverberation to drop 60 dB. A "good" RT-60 is usually considered 300 Milliseconds or less.

xtz measures the RT-60 on the full range measurement, but just in the range from 125 Hz up. However the spectral decay in the bass range can been seen on the FR graphs, in the upper right corner, (and by clicking on it to exchange it with the FR graph.) Here is an example:


Note the scale on the right in dB. The color represents the SPL. Red is 10 dB down. Yellow is 20 dB down. Cyan is 30 dB down, etc. In the above graph, there is a room mode at ~32 Hz. This mode resonates out to about 225 ms., but the RT-60 would be well under 300 ms. This would be very good response in this range.

Here is a chart of the pre-Audyssey response in my room:


In the spectral decay chart in the upper right, you can see the prolonged resonances, with red and yellow out past 100 ms. at 20 and 40 Hz, and cyan and blue all the way across the chart.

Here is the post-Audyssey response:

Not the reduction in spectral decay times in the 20 and 40 Hz ranges. This is the effect of the frequency and time domain filters in the Audyssey algorithm. IME, this time domain correction is just as important as the frequency domain improvement seen the FR charts.

Craig
Reply
Reply
post #54 of 291
8/5/10 at 9:50pm
John,

That was a wonderful explanation. I now understand what I'm looking at with that 2D waterfall.

So our ultimate goal is to have as much dark blue, as early as possible, and have the flattest FR curve we can more or less.

Thanks again!
Reply
Reply
post #55 of 291
8/5/10 at 9:51pm
Quote:


Agreed. Plus, Audyssey is a multi-point measurement and EQ system. If the outboard EQ is a single point EQ, the boosts and cuts employed by the outboard EQ may actually be "re-corrected" by Audyssey, due to measurements it found in other locations.

Thanks for pointing this out again. I had forgotten that Audyssey is a multi-point system. I was stuck in thinking about optimising the freq response for my main listening position.

I suppose in doing a multi-point measurement, Audyssey tries to achieve the best compromise for all the measurement points? Sort of like a "jack of all trades, master of none" approach?

Quote:


Be sure you are looking at the 3-position curve, not just the primary LP. I would expect the 3-position curve to be flatter than the single position curve.

Indeed, the 3 position XTZ graph is much flatter than the primary LP. Like I realised above, I was stuck in thinking mostly about the primary LP and trying to optimise for that.

Quote:


If you want to play around, try running Audyssey with 8 positions clustered very closely around the primary LP, like within a few inches of each other. Then measure the response with xtz with a single position measurement at the primary LP. I'll bet it will be very flat. Re-run it again the "correct" way, and the flat response at the primary LP will not be as good, but the 3-position measurement will be better. Then, pick your poison.

Thanks for that. I'll try it and see how it goes. Then pick my poison

I suppose the only way to achieve flatter response across multiple seating positions is to have multi-subs? Put another way, with only 1 sub there is no way to have optimal freq response for multiple positions?
Reply
Reply
post #56 of 291
8/5/10 at 10:07pm
Quote:
Originally Posted by fsrenduro View Post

REW is great but you'll need to invest in a calibrated mic, mic preamp and USB input to your computer to get to a level playing field comparison with XTZ.

Just comparing the programs REW does have more flexibility. It allows for more measurement locations and has the ability to download EQ settings to certain EQ's via midi (Behringer stuff), etc.

Both are great. XTZ takes the thinking out of what mic, soundcard and preamp you'll need as well as having to get them all properly intergrated. You could probably build a great measuring system around REW for close to the price of an XTZ if you don't mind the integrating and buying of different pieces.

I agree. For me, the main advantage of XTZ is that it's a complete package and well integrated. No need to buy anything else and figure out how to marry them all together.

That said, I do note some limitations especially on the software side. I do have a wish list for a ver 3.0 of the software (crossing my fingers it'll come out in future):

1. Option to have un-smoothed graph to show more detail. The present 1/6 octave smoothing is still more smoothing than I would prefer.

2. Ability to label and adjust the graph scale. At present it's only labelled at 16, 31.5, 63, 125 and 250 with gridlines at 1/3 octave. I would love an option to label each gridline and change the gridlines to every 10Hz instead of 1/3 octave.

3. Having a 3D waterfall would be nice.

4. Easier way to set the test tone volume and which is user definable. The abrupt cut off at present is disruptive and I'm not sure what is the cut off point.
Reply
Reply
post #57 of 291
8/5/10 at 10:21pm
Quote:
Originally Posted by craig john View Post

The RT-60 is the usual measurement used to access the spectral decay. The RT-60 is the time it takes for reverberation to drop 60 dB. A "good" RT-60 is usually considered 300 Milliseconds or less.

...snip...

Thanks for posting your charts and explaining. Just curious:

1. Is the chart for just the sub alone or with mains included also?
2. 1 position or 3 position measurement?

To share here's my chart, XTZ plugged directly to sub input, mic at main LP only (1 pos measurement):


Reply
Reply
post #58 of 291
8/6/10 at 5:35am
Thread Starter 
Quote:
Originally Posted by jchong View Post

I suppose the only way to achieve flatter response across multiple seating positions is to have multi-subs? Put another way, with only 1 sub there is no way to have optimal freq response for multiple positions?

Multiple subs give you best chance for smooth response at multiple seats. Audyssey makes an attempt at smoothing the response for a listening "area". However, the better the initial response, the better the Audyssey result. Multiple subs, placed well around the LP give the best chance for that.

My graphs show the response of 3 Seaton Sound Submersives, placed randomly around the listening position.

Craig
Reply
Reply
post #59 of 291
8/6/10 at 5:37am
Thread Starter 
Quote:
Originally Posted by jchong View Post

3. Having a 3D waterfall would be nice.

Have you loked at the "Full Range" tab? It shows a 3-D waterfall. Of course, you need to run through your speakers to get a true full range measurement.

Craig
Reply
Reply
post #60 of 291
8/6/10 at 5:52am
Thread Starter 
Quote:
Originally Posted by jchong View Post

Thanks for posting your charts and explaining. Just curious:

1. Is the chart for just the sub alone or with mains included also?
2. 1 position or 3 position measurement?

My charts are subs and speakers, and 3-position. I use an 80 Hz crossover on my speakers, (Atlantic Technology 8200e's). While it's interesting to see the upper range response of the sub when it is given the full xtz signal, I want to see how the sub blends with the speakers at the crossover. xtz is an excellent tool to optimize the response around the crossover point. (See Mark Seaton's post I quoted above.)

This weekend I will try his suggestion of using PLIIx Movie to measure the response of each speaker and sub independently and post the result. I'll also try to get some charts that show how to optimize the response around the crossover.

Quote:
Originally Posted by jchong View Post

To share here's my chart, XTZ plugged directly to sub input, mic at main LP only (1 pos measurement):



That's not bad at all. I would bet if you ran the signal through the entire system and applied Audyssey, you could get that from +/- 6dB to +/- 3dB.

What speakers do you have? What are your Bass Management settings?

Craig
Reply
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Audio theory, Setup and Chat

AVS Top Picks

AVS › AVS Forum › Audio › Audio theory, Setup and Chat › The "Official" xtz Room Analyzer Thread