"Official" Audyssey thread (FAQ in post #51779) - Page 19 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #541 of 71855 Old 01-04-2008, 06:21 PM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

I own the outboard Audyssey Pro SEQ, and I tried pointing the mic so that the tip faced the front speakers and then moved it between chirps to face the rears. Surprisingly the only difference this made was in the very highest frequencies (above 10 kHz) and only about a 2 dB difference. This may be different with the cheaper mic supplied with the AVR's or with more directional speakers, but pointing the mic made so little of a difference that I didn't bother when I reran the calibration.

I have Monitor Audio GS, direct radiating speakers for what it is worth.

- Tim

That is interesting. I think the cheaper mics do make a difference. I wasn't going to go into the details but since you brought this up, I think I will. I have an Onkyo TX-SR805 which includes one of those hockey puck mics:



My speakers are 5 identical M&K S-150's which as you may know are intentionally limited in their dispersion. This charactaristic may have contributed to my problem. Anyway, my center channel speaker sits beneath my TV near the floor. It is angled up to the listener's ear position so it sounds fine when sitting at the listening position (couch) but this caused problems during calibration. The problem is that the base of the mic is very large and after it is mounted on a tripod and pointed directly at the ceiling (per the instructions), the direct sound coming from the center channel was nearly totally obscured by the base of the mic and the tripod. I suspected this caused a high frequency roll-off during measurement that Audyssey was trying to correct for by boosting the highs! This definitely seems to have been my problem as the sound was brittle, harsh and metallic.

I found that if I angled the mic on the tripod so the mic opening was at a 90 degree angle with the center speaker baffle, the sound could perfectly graze the mic opening. The result was a MUCH smoother, more natural calibration in the center channel without any of the brittle, metallic character of prior calibrations. Unfortunately, my Onkyo gives me no indication of what corrective EQ Audyssey is applying and I don't have a spectrum analyzer to see see the difference so I don't really know where the EQ errors were but let me tell you I could hear them and I hated Audyssey because I thought that is just how it worked. I was pleased to find I was wrong.

Your mic however, is thin and tall and would not have obscured the sound from a low sitting speaker the way the hockey puck mics do. I suspect if I could have used a proper mic like yours that I would not have had any problems at all.

Of course this is kind of the opposite problem mentioned by the OP but it still relates to mic angle and proper grazing.

Mike
BigPines is offline  
Sponsored Links
Advertisement
 
post #542 of 71855 Old 01-04-2008, 08:13 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Great job Mike! You are an admirable problem solver and and equally thorough communicator. I could see how the base of that puck would get in the way of a low-sitting speaker.

The mic that is used with the Audyssey Pro unit is a very nice professional omni-directional type, and not just some cheap Behringer unit either.

Say, since you seem to be a good logical thinker, I'd like your opinion on something. Audyssey Pro is detecting a large dip in the upper bass response at several seating positions and it is applying a +7 to +9 dB boost centered at 75 Hz, spanning from 60 to 90 Hz (verified with an RTA).

While the boost does effectively flatten the graph, it also sounds like there is a peak in that frequency range with the EQ on. I have tried several different mic positions, even taking some in front of the couch where I know there is no 75 Hz dip but it doesn't offset the corrections made much.

Frankly I'm surprised that Audyssey would try to flatten such a large dip, as this is typically inadvisable, but it does and I can find no way to trick it into not doing it or doing it less.

Any thoughts?
hifisponge is offline  
post #543 of 71855 Old 01-04-2008, 09:35 PM
AVS Special Member
 
rnrgagne's Avatar
 
Join Date: Jan 2003
Location: Canada
Posts: 6,586
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 44
The first installers kit I got, actually has a re-badged Behringer ECM8000. The new kit has a skinnier mic that I'm not familiar with. If you look a the pic of the mic above the pick-up is recessed. What both the ECM and the new mic both have that likely makes "grazing" less important is the cut outs around the pick ups' circumference
rnrgagne is offline  
post #544 of 71855 Old 01-04-2008, 09:56 PM
AVS Special Member
 
Jeremy Anderson's Avatar
 
Join Date: Jan 2003
Location: Mobile, AL
Posts: 1,682
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 348
Quote:
Originally Posted by hifisponge View Post

Great job Mike! You are an admirable problem solver and and equally thorough communicator. I could see how the base of that puck would get in the way of a low-sitting speaker.

The mic that is used with the Audyssey Pro unit is a very nice professional omni-directional type, and not just some cheap Behringer unit either.

Say, since you seem to be a good logical thinker, I'd like your opinion on something. Audyssey Pro is detecting a large dip in the upper bass response at several seating positions and it is applying a +7 to +9 dB boost centered at 75 Hz, spanning from 60 to 90 Hz (verified with an RTA).

While the boost does effectively flatten the graph, it also sounds like there is a peak in that frequency range with the EQ on. I have tried several different mic positions, even taking some in front of the couch where I know there is no 75 Hz dip but it doesn't offset the corrections made much.

Frankly I'm surprised that Audyssey would try to flatten such a large dip, as this is typically inadvisable, but it does and I can find no way to trick it into not doing it or doing it less.

Any thoughts?

You didn't ask me, but I figured I'd offer a thought or two:

To logically muddle through this, we'd probably need more info; for instance, what crossover points you're using, if any, for all channels. If you're using a crossover between 60-90Hz and detecting that dip from that particular channel, it could be phase cancellation between the speaker and sub across the crossover range. In that case, you wouldn't know if Audyssey was applying a boost to correct this or simply altering time response to align phase, thus preventing the phase cancellation. Were you verifying via RTA by measuring the effect on in-room response or by analyzing from the pre-out itself to see the boost at the output stage? Also, was this boost applied to individual channels or just the subwoofer?

I can't speak for the Pro variant, but I found that mirroring my initial three MultEQ XT measurement positions a foot or so further into the room helped a lot with problems like this. That's especially true if your main seating happens to be within 18" or so of room boundaries as the seating in smaller home theaters can be.

Could it be that you're just not used to hearing the newly flattened response? If that dip is inherent to the seating positions pre-Audyssey, it may just be that you're not accustomed to the way it SHOULD sound and are perceiving that as a peak. Just wondering.

"Never believe any quote you read on the internet." - Abraham Lincoln
Jeremy Anderson is offline  
post #545 of 71855 Old 01-04-2008, 11:09 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,355
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 134
Mike, got it, thanks for the explanation.

Seems like a lot of HT setups would be inherently problematic in this regard with low mounted centers and perhaps L/r's, and high-mounted surrounds.

Noah
noah katz is offline  
post #546 of 71855 Old 01-05-2008, 12:58 AM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Jeremy Anderson View Post

You didn't ask me, but I figured I'd offer a thought or two:

To logically muddle through this, we'd probably need more info; for instance, what crossover points you're using, if any, for all channels. If you're using a crossover between 60-90Hz and detecting that dip from that particular channel, it could be phase cancellation between the speaker and sub across the crossover range. In that case, you wouldn't know if Audyssey was applying a boost to correct this or simply altering time response to align phase, thus preventing the phase cancellation. Were you verifying via RTA by measuring the effect on in-room response or by analyzing from the pre-out itself to see the boost at the output stage? Also, was this boost applied to individual channels or just the subwoofer?

I can't speak for the Pro variant, but I found that mirroring my initial three MultEQ XT measurement positions a foot or so further into the room helped a lot with problems like this. That's especially true if your main seating happens to be within 18" or so of room boundaries as the seating in smaller home theaters can be.

Could it be that you're just not used to hearing the newly flattened response? If that dip is inherent to the seating positions pre-Audyssey, it may just be that you're not accustomed to the way it SHOULD sound and are perceiving that as a peak. Just wondering.

Those are certainly logical thoughts, but I'm only hearing the problem with the sub on. When I run the L/R speakers full range, it goes away. The more that I look at the graphs from the RTA measurements I took, the more confused I become. Real RTA graphs are not as clean and pretty as the ones that Audyssey generates. I'm not sure that I'm just seeing what I want to see. When I look at the RTA graphs for the sub channel alone, it looks like Audyssey is boosting the dip between 70-90 Hz. But when I look at the combined measurement of the sub along with the Left and Right speakers there is no sign of this boost.

All measurements were taken of the in-room response, with Audyssey on and off, not taken from the output stage of the Audyssey unit.

Out of 16 measurement positions with Audyssey (not the RTA) I intentionally took 3-4 measurements two feet in front of the couch where I know the bass response is a bit more smooth. My listening seat (couch) places the listener's head about 40" from the back wall. The room is 23' x 13' x 9.

The bottom line is that with Audyssey on, I hear woofy, thumpy upperbass and the only way I have been able to somewhat remedy the problem is to create a gap in the x-over between the sub and the mains. I originally had the x-over set to 80Hz for all speakers, but changed the sub to 60Hz to create hole.

Yargh!
hifisponge is offline  
post #547 of 71855 Old 01-05-2008, 10:06 AM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

Those are certainly logical thoughts, but I'm only hearing the problem with the sub on. When I run the L/R speakers full range, it goes away. The more that I look at the graphs from the RTA measurements I took, the more confused I become. Real RTA graphs are not as clean and pretty as the ones that Audyssey generates. I'm not sure that I'm just seeing what I want to see. When I look at the RTA graphs for the sub channel alone, it looks like Audyssey is boosting the dip between 70-90 Hz. But when I look at the combined measurement of the sub along with the Left and Right speakers there is no sign of this boost.

All measurements were taken of the in-room response, with Audyssey on and off, not taken from the output stage of the Audyssey unit.

Out of 16 measurement positions with Audyssey (not the RTA) I intentionally took 3-4 measurements two feet in front of the couch where I know the bass response is a bit more smooth. My listening seat (couch) places the listener's head about 40" from the back wall. The room is 23' x 13' x 9.

The bottom line is that with Audyssey on, I hear woofy, thumpy upperbass and the only way I have been able to somewhat remedy the problem is to create a gap in the x-over between the sub and the mains. I originally had the x-over set to 80Hz for all speakers, but changed the sub to 60Hz to create hole.

Yargh!

Wow! It doesn't seem like you should have to do that does it?

This is a puzzling problem. So you are saying that the graphs look flat but it sounds chesty in the upper bass? How long have you been listening in that room? How long ago did you get Audyssey?

I know you won't be wild about hearing this but my first thought is along the lines of Jeremy's - that perhaps you have become accustomed to that dip and now it sounds too chesty since it has been flattened out. My experience is that Audyssey is very accurate if it has been set up correctly so I am at a bit of a loss as to why this would happen. It sounds like you are doing all the right things and as long as your RTA is showing flat FR, I would think Audyssey has done a good job. Are you taking the RTA measurements from the primary listening position I assume?

Perhaps this is a question Chris from Audyssey could answer. Sorry, I wish I could help with this one.

Mike
BigPines is offline  
post #548 of 71855 Old 01-05-2008, 10:35 AM
AVS Special Member
 
rnrgagne's Avatar
 
Join Date: Jan 2003
Location: Canada
Posts: 6,586
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 2 Post(s)
Liked: 44
To quote Kal from his review of the SEQ the graphs it produces are "optimistic".
rnrgagne is offline  
post #549 of 71855 Old 01-05-2008, 12:30 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
OK, bear with me, as I am still learning how to use the RTA program I used to measure the effects of Audyssey. I have to say first that it appears I was wrong about Audyssey boosting the a dip in the upper bass. I did not properly scale the graphs so that the output with Audyssey ON was at the same level as with it OFF. Making this adjustment to the graphs leaves me even more confused though. At least when I thought Audyssey was boosting the upper bass, I had a possible reason for the thumpy, woofy bass I hear when Audyssey is ON. Now, I've got nothing.

Anyway, here are some screen grabs from the RTA:

Subwoofer Only (measurements taken at main listening seat)

Blue line = Audyssey OFF
Purple line = Audyssey ON

Subwoofer Only (measurements taken 1 foot left of main listening seat)

Green line = Audyssey OFF
Purple line = Audyssey ON

L & R + Sub Average (measurements at four locations around the main listening seat)

Dark blue line = Audyssey OFF
Light blue line = Audyssey ON

Based on that last graph, it seems that I may need to adjust the phase or polarity of the sub to get a better blend between the mains and sub at 70Hz. Which is odd because this is the area that supposedly sounds peaky. But it wouldn't be the first time I've been wrong.
hifisponge is offline  
post #550 of 71855 Old 01-05-2008, 01:31 PM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
What RTA did you use to make these graphs?

Man, that last graph looks pretty dang sweet to me. With Audyssey engaged, everything is within 10db from 20Hz up to around 250Hz. I'd bet that is better than what my system would measure. Too bad your ears do not agree. I guess that makes them less useful.

Audyssey should be adjusting the phase if necessary so I doubt that it the problem. Very strange indeed.

Mike
BigPines is offline  
post #551 of 71855 Old 01-05-2008, 03:37 PM
AVS Special Member
 
Jeremy Anderson's Avatar
 
Join Date: Jan 2003
Location: Mobile, AL
Posts: 1,682
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 348
Based on the graphs, I have a thought: With Audyssey off, you had a huge broad room-induced peak centered around 30Hz. Accordingly, if you were setting subwoofer level with a SPL meter pre-Audyssey, you were likely getting so much energy below 50Hz that the crucial 40-80Hz range was weak... and that was the sound you were accustomed to hearing.

Once that peak was flattened out and the level brought up to match the other channels, this naturally means that mid-bass is now louder than you were used to hearing it before. You no longer have that low frequency peak affecting the calibration, meaning you're getting a more even response across the frequency range. All three graphs show a very nice improvement with the bass. Your last graph should verify that it isn't a phase cancellation issue, since you're getting improved response with Audyssey on; I would guess that this is more due to proper phase alignment than unnecessarily high boosting on Audyssey's part, since you're seeing that improvement in your crossover range.

Your pre-Audyssey response looks like a very steep house curve more than the desired flat response. Some people do this intentionally, because if we're being honest, most people aren't accustomed to hearing flat response with their subwoofers. Most people are accustomed to hearing the peaks caused by our rooms and they think that's the way it's supposed to sound.

So my opinion is that you aren't getting a boost in mid-bass. You're getting a reduction of deep bass and then a boost of the overall level to compensate, which makes mid-bass sound louder than you're accustomed to hearing. Give yourself a little time to get used to the way it sounds and then do some critical music listening, paying attention to instruments that fall in the lower ranges. You should hear a more consistent level between notes, particularly with bass guitar.

"Never believe any quote you read on the internet." - Abraham Lincoln
Jeremy Anderson is offline  
post #552 of 71855 Old 01-05-2008, 06:05 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Jeremy -

You obviously know your stuff, and given the problem, I would have given the same advice. However, I have been EQing my sub with a parametric EQ for many years prior to putting the Audyssey Pro unit in the system.

I have had my Velodyne DD15 subwoofer (which has a parametric EQ and RTA built into it) for about three years and used a Behringer EQ prior to that. With the Velodyne EQ I pulled down the peak at 30Hz and I put a mild boost between 40 & 50Hz, so I know what relatively flat bass sounds like prior to Audyssey.

What I am hearing with the critical music listening I have done so far are some "hot notes" in the mid/upper bass (not sure which). In other words, certain bass notes are elevated in level compared to those that neighbor it.

Got any recommendations for good bass guitar discs that I can use to verify the evenness of the frequency response?

Thanks for your thoughts on my problem.

- Tim
hifisponge is offline  
post #553 of 71855 Old 01-05-2008, 06:13 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by BigPines View Post

What RTA did you use to make these graphs?

Man, that last graph looks pretty dang sweet to me. With Audyssey engaged, everything is within 10db from 20Hz up to around 250Hz. I'd bet that is better than what my system would measure. Too bad your ears do not agree. I guess that makes them less useful.

Audyssey should be adjusting the phase if necessary so I doubt that it the problem. Very strange indeed.

Mike

The graphs do look a lot better with Audyssey ON, don't they? Shame that is not what I hear.

The RTA I use is the freeware "Room EQ Wizard". It's a great program, that is very well thought-out and easy to use.

http://www.hometheatershack.com/roomeq/

You should check it out. You seem like the type that would find it interesting and useful. You will need to pick up a measurement mic and a phantom power supply to use it, which should run you under $100 for the pair.

http://www.zzounds.com/item--BEHECM8000

http://www.zzounds.com/item--ARTPHANIII

Cheers,

- Tim
hifisponge is offline  
post #554 of 71855 Old 01-05-2008, 06:25 PM
AVS Special Member
 
Jeremy Anderson's Avatar
 
Join Date: Jan 2003
Location: Mobile, AL
Posts: 1,682
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 348
Quote:
Originally Posted by hifisponge View Post

Jeremy -

You obviously know your stuff, and given the problem, I would have given the same advice. However, I have been EQing my sub with a parametric EQ for many years prior to putting the Audyssey Pro unit in the system.

I have had my Velodyne DD15 subwoofer (which has a parametric EQ and RTA built into it) for about three years and used a Behringer EQ prior to that. With the Velodyne EQ I pulled down the peak at 30Hz and I put a mild boost between 40 & 50Hz, so I know what relatively flat bass sounds like prior to Audyssey.

What I am hearing with the critical music listening I have done so far are some "hot notes" in the mid/upper bass (not sure which). In other words, certain bass notes are elevated in level compared to those that neighbor it.

Got any recommendations for good bass guitar discs that I can use to verify the evenness of the frequency response?

Thanks for your thoughts on my problem.

- Tim

Hmm... well, I'm not really sure where that leaves you. My only other thought would be to reposition the sub to see if you can get a better uncorrected response before running Audyssey. Or perhaps do some acoustic treatment to smooth things out. That's a ginormous peak you have at the low end! I'm surprised Audyssey was able to reduce it as much as it did.

As for bass guitar discs, I'm no help there. My listening preferences quickly jump between punk rock and bluegrass, and are somewhat obscure even in those genres.

"Never believe any quote you read on the internet." - Abraham Lincoln
Jeremy Anderson is offline  
post #555 of 71855 Old 01-05-2008, 07:31 PM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Tim
Now that you have a RTA and are confident with using it, I would forget Audyssey for the moment. Try this:

1. With sub off, run RTA through ONE of your mains (say the left). Now move the speaker into different positions on a 8" grid on the floor. ie move the speaker 8" at a time, and re-test. The grid should cover an area of say 3' x 3' (if your room allows). Your RTA results should show you the best position. You then refine by re-testing at 4" positions around this "best" position to find your final position.

2. Repeat with the right speaker. If the room and furnishings/curtains,etc is symetrical, then the right should be placed in the mirror position to the left and just checked. Otherwise you will have to test it as per the right speaker.

3. Once you are happy with the response readings of the mains, then repeat the procedure with the sub (with the mains off).

What this does is tries to position the speakers where they receive minimal influence from the standing waves and other sound anomalies in the room.
Another aid to taming room effects is to place sound absorbers at the first reflection points on the walls and ceiling (and floor if you don't have heavy carpet). This is well covered in other forums in the AVS Forum.

Once all this is done, then run Audyssey. Hopefully then all Audyssey will be doing will be smoothing the response, not trying to cope with roller-coaster type response curves.

Hope this helps
Nordo is offline  
post #556 of 71855 Old 01-05-2008, 07:49 PM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

The graphs do look a lot better with Audyssey ON, don't they? Shame that is not what I hear.

The RTA I use is the freeware "Room EQ Wizard". It's a great program, that is very well thought-out and easy to use.

http://www.hometheatershack.com/roomeq/

You should check it out. You seem like the type that would find it interesting and useful. You will need to pick up a measurement mic and a phantom power supply to use it, which should run you under $100 for the pair.

http://www.zzounds.com/item--BEHECM8000

http://www.zzounds.com/item--ARTPHANIII

Cheers,

- Tim

Thanks,

Actually, I know about REW. I have not tried it yet but I always thought I would check it out one day.

Mike
BigPines is offline  
post #557 of 71855 Old 01-05-2008, 08:39 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Nordo -

That's a fantasic system for speaker placement. I wish I could put it to use. Unfortunately, my speaker placement is dictated by the room layout and all of the doors within it. My system is in a bonus room that has a door to the kitchen, one to the garage and another to the back porch. It's amazing that I could get a system to work in this room at all.

Hence my reason for looking for an electronic solution for my room and speaker placement problems.

I'm going to do more experimentation tomorrow morning and I'll post my findings.
hifisponge is offline  
post #558 of 71855 Old 01-05-2008, 08:49 PM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
When I set up my room and speakers, I used a brilliant program called ETF. You can use the standard Radio Shack SPL meter with it. It has a response calibration adjustment for the RS meter built in. This saves having to buy a special testing mic. BTW, I have read posts on other forums where the response of the SPL meter has been tested as being very good. The ETF program uses your standard sound card on your computer, but uses internal feed back to counteract any anomalies your sound card has.
The downloadable trial version allows you to carry out all the basic tests. but you can't layer your result curves onto the one page to compare you results (like you can with the full program). However using the trial version you can still capture the monitor screen and then print it to save your result.

This is the program's home page.
The website also has a fine tutorial.
Nordo is offline  
post #559 of 71855 Old 01-05-2008, 09:34 PM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Mike
Just further on the laser level you use on the tripod, can you post a photo of the way you use it?
I'm trying to figure out how you place it/fix it to the tripod head while the mic is on the head?
Nordo is offline  
post #560 of 71855 Old 01-05-2008, 09:43 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Nordo View Post

When I set up my room and speakers, I used a brilliant program called ETF. You can use the standard Radio Shack SPL meter with it. It has a response calibration adjustment for the RS meter built in. This saves having to buy a special testing mic. BTW, I have read posts on other forums where the response of the SPL meter has been tested as being very good. The ETF program uses your standard sound card on your computer, but uses internal feed back to counteract any anomalies your sound card has.
The downloadable trial version allows you to carry out all the basic tests. but you can't layer your result curves onto the one page to compare you results (like you can with the full program). However using the trial version you can still capture the monitor screen and then print it to save your result.

This is the program's home page.
The website also has a fine tutorial.

Hello again -

I own ETF as well as REW, but I like the usability and interface of REW better, and it's free. You can get away with the Rat Shack SPL meter for bass measurements, but I can assure you that it is useless for full frequency range readings (it has highly inaccurate treble response). Oh, you can use the SPL meter with REW as well and it does most if not all that ETF does, including waterfall plots, averaging, smoothing, impulse response and so on.

- Tim
hifisponge is offline  
post #561 of 71855 Old 01-06-2008, 02:42 AM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

Hello again -

I own ETF as well as REW, but I like the usability and interface of REW better, and it's free. You can get away with the Rat Shack SPL meter for bass measurements, but I can assure you that it is useless for full frequency range readings (it has highly inaccurate treble response). Oh, you can use the SPL meter with REW as well and it does most if not all that ETF does, including waterfall plots, averaging, smoothing, impulse response and so on.

- Tim

Several years ago, there was a test done comparing the Rat Shack SPL meter with a high quality high resolution laboratory frequency meter. The Rat Shack meter compared quite well with the laboratory meter across the range 10Hz to 20,000Hz. From this testing they produced a calibration adjustment table for the readings you would take with the Rat Shack meter. This testing, and the resultant table was reported in a forum called "The Bass List" (now called DIYSpeakers.Net).

The ETF uses this adjustment table to calibrate readings when using the Rat Shack meter (as can anyone using the RS meter to manually record sound levels of variously frequencies). If anyone wants, I can probably dig up the adjustment table.

The caveat is that the test and the resultant adjustment table is approximately 10 years old, and the circuit/electronics of the Rat Shack SPL meter may have changed probably producing a totally different response curve from the meter.
Nordo is offline  
post #562 of 71855 Old 01-06-2008, 02:24 PM
Advanced Member
 
David Aiken's Avatar
 
Join Date: Mar 2002
Posts: 617
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 21
Quote:
Originally Posted by Nordo View Post

Several years ago, there was a test done comparing the Rat Shack SPL meter with a high quality high resolution laboratory frequency meter. The Rat Shack meter compared quite well with the laboratory meter across the range 10Hz to 20,000Hz. From this testing they produced a calibration adjustment table for the readings you would take with the Rat Shack meter. This testing, and the resultant table was reported in a forum called "The Bass List" (now called DIYSpeakers.Net).

The ETF uses this adjustment table to calibrate readings when using the Rat Shack meter (as can anyone using the RS meter to manually record sound levels of variously frequencies). If anyone wants, I can probably dig up the adjustment table.

The caveat is that the test and the resultant adjustment table is approximately 10 years old, and the circuit/electronics of the Rat Shack SPL meter may have changed probably producing a totally different response curve from the meter.

There is another caveat and it is important.

High quality, high resolution meters have much more extensive quality control than the Radio Shack meter and there is probably much less variation across individual meters. Before use, the high quality meter is recalibrated using a separate calibration device in order to ensure accuracy of readings. There is no such calibration device for the Radio Shack meter and the high level meters do need to be adjusted from time to time. Their accuracy does measurably deteriorate over time. Finally, the high quality meters also need to be sent for servicing and a more extensive recalibration from an accredited service centre each year. Professionals do this as a matter of course and part of the protocol behind professional testing is that the meter used is of a professional standard, professionally serviced and checked/recalibrated within the last 12 months, and recalibrated with its matching recalibration device prior to each set of tests.

The calibration table you refer to is really only valid for that one particular Radio Shack meter used, and really only for its use very soon after the calibration table was prepared.

In practice, what will suffer the most with time is accuracy on SPL. Radio Shack meters would be expected to become increasingly less accurate over time. If you're using them for matching levels in a surround system or for measuring speaker response that's not much of an issue. Levels will be matched accurately if each channel delivers the same readingit doesn't matter whether that reading is, say, 80 dB as shown by the meter or even something as innacurate as 75 dB, and the meter would not be that far our. Provided measurements are identical, channels will be matched. Speaker response curves will also be reliable but the actual levels shown will not.

On the other hand, if you're using the meter for testing SPLs for legal compliance with hearing protection requirements or you're doing laboratory tests which require accurate measurement of SPLs, then the Radio Shack meter is not reliable and it's inappropriate to those kinds of use.
David Aiken is offline  
post #563 of 71855 Old 01-06-2008, 07:58 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by hifisponge View Post

The graphs do look a lot better with Audyssey ON, don't they? Shame that is not what I hear.

I'm 90% sure that I have solved the problem I was having with peaky sounding upper bass with Audyssey ON. This was a rather tangled knot of a problem, but no one said this hobby was easy if you want to do it right.

The short explanation is that all it took was moving the crossover from 80 to 60 Hz for both the main speakers and the sub.

The long explanation is as follows:

I started by setting the crossover where I wanted it -- at 80 Hz for all speakers including the sub. I then used the RTA built into my Velodyne DD sub and measured the effects Audyssey ON and OFF at the couch. The first thing I noticed was that with Audyssey ON the response looked virtually ruler flat! Flatter than I have ever achieved using the parametric filters in the sub. Normally this would be great, but of course I didn't like what I was hearing. So I turned Audyssey OFF and saw that there was a broad dip centered at 80 Hz, like I was used to getting. Obviously, Audyssey saw this dip and tried to correct it. This is where it gets complicated, but I will try my best.

The first problem was that I had been using the dynamic loudness function of my Lexicon prepro prior to Audyssey. I always made sure to turn it off while calibrating and measuring, but turned it back on without thought right after. With Audyssey ON, the dynamic loudness function was elevating the overall bass too much from 100 Hz down and making the problem I was hearing in the upper bass worse-- duh. I always liked the dynamic loudness function without Audyssey, but it was too much of a good thing with Audyssey engaged.

Along the way, I discovered that I am much more tolerant of excessive low bass (below 40 Hz) than I am of any boost in the upper bass. Low bass has no tone, it only adds weight to the sound, whereas peaks or elevations in the upper bass do have tone and stand out much more as hot notes.

When I turned off the dynamic loudness, things got a good bit better, but the upper bass still sounded a little too hot. What's interesting is that at the couch, the bass sounded very close to flat, but the moment you stepped three feet away from the couch into the room, the upper bass demon returned in full force. I think there is some psychology involved here because I have to walk across the room from my couch to turn Audyssey OFF. And right next to the AV cabinet is what I call "the bass pit". Bass energy tends to pool up right there and it sounds awful. Part of what I believe was happening was I was carrying the memory of the elevated upper bass in the majority of the room with me as I walked to sit down to listen at the couch. My focus was on the excessive upper bass as I walked through the rrom, so I was possibly hearing it when it wasn't there at the couch or would otherwise not find it objectionable.

I decided to take some measurements around the middle of the room. Sure enough, with Audyssey ON there was either large peaks at 45 Hz and 75 Hz, or a broad elevation from 45-90 Hz depending on the location. With Audyssey OFF, the peak from 60-90 Hz was greatly reduced but from 40-60 Hz it remained. Remember I find elevated low bass much less objectionable than upper bass so this sounded better. I may be wrong here, but I also have a feeling that all of that excess upper bass energy in the room everywhere except at the couch was in fact being heard somehow at the couch position.

Knowing now that I wanted to try and at least reduce the upper bass, I started playing with the cross over points. I moved the sub crossover down to 70 Hz and kept the other crossovers at 80 Hz to try to force a dip in the upper bass. Didn’t work. So I moved the sub crossover to 60 Hz. Much better, but now it sounded like there was a bit of hole. So finally I moved the crossover for just the L/R speakers down to 60 Hz and wah-lah, it all snapped together—better bass than I have ever heard in my room. The low end was all there, the upper bass sounds pitch-perfect and I now have punch and kick. The upper bass even sounds pretty good throughout the rest of the room, and tolerable in the bass pit.

I don’t know why I couldn’t just pick any crossover point that I wanted, and I don’t know why I heard excessive bass between 60 and 80 Hz when there is nothing in the graphs that says I should (at the couch anyway), but I’m glad I have a solution. Still, I’m the type that likes, no needs, to know “why”.

Thoughts welcome.

Hope that was at least a little bit interesting or helpful to someone other than me.

Cheers,

- Tim
hifisponge is offline  
post #564 of 71855 Old 01-06-2008, 11:31 PM
AVS Special Member
 
Jeremy Anderson's Avatar
 
Join Date: Jan 2003
Location: Mobile, AL
Posts: 1,682
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 348
But... doesn't setting the sub's crossover to 60Hz essentially throw away the upper half of the LFE channel?

"Never believe any quote you read on the internet." - Abraham Lincoln
Jeremy Anderson is offline  
post #565 of 71855 Old 01-06-2008, 11:47 PM
AVS Special Member
 
hifisponge's Avatar
 
Join Date: Jan 2004
Location: Redmond, WA
Posts: 7,516
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Jeremy -

Not in my prepro. The .1 channel is seperate from the crossover between the mains and the sub. I still also have my center and surrounds set at 80 Hz, so you would think that 80-60 Hz would get tossed too.

It works like this:

* 60 Hz and below from the mains goes to the sub
* 80-60 Hz from the center and surrounds goes to the mains and 60 and below to the sub
* 120 Hz and below in the LFE channel goes to the sub

- Tim
hifisponge is offline  
post #566 of 71855 Old 01-07-2008, 05:09 AM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

Jeremy -

Not in my prepro. The .1 channel is seperate from the crossover between the mains and the sub. I still also have my center and surrounds set at 80 Hz, so you would think that 80-60 Hz would get tossed too.

It works like this:

* 60 Hz and below from the mains goes to the sub
* 80-60 Hz from the center and surrounds goes to the mains and 60 and below to the sub
* 120 Hz and below in the LFE channel goes to the sub

- Tim

Keep in mind too that a nominated xover point, say 80Hz, is not a sudden cutoff. The level is 3db down at that frequency, then continues to reduce on a gradient that varies depending on the type of xover used. So the main, say, would still be contributing to sound levels well below 80Hz. In a similar fashoin the sub would still be contributing well above it's set xover frequency. If you cross them both at the same frequency, there is a good chance you could get a peak at that frequency where the mains and the sub are both producing sound. This is where phase becomes important. If the mains and sub are out of phase at the xover point, you can get a bad suck-out.
Nordo is offline  
post #567 of 71855 Old 01-07-2008, 05:22 AM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

-------------
What's interesting is that at the couch, the bass sounded very close to flat, but the moment you stepped three feet away from the couch into the room, the upper bass demon returned in full force. I think there is some psychology involved here because I have to walk across the room from my couch to turn Audyssey OFF. And right next to the AV cabinet is what I call "the bass pit". Bass energy tends to pool up right there and it sounds awful. Part of what I believe was happening was I was carrying the memory of the elevated upper bass in the majority of the room with me as I walked to sit down to listen at the couch. My focus was on the excessive upper bass as I walked through the rrom, so I was possibly hearing it when it wasn't there at the couch or would otherwise not find it objectionable.

I decided to take some measurements around the middle of the room. Sure enough, with Audyssey ON there was either large peaks at 45 Hz and 75 Hz, or a broad elevation from 45-90 Hz depending on the location. With Audyssey OFF, the peak from 60-90 Hz was greatly reduced but from 40-60 Hz it remained. Remember I find elevated low bass much less objectionable than upper bass so this sounded better. I may be wrong here, but I also have a feeling that all of that excess upper bass energy in the room everywhere except at the couch was in fact being heard somehow at the couch position.
-------------
Thoughts welcome.

Hope that was at least a little bit interesting or helpful to someone other than me.

Cheers,

- Tim

I wish I had sensitive ears like you

By the sound of it, you have really bad standing waves in your room from the sub. While playing a low tone, or some music with fairly continuous bass, you should be able to walk from one end of the room to the other and experience zones of very loud bass, and zones of very quiet bass.
I'm not familar with all the other forums on AVSforum, but there will certainly be one that will help you tame these standing waves using "bass traps", relocating the sub, etc. If you can't tame them, then at least position your main listening seat to a zone of medium loudness. Alternatively, move the sub to achieve the same effect.
Sounds like your having lots of fun
Nordo is offline  
post #568 of 71855 Old 01-07-2008, 05:27 AM
Advanced Member
 
Nordo's Avatar
 
Join Date: Feb 2004
Location: Noosa,Australia
Posts: 755
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by David Aiken View Post

There is another caveat and it is important.

High quality, high resolution meters have much more extensive quality control than the Radio Shack meter and there is probably much less variation across individual meters. Before use, the high quality meter is recalibrated using a separate calibration device in order to ensure accuracy of readings. There is no such calibration device for the Radio Shack meter and the high level meters do need to be adjusted from time to time. Their accuracy does measurably deteriorate over time. Finally, the high quality meters also need to be sent for servicing and a more extensive recalibration from an accredited service centre each year. Professionals do this as a matter of course and part of the protocol behind professional testing is that the meter used is of a professional standard, professionally serviced and checked/recalibrated within the last 12 months, and recalibrated with its matching recalibration device prior to each set of tests.

The calibration table you refer to is really only valid for that one particular Radio Shack meter used, and really only for its use very soon after the calibration table was prepared.

In practice, what will suffer the most with time is accuracy on SPL. Radio Shack meters would be expected to become increasingly less accurate over time. If you're using them for matching levels in a surround system or for measuring speaker response that's not much of an issue. Levels will be matched accurately if each channel delivers the same readingit doesn't matter whether that reading is, say, 80 dB as shown by the meter or even something as innacurate as 75 dB, and the meter would not be that far our. Provided measurements are identical, channels will be matched. Speaker response curves will also be reliable but the actual levels shown will not.

On the other hand, if you're using the meter for testing SPLs for legal compliance with hearing protection requirements or you're doing laboratory tests which require accurate measurement of SPLs, then the Radio Shack meter is not reliable and it's inappropriate to those kinds of use.

What you say makes a lot of sense.
However, the ETF is really only using the mic of the meter. Although I can see how it's response curve could change over time, and/or the effect of poor quality control in the manufacture of what would be a very cheap mic.
Nordo is offline  
post #569 of 71855 Old 01-07-2008, 07:34 AM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by Nordo View Post

Mike
Just further on the laser level you use on the tripod, can you post a photo of the way you use it?
I'm trying to figure out how you place it/fix it to the tripod head while the mic is on the head?

I'll try to post a photo later tonight. It really is not as technical as it sounds. If you don't have a laser level, a flashlight would work just fine too. Just hold it on the top of the tripod head so the laser/light is pointing at a 90 degree angle in relation to the mic opening and make your changes. The laser/light is not attached to the tripod in any way. You just hold it in place on the flat part of the tripod head when you need it.

Again, this is only intended to help you get the APPROXIMATE proper angle set quickly so you can get out of the room before the next chirps start.

Mike
BigPines is offline  
post #570 of 71855 Old 01-07-2008, 07:48 AM
Advanced Member
 
BigPines's Avatar
 
Join Date: Jul 2007
Posts: 749
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hifisponge View Post

* 80-60 Hz from the center and surrounds goes to the mains and 60 and below to the sub

Um, are you sure about this one? I have never seen a bass management implementation that works like this (not that that means a whole lot). What pre/pro are you using?

Mike
BigPines is offline  
Reply Receivers, Amps, and Processors

Tags
Audyssey , Receivers Amplifiers , Kef Kht1005 2se 5 1 Subwoofer Satellite System With C4 Subwoofer Gloss White , 5 6 7 1 7 2 Or 8 1 8 2 One Or Two Subwoofer Compatible 16 Banana Post 2 Rca Speaker Wall Plate For H
Gear in this thread

    Thread Tools
    Show Printable Version Show Printable Version
    Email this Page Email this Page


    Forum Jump: 

    Posting Rules  
    You may not post new threads
    You may not post replies
    You may not post attachments
    You may not edit your posts

    BB code is On
    Smilies are On
    [IMG] code is On
    HTML code is Off
    Trackbacks are Off
    Pingbacks are Off
    Refbacks are Off