|
|
Post by moodyman on May 11, 2010 20:02:48 GMT -5
If the EQ does nothing then it might be a problem with the input signal, settings or the firmware. As to the graphic EQ itself: It's impossible to correct low frequency room modes with a graphical EQ. A parametric EQ is needed that is variable in level, center frequency and bandwidth. This is something that is not open for discussion as Lonnie suggested in another thread. This is physics. With all due respect you are simply wrong. It is NOT "impossible" to use a graphic EQ to control low frequency room modes. Is it simpler, easier and more flexible to do so with a PEQ? Yes. of course. But a GEQ can absolutely perform a similar function. Acoustics is acoustics, and yes, it's physics. Please stop with the FUD. What??  How do you adjust a frequency with a GEQ for which you have no "slider" for?? Especially if you don't want to mess with the other freq's?? |
|
Lonnie
Emo Staff 
admin
Pay no attention to the man behind the curtain
Posts: 6,999
|
Post by Lonnie on May 11, 2010 21:18:50 GMT -5
I was using my Rives audio test disc to check the curve of my new 15" sub last night. I found I had about a 5db spike at 25hz. Tried to smooth this out with the UMC-1, but taking 22hz and 28hz all the way down to -10 did absolutely nothing. I believe this is just stereo audio though, not a dedicated .1 LFE track. So how the hell am I supposed to use the sub EQ? If the EQ does nothing then it might be a problem with the input signal, settings or the firmware. As to the graphic EQ itself: It's impossible to correct low frequency room modes with a graphical EQ. A parametric EQ is needed that is variable in level, center frequency and bandwidth. This is something that is not open for discussion as Lonnie suggested in another thread. This is physics. You are correct, Physics are physics and you can't change them. Thus you CAN make any adjustments with a graphic that can be done with a parametric. All EQs work the same way, period. The difference is that a parametric allows you to select the center freq, gain and sometimes Q. Where a graphic has set center freq and Qs. Bottom line is the phase shift and the way they work are identical in every regard. If you have a null point in your room there isn't an EQ in the world that can fix it. |
|
|
|
Post by markus on May 11, 2010 21:42:07 GMT -5
If the EQ does nothing then it might be a problem with the input signal, settings or the firmware. As to the graphic EQ itself: It's impossible to correct low frequency room modes with a graphical EQ. A parametric EQ is needed that is variable in level, center frequency and bandwidth. This is something that is not open for discussion as Lonnie suggested in another thread. This is physics. You are correct, Physics are physics and you can't change them. Thus you CAN make any adjustments with a graphic that can be done with a parametric. All EQs work the same way, period. The difference is that a parametric allows you to select the center freq, gain and sometimes Q. Where a graphic has set center freq and Qs. Bottom line is the phase shift and the way they work are identical in every regard. If you have a null point in your room there isn't an EQ in the world that can fix it. Lonnie So the bottom line is that graphic and parametric EQs are pointless because they can't correct room modes? If that would be true, one has to wonder why you implemented any EQs in the first place? Fortunately that is not true and parametric EQs can correct modal peaks (not dips) at low frequencies as long as level, center frequency and bandwidth are freely adjustible (graphic EQs don't provide that - a 1/3-octave EQ is too coarse). There are a couple of AES papers that describe just that (e.g. Celestinos/Nielsen). I guess you need to read up on it. There are several EQ solutions on the market (REW+BFD, Anti-Mode 8033) that work very well. I can post measurements of my own room any time. Best, Markus |
|
Lonnie
Emo Staff
admin
Pay no attention to the man behind the curtain
Posts: 6,999
|
Post by Lonnie on May 11, 2010 21:58:05 GMT -5
You are correct, Physics are physics and you can't change them. Thus you CAN make any adjustments with a graphic that can be done with a parametric. All EQs work the same way, period. The difference is that a parametric allows you to select the center freq, gain and sometimes Q. Where a graphic has set center freq and Qs. Bottom line is the phase shift and the way they work are identical in every regard. If you have a null point in your room there isn't an EQ in the world that can fix it. Lonnie So the bottom line is that graphic and parametric EQs are pointless because they can't correct room modes? If that would be true, one has to wonder why you implemented any EQs in the first place? Fortunately that is not true and parametric EQs can correct modal peaks (not dips) at low frequencies as long as level, center frequency and bandwidth are freely adjustible (graphic EQs don't provide that - a 1/3-octave EQ is too coarse). There are a couple of AES papers that describe just that (e.g. Celestinos/Nielsen). I guess you need to read up on it. There are several EQ solutions on the market (REW+BFD, Anti-Mode 8033) that work very well. I can post measurements of my own room any time. Best, Markus EQs at best can correct for small dips or peaks in the acoustic response which is why they were put in there. But you can't dump enough energy into a null to correct it. Now to make a blanket statement that graphic EQs are useless (as you have said many times) is blaintly incorrect. If you do a little research into the work of Dr. Jung, I beleive you will find that either one can acomplish the tast at hand. |
|
|
|
Post by dcibel on May 11, 2010 22:22:01 GMT -5
Just to add, there is a good reason why a single EQ cannot adequately correct for room modes. The problem is the room. If you take EMO-Q as an example, take a mic and measure the frequency response at a specific location in the room, then apply EQ so that the response is flat. What happens when you move a couple feet to the side of the mic? You are in a different location in the room, standing wave nulls and peaks have moved and so has their amplitude. Applying EQ to a single point does just that, EQ that specific point. Once you move out of that location the situation changes. The idea of EMO-Q is to give you a good idea of what corrections should be made to that single point, then you get to suit your personal needs.
What one could do is run EMO-Q in several listening locations in the room, and see how the results compare. Provided the measurements are accurate, it can give you an idea of how the sound changes from one point in the room to another.
I have studied a bit of Dr. Earl Geddes work, and believe that his solution to bass room modes is a good one. His solution is to use multiple subs placed around the room. Three should be good. Since you have multiple, they don't necessarily need to be big and powerful, as their sound will add together, each working less than one alone. Since they are in different location in the room, the room modes they generate will also be different, and the sound you hear will be a closer approximation of flat. I think that there may be more peaks and dips, but the difference in amplitude between the peaks and dips will be much less, resulting in better sound. More importantly better sound all over the room.
|
|
|
|
Post by markus on May 11, 2010 22:23:35 GMT -5
Lonnie So the bottom line is that graphic and parametric EQs are pointless because they can't correct room modes? If that would be true, one has to wonder why you implemented any EQs in the first place? Fortunately that is not true and parametric EQs can correct modal peaks (not dips) at low frequencies as long as level, center frequency and bandwidth are freely adjustible (graphic EQs don't provide that - a 1/3-octave EQ is too coarse). There are a couple of AES papers that describe just that (e.g. Celestinos/Nielsen). I guess you need to read up on it. There are several EQ solutions on the market (REW+BFD, Anti-Mode 8033) that work very well. I can post measurements of my own room any time. Best, Markus EQs at best can correct for small dips or peaks in the acoustic response which is why they were put in there. But you can't dump enough energy into a null to correct it. Now to make a blanket statement that graphic EQs are useless (as you have said many times) is blaintly incorrect. If you do a little research into the work of Dr. Jung, I beleive you will find that either one can acomplish the tast at hand. I've never said that dips can or should be corrected with EQs. Did you read that in any of my posts? What I've said is that modal peaks at low frequencies can be corrected with EQs. In order to do this, a parametric EQ is required. It would be a matter of luck if a graphic EQ would provide the exact frequency and bandwidth for correct peak attenuation. So my assessment of graphical EQs is correct for 99.9% of all real world cases. The EQs in the UMC-1 makes low frequency modal peak correction especially complicated because it provides EQs for single speakers only but what we need is a EQ for the summed low frequency channel. Additionally the EQs for the mains and surrounds are simply too coarse and they add in an unpredictable manner that depends on speaker distance settings. This peak requires an EQ at 49Hz but the UMC offers only 44 or 56Hz:  |
|
|
|
Post by mysterymachine on May 11, 2010 22:31:16 GMT -5
markus, I love your detailed posts but can't you and Lonnie agree to disagree before this turns into AVSF***m?  Doesn't it boil down to A) Nothing but room treatments can truely fix the issues B) Parametric EQs are better than graphic EQs at correcting room issues but that doesn't make graphic EQs worthless. Besides there is more to the UMCs EQs than fixing room issues - they can be used to tweak the sound to people's liking (hence the multiple EQs that you can save) having a perfectly flat frequency curve may not be everyone's cup of tea. |
|
|
|
Post by markus on May 11, 2010 22:31:19 GMT -5
|
|
|
|
Post by markus on May 11, 2010 22:34:05 GMT -5
Doesn't it boil down to A) Nothing but room treatments can truely fix the issues B) Parametric EQs are better than graphic EQs at correcting room issues but that doesn't make graphic EQs worthless. I would agree to that if it would be true but it's not. |
|
Lonnie
Emo Staff
admin
Pay no attention to the man behind the curtain
Posts: 6,999
|
Post by Lonnie on May 11, 2010 22:39:57 GMT -5
markus, I love your detailed posts but can't you and Lonnie agree to disagree before this turns into AVSF***m? Doesn't it boil down to A) Nothing but room treatments can truely fix the issues B) Parametric EQs are better than graphic EQs at correcting room issues but that doesn't make graphic EQs worthless. Besides there is more to the UMCs EQs than fixing room issues - they can be used to tweak the sound to people's liking (hence the multiple EQs that you can save) having a perfectly flat frequency curve may not be everyone's cup of tea. I agree, it is probably best to agree to disagree. I see Marcus's points and there are any number of papers/ test that have been done both for and against it. So before it gets out hand, I will just say that I understand what Marcus is saying, but I do not completely agree with it. To that end, everyone can chose to use whichever one works best for them. |
|
|
|
Post by markus on May 11, 2010 22:43:28 GMT -5
I agree, it is probably best to agree to disagree. I see Marcus's points and there are any number of papers/ test that have been done both for and against it. So before it gets out hand, I will just say that I understand what Marcus is saying, but I do not completely agree with it. To that end, everyone can chose to use whichever one works best for them. So we don't get a parametric EQ for the summed subwoofer channel? |
|
|
|
Post by mysterymachine on May 11, 2010 23:11:57 GMT -5
So we don't get a parametric EQ for the summed subwoofer channel? We do, its called the Anti-mode 8033 and with the way the euro's been weakening with the Greece meltdown it keeps getting cheaper and cheaper ;D (you never asked if we get it in the UMC) |
|
|
|
Post by darien87 on May 12, 2010 0:27:13 GMT -5
markus, I love your detailed posts but can't you and Lonnie agree to disagree before this turns into AVSF***m? Doesn't it boil down to A) Nothing but room treatments can truely fix the issues B) Parametric EQs are better than graphic EQs at correcting room issues but that doesn't make graphic EQs worthless. Besides there is more to the UMCs EQs than fixing room issues - they can be used to tweak the sound to people's liking (hence the multiple EQs that you can save) having a perfectly flat frequency curve may not be everyone's cup of tea. I agree, it is probably best to agree to disagree. I see Marcus's points and there are any number of papers/ test that have been done both for and against it. So before it gets out hand, I will just say that I understand what Marcus is saying, but I do not completely agree with it. To that end, everyone can chose to use whichever one works best for them. Thanks for chiming in here Lonnie. However all this arguing back and forth about with is better, GEQ or PEQ doesn't answer my question about why the sub EQ in the UMC-1 doesn't work. As I noted, I was trying to bring my 25hz spike down a bit by taking 22hz and 28hz all the way to -10. Nothing changed. Shouldn't my 25hz peak come down at least a little bit or am I missing something? |
|
|
|
Post by weeman on May 12, 2010 0:44:46 GMT -5
I agree, it is probably best to agree to disagree. I see Marcus's points and there are any number of papers/ test that have been done both for and against it. So before it gets out hand, I will just say that I understand what Marcus is saying, but I do not completely agree with it. To that end, everyone can chose to use whichever one works best for them. Thanks for chiming in here Lonnie. However all this arguing back and forth about with is better, GEQ or PEQ doesn't answer my question about why the sub EQ in the UMC-1 doesn't work. As I noted, I was trying to bring my 25hz spike down a bit by taking 22hz and 28hz all the way to -10. Nothing changed. Shouldn't my 25hz peak come down at least a little bit or am I missing something? the sub eq only works for the LFE .1 channel and has no effect when playing stereo sources. |
|
ratmice
Emo VIPs 
I'm not an actor, but I play one on TV.
Posts: 1,853
|
Post by ratmice on May 12, 2010 8:12:58 GMT -5
Markus - Wouldn't a graphic EQ with adjustments at 1hz intervals be just as good as a PEQ. You could approximate any PEQ curve with a system like that. What you mean is that , as generally implemented, GEQ is too coarse an adjustment to fix every room, it does not mean that PEQ is inherently better than GEQ - or that GEQ is useless. This nag is dead!
|
|
|
|
Post by markus on May 12, 2010 8:35:05 GMT -5
Markus - Wouldn't a graphic EQ with adjustments at 1hz intervals be just as good as a PEQ. You could approximate any PEQ curve with a system like that. What you mean is that , as generally implemented, GEQ is too coarse an adjustment to fix every room, it does not mean that PEQ is inherently better than GEQ - or that GEQ is useless. This nag is dead! If that GEQ would have adjustable Q (bandwidth), then it could work because this approximates a PEQ Have you ever seen a GEQ that is adjustable in 1Hz increments? A EQ needs to match exactly the frequency and bandwidth of a low frequency modal peak to correct it. The UMC provides EQs at 31.5, 63, 125, 190, 250, 500, 1000, 2200, 4500, 9000, 18000 Hz with a fixed Q for the mains. This can't work. If you don't believe me, then get REW and try it out for yourself. So, of what use is a GEQ in room acoustics? Create a room curve? Maybe. Fix high Q low frequency modal problems? No. |
|
|
|
Post by darien87 on May 12, 2010 10:15:59 GMT -5
Thanks for chiming in here Lonnie. However all this arguing back and forth about with is better, GEQ or PEQ doesn't answer my question about why the sub EQ in the UMC-1 doesn't work. As I noted, I was trying to bring my 25hz spike down a bit by taking 22hz and 28hz all the way to -10. Nothing changed. Shouldn't my 25hz peak come down at least a little bit or am I missing something? the sub eq only works for the LFE .1 channel and has no effect when playing stereo sources. Thanks for the info Weeman. You wouldn't happen to know where I can get a test disc that has tones specifically for the LFE channel? |
|
|
|
Post by markus on May 12, 2010 12:45:00 GMT -5
|
|
|
|
Post by loopinfool on May 12, 2010 12:46:18 GMT -5
You wouldn't happen to know where I can get a test disc that has tones specifically for the LFE channel? Check out my post near the bottom of the previous page in this thread. There are also a number of commercial "test discs" that have such tones. I think some of the "THX Optimizer" menus on some DVDs do, too. - LoopinFool |
|
|
|
Post by loopinfool on May 12, 2010 12:50:05 GMT -5
the sub eq only works for the LFE .1 channel and has no effect when playing stereo sources. Lonnie, Will it ever be possible for the UMC-1 to apply its EQs to the channel outputs rather than the inputs -- so this very issue (and others) can be resolved? "We're looking into it" or "We're waiting to hear back from Cirrus" are perfectly acceptable answers... ;D - LoopinFool |
|
How do you adjust a frequency with a GEQ for which you have no "slider" for?? Especially if you don't want to mess with the other freq's??
Doesn't it boil down to A) Nothing but room treatments can truely fix the issues B) Parametric EQs are better than graphic EQs at correcting room issues but that doesn't make graphic EQs worthless. 
