Thread: [Drc-fir-users] My new system (slightly OT)
Brought to you by:
dsbragio
From: Denis S. <d.s...@in...> - 2008-04-21 08:46:59
|
Hello, after six months of hard work, here are some pictures of my new system: http://servizi.infotecna.it/ListeningRoom/ This is not completely off topic for some simple reasons. Lets start from the beginning. The system has been built from the ground up with the "getting the best without equalization" target in mind. For example: 1) The room geometry has been optimized for uniform modal distribution, Bonello criteria, and so on. 2) The left wall is made of plasterboard filled with rockwool, so it behaves pretty much like a panel bass trap. 3) Same for the wood ceiling, though more rigid, so working a bit more in the mid-bass instead of the bass. 4) The room has been extensively damped, as can be seen from the pictures, leaving a gap between the adsorbers and the wall, to extend the absorption at the lower frequencies, "crossing" with the effect of the "bass trap like" elements. 5) The adsoprtion panels have been concentrated in the primary reflection points, basicly there's no single direct patch left untreated. 6) An heavy carpet (2.5 Cm thick) has ben placed on the floor. 7) Speakers with controlled uniform dispersion and flat direct sound have been adopted. 8) Corner placed subwoofers have been used, so that the room gets properly excited in its modal region. Many other details not worth to mention have been optimized, basicly everything has been done following the theory to make a good listening room. And indeed the results are excellent both from a measured and a listening point of view. For example just take a look at the reverberation time: http://servizi.infotecna.it/ListeningRoom/RT60.png Is is almost perfectly flat at 0.3 s above 100 Hz, as per the ITU and other guidelines. I won't spend much time describing the listening impressions, it's simply an excellent reproduction. Nevertheless, DRC is still there. It still makes a great improvement, it still makes the difference between an excellent reproduction and a "lifelike" one. I initially dreamed about optimizing the system up to the point of not needing DRC at all, but after a while I gave up. The test who finally convinced me that it was not worth the effort was a comparison between a minimum phase DRC correction and a mixed phase one. I already knew that with the "traditional" optimizations I couldn't get much more than what you get with a minimum phase correction, let alone that it is already a real challenge to get even that kind of fine tuning. So basicly a minimum phase correction was a good reference to get an idea of what could be done. Well, a mixed phase correction still makes a difference, even in a "no compromise" system like this. Of course the difference is much smaller than in an untreated room, but it's still clear, I wouldn't have any problem in identifying it in a DBT. The bass and mid bass is tighter and cleaner, making the difference. I think there's also some small difference in the mids, but I won't bet on this, this is much smaller. So now I decided to change direction. I will optimize the system in order to let DRC do it best. Here you can find some preliminary results (30 Mb): http://servizi.infotecna.it/ListeningRoom/drc-graphs.pdf You will notice some weird things in those measures. It's because I took the wrong files to do the graphs. This was an experimental correction with some weird parameters and when I realized it I had no more time to build the graphs again with the right files. Nevertheless it's enough to get a picture of the results. Just take a look at the 5 ms impulse response and you get an idea of how "clean" the results are. Now it's time to get back to the DRC development. ;) Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 09:44:16
|
Where is the big flat telly going to go? Ed W |
From: Denis S. <d.s...@in...> - 2008-04-21 09:57:35
|
On Mon, April 21, 2008 11:43, Ed W wrote: > Where is the big flat telly going to go? Audio only, sorry... :) -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Bernt <be...@ju...> - 2008-04-21 10:06:13
|
Congratulations on your new room Denis! IMO, nothing compares to a well-tuned system in a well-damped room. My room has an acoustically bad geometry, but the combination of damping and room correction works better than almost any room with long decay. Your room is probably acoustically more fit, but even though I have an idea of the improvements your hearing :-) FIY It takes forever to download the images so I skipped the middle part. Is there new hardware there too? The combination of acoustic treatment and DRC does bring with it a few new questions regarding the optimal mix of the two. Vennlig hilsen / kind regards Bernt -----Original Message----- From: drc...@li... [mailto:drc...@li...] On Behalf Of Denis Sbragion Sent: 21. april 2008 10:47 To: drc...@li... Subject: [Drc-fir-users] My new system (slightly OT) Hello, after six months of hard work, here are some pictures of my new system: http://servizi.infotecna.it/ListeningRoom/ This is not completely off topic for some simple reasons. Lets start from the beginning. The system has been built from the ground up with the "getting the best without equalization" target in mind. For example: 1) The room geometry has been optimized for uniform modal distribution, Bonello criteria, and so on. 2) The left wall is made of plasterboard filled with rockwool, so it behaves pretty much like a panel bass trap. 3) Same for the wood ceiling, though more rigid, so working a bit more in the mid-bass instead of the bass. 4) The room has been extensively damped, as can be seen from the pictures, leaving a gap between the adsorbers and the wall, to extend the absorption at the lower frequencies, "crossing" with the effect of the "bass trap like" elements. 5) The adsoprtion panels have been concentrated in the primary reflection points, basicly there's no single direct patch left untreated. 6) An heavy carpet (2.5 Cm thick) has ben placed on the floor. 7) Speakers with controlled uniform dispersion and flat direct sound have been adopted. 8) Corner placed subwoofers have been used, so that the room gets properly excited in its modal region. Many other details not worth to mention have been optimized, basicly everything has been done following the theory to make a good listening room. And indeed the results are excellent both from a measured and a listening point of view. For example just take a look at the reverberation time: http://servizi.infotecna.it/ListeningRoom/RT60.png Is is almost perfectly flat at 0.3 s above 100 Hz, as per the ITU and other guidelines. I won't spend much time describing the listening impressions, it's simply an excellent reproduction. Nevertheless, DRC is still there. It still makes a great improvement, it still makes the difference between an excellent reproduction and a "lifelike" one. I initially dreamed about optimizing the system up to the point of not needing DRC at all, but after a while I gave up. The test who finally convinced me that it was not worth the effort was a comparison between a minimum phase DRC correction and a mixed phase one. I already knew that with the "traditional" optimizations I couldn't get much more than what you get with a minimum phase correction, let alone that it is already a real challenge to get even that kind of fine tuning. So basicly a minimum phase correction was a good reference to get an idea of what could be done. Well, a mixed phase correction still makes a difference, even in a "no compromise" system like this. Of course the difference is much smaller than in an untreated room, but it's still clear, I wouldn't have any problem in identifying it in a DBT. The bass and mid bass is tighter and cleaner, making the difference. I think there's also some small difference in the mids, but I won't bet on this, this is much smaller. So now I decided to change direction. I will optimize the system in order to let DRC do it best. Here you can find some preliminary results (30 Mb): http://servizi.infotecna.it/ListeningRoom/drc-graphs.pdf You will notice some weird things in those measures. It's because I took the wrong files to do the graphs. This was an experimental correction with some weird parameters and when I realized it I had no more time to build the graphs again with the right files. Nevertheless it's enough to get a picture of the results. Just take a look at the 5 ms impulse response and you get an idea of how "clean" the results are. Now it's time to get back to the DRC development. ;) Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javao ne _______________________________________________ Drc-fir-users mailing list Drc...@li... https://lists.sourceforge.net/lists/listinfo/drc-fir-users |
From: Denis S. <d.s...@in...> - 2008-04-21 10:21:46
|
Hello Bernet, On Mon, April 21, 2008 12:05, Bernt wrote: > FIY It takes forever to download the images so I skipped the middle part. Is > there new hardware there too? it's on our server, connected by an ADSL line (i.e. 256 Kbits only), sorry. What do you mean by new hardware? Almost anything there is new. The only parts that have been preserved from the previous system are the CD Player and the convolver PC. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Denis S. <d.s...@in...> - 2008-04-21 10:35:25
|
Hello Bernt, On Mon, April 21, 2008 12:05, Bernt wrote: > IMO, nothing compares to a well-tuned system in a well-damped room. My room > has an acoustically bad geometry, but the combination of damping and room > correction works better than almost any room with long decay. ... sorry, I overlooked this. Don't get me wrong: I was not saying that the passive treatment isn't important. If I had to choose between an undamped room with DRC and a propelry damped room without DRC I would choose the second one, without the slightest esitation. The room still provides the greatest improvement, no question about this. What I was trying to say is that the common belief that with a properly damped room you can completely avoid equalization in my opinion is wrong. Even when everything is tuned down following closely everything needed to get a good listening room, equalization still has its role, and it isn't a marginal one. Furthermore now I'm convinced that to get the real best you can't simply go for minimum phase equalization, like the one provided by a good parametric equalizer. A complete mixed phase equalization is still needed, and the benefits are still worth the effort even in a properly tuned room, not only in the typical average listening room. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 10:42:59
|
How are you crossing over the bass to the Genelecs? Most of the Gs have a steep 40Hz active highpass filter I think? Did you disable this or simply design a lowpassfilter for the bass which was complementary? There is still a bit of group delay in the bass right? Is this more than you would expect for what appears to be a sealed bass sub system? How much of it is a function of the crossovers in use though? Looks nice - interested to see some more measurements! Ed W |
From: Denis S. <d.s...@in...> - 2008-04-21 11:07:08
|
Hello Ed, On Mon, April 21, 2008 12:42, Ed W wrote: > How are you crossing over the bass to the Genelecs? Most of the Gs have it's a simple 80 Hz LR4, choosen more for dynamic range considerations than anything else. The dynamic limit of the Gens woofer is reached around 60 Hz, so I stayed a bit above it to avoid any low end compression by the Gens woofer. This way the system should be able to provide > 120 dB SPL from 10 Hz up. > a steep 40Hz active highpass filter I think? Did you disable this or > simply design a lowpassfilter for the bass which was complementary? It's at 33 Hz for the 1038. I didn't disable it. Considering that the XOver is more than an octave above, it doesn't cause any trouble. Most of the problems in this region are still caused by the room, not the speakers. At 80 Hz you are deep in the middle of the modal region of the room, and the room modes dominate any speaker issue by a wide margin. > There is still a bit of group delay in the bass right? Is this more > than you would expect for what appears to be a sealed bass sub system? > How much of it is a function of the crossovers in use though? For sure the complex interaction between the XOver, the subsonic filter, the vented tuning of the Gens and the sealed tuning of the sub is causing all sort of phase anomalies in this region. But again this is nothing compared to the long tails caused by the room modes, which are far worse than any XOver/speaker issues, despite the excellent damping of the room as proven by the RT60. You can see this with an RT60 computed with an higher frequency resolution: http://servizi.infotecna.it/ListeningRoom/RT60-HR.png This starts to gets affected by the measurement noise, so it's less reliable, take it with cautions, but the spikes you see in the 60-400 Hz range are caused by the room resonances. This are far more problematic than speakers problems. The only way to compensate them would be the use of some finely tuned resonator, but this is beyond what I can do in my room. Considering that I have not even a way to measure those issues, because they get buried in the room anomalies, I simply selected the xover settings needed to let DRC work at its best. I'm still unsatisfied from this point of view, I think there's still some margin for improvements. > Looks nice - interested to see some more measurements! As soon as I finish the fine tuning. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Uli B. <uli...@gm...> - 2008-04-21 11:08:45
|
Denis, 1. can you downsize the pictures as > 2 MB is not really necessary to view with a normal screen. This will save a lot of time. Also 30 MB for the pdf is pretty big. 2. Can you also put the pulse responses on your server for download? Anyway congrats for your new room. I'm envy. Uli |
From: Denis S. <d.s...@in...> - 2008-04-21 11:14:47
|
Hello Uli, On Mon, April 21, 2008 13:08, Uli Brueggemann wrote: > 1. can you downsize the pictures as > 2 MB is not really necessary to view > with a normal screen. This will save a lot of time. Also 30 MB for the pdf > is pretty big. > > 2. Can you also put the pulse responses on your server for download? bit busy now, I'll try to do it in the afternoon... Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 11:32:46
|
Uli Brueggemann wrote: > Denis, > > 1. can you downsize the pictures as > 2 MB is not really necessary to > view with a normal screen. This will save a lot of time. Also 30 MB > for the pdf is pretty big. I mirrored denis's files here: http://www.duffroomcorrection.com/public/ Bandwidth is quite large on that site... Ed W |
From: Denis S. <d.s...@in...> - 2008-04-21 11:46:11
|
On Mon, April 21, 2008 13:32, Ed W wrote: > Bandwidth is quite large on that site... Thanks, our DSL link was crying for relief... :) -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Denis S. <d.s...@in...> - 2008-04-21 11:42:00
|
Hello Uli, On Mon, April 21, 2008 13:08, Uli Brueggemann wrote: > 1. can you downsize the pictures as > 2 MB is not really necessary to view > with a normal screen. This will save a lot of time. Also 30 MB for the pdf > is pretty big. here they are, some smaller versions: http://servizi.infotecna.it/ListeningRoom/Small/ > 2. Can you also put the pulse responses on your server for download? http://servizi.infotecna.it/ListeningRoom/IR.png Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Uli B. <uli...@gm...> - 2008-04-21 11:45:29
|
Hi Denis, sorry, I've not been precise. I did not mean a picture but the uncorrected pulse response data (pcm-file) :) Uli On Mon, Apr 21, 2008 at 1:41 PM, Denis Sbragion <d.s...@in...> wrote: > Hello Uli, > > On Mon, April 21, 2008 13:08, Uli Brueggemann wrote: > > 1. can you downsize the pictures as > 2 MB is not really necessary to > view > > with a normal screen. This will save a lot of time. Also 30 MB for the > pdf > > is pretty big. > > here they are, some smaller versions: > > http://servizi.infotecna.it/ListeningRoom/Small/ > > > 2. Can you also put the pulse responses on your server for download? > > http://servizi.infotecna.it/ListeningRoom/IR.png > > > Bye, > > -- > Denis Sbragion > InfoTecna > Tel: +39 0362 805396, Fax: +39 0362 805404 > URL: http://www.infotecna.it > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > _______________________________________________ > Drc-fir-users mailing list > Drc...@li... > https://lists.sourceforge.net/lists/listinfo/drc-fir-users > |
From: Denis S. <d.s...@in...> - 2008-04-21 11:48:09
|
Hello Uli, On Mon, April 21, 2008 13:45, Uli Brueggemann wrote: > sorry, I've not been precise. > I did not mean a picture but the uncorrected pulse response data (pcm-file) > :) ops, sorry. :) I don't have them with me, I left them at home. I'll place them on-line tomorrow. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 13:18:38
|
What are you using as a sub? Looks like a smallish driver (12"?) in sonotube equivalent? Difficult to get scale though! No room for an IB then? I should imagine that the sub integrates well with that large woofer on the genelec (15"?). A possible problem with my setup is that I only have a 5" woofer on my mains cutting in to a 15" based IB, so directivity is quite different... :-( One idea on your system might be to run the woofers up quite a bit higher, eg >100Hz should be possible with a stereo amp - you might get quite good results Ed W |
From: Denis S. <d.s...@in...> - 2008-04-21 13:32:03
|
Hello Ed, On Mon, April 21, 2008 15:03, Ed W wrote: > What are you using as a sub? Looks like a smallish driver (12"?) in > sonotube equivalent? Difficult to get scale though! no, they are 15". Exodus Audio Tempest-X: http://www.diycable.com/main/product_info.php?cPath=24_93&products_id=652 XBL^2 motor, 27 mm XMax, they're a bargain for their price. A bit of a nightmare to ship them from the US, which also doubled the price, but it's been worth the effort. Consider that the distortion is so low that gets buried in the room noise and I haven't been able to measure it. As the driver name suggest these are the same people of Adire Audio. After Adire ceased operation they resurrected with this new line of subs. The cabinet is made from duct pipes normally used for water ducts, filled with polyester fiber to avoid resonance problems. They are polyethylene pipes, about 25 mm thick. Each of the two arrays is around 300 Kg, and it is completely dead. The coupling between the arrays helps further reducing any residual vibration. The whole sub is 1.8 m tall and more than 3.2 m large, quite big. > No room for an IB then? I had to abandon the IB idea. I tried placing a normal sub where the IB mouth was, and the results were unacceptable. There was an enormous hole around 30 Hz, that no EQ would have been able to fill. Add to this the fact the the mouth was behind the listening position, with all its load of further problems, and you see why I resorted to a "standard" (well, not so standard :) solution. > I should imagine that the sub integrates well with that large woofer on > the genelec (15"?). A possible problem with my setup is that I only > have a 5" woofer on my mains cutting in to a 15" based IB, so > directivity is quite different... :-( Yes, no coupling problems. Everything has been designed to integrate pretty well. I was more concerned about the dynamic issues than directivity, because I wanted to achieve greater than 120 dB SPL at all frequencies. Having the subs directly behind the mains made this much more easy, allowing for an almost free selection of the crossover frequency. With the sub on the back I was forced to use a really low xover frequency (< 60 Hz). > One idea on your system might be to run the woofers up quite a bit > higher, eg >100Hz should be possible with a stereo amp - you might get > quite good results The sub is already stereo driven by a QSC RMX4050HD (1.400 W/Ch, 2.200 W/Ch peak). I want to try experimenting a bit with the crossover frequencies to see which ones provides the best results. For example without DRC I got the best results using a 160 Hz xover, with the sub level about 10 dB below the main levels. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 14:25:58
|
Hi > no, they are 15". Exodus Audio Tempest-X: > > http://www.diycable.com/main/product_info.php?cPath=24_93&products_id=652 > Jeepers! I have previous generation Tempests and they are certainly good drivers. Actually they only need about 150W to get them to bang the stops so your amp is quite well rated! Would need to think about it a bit more, but I wouldn't have thought you have enough of them to hit 120dB at 10Hz though (at least not in closed box form?). Not to discredit that awesome bunch of speakers though - certainly they will go waaaay louder than most people need... It's quite fun looking for stuff which goes loud and deep enough to test these woofers. I have 4x in IB form - distortion is incredibly low and you can fire them up at 10Hz and hear.... Nothing at all... However, after about 45 seconds you have a nasty headache and feel rather quesy and it's time for some fresh air... I have double glazed windows and they do flex quite a lot... However, I last setup my system with the default DRC templates and these target a declining target at 10Hz. If you remember we discussed bottoming out at 10Hz hence the reason for the default targets. I guess with your system you are now pretty happy that this is difficult with low of high spec drivers!! So I can probably get a lot more out of my setup if I tune my targets a bit more. Also I suspect that my ECM 8000 doesn't measure down <20Hz with decent flatness... > I had to abandon the IB idea. I tried placing a normal sub where the IB mouth > was, and the results were unacceptable. There was an enormous hole around 30 > Hz, that no EQ would have been able to fill. This is because of room modes? Are you finding that having 6 drivers is giving you much better node excitation in your current setup? > The sub is already stereo driven by a QSC RMX4050HD (1.400 W/Ch, 2.200 W/Ch > peak). I want to try experimenting a bit with the crossover frequencies to see > which ones provides the best results. For example without DRC I got the best > results using a 160 Hz xover, with the sub level about 10 dB below the main > levels. > Yeah - this is what Lyngdorf advocates - woofers driven well and time aligned can often go a lot higher than people think and take a ton of weight off the main drivers (improving them quite a lot). 160Hz seems impressively high for such a widely dispersed driver array. Lyngdorf go up to about 300+Hz on their demo setups though, so definitely it's a workable design! Good luck! Ed W |
From: Denis S. <d.s...@in...> - 2008-04-21 15:06:11
|
Hello Ed, On Mon, April 21, 2008 16:25, Ed W wrote: > Jeepers! I have previous generation Tempests and they are certainly > good drivers. Actually they only need about 150W to get them to bang > the stops so your amp is quite well rated! in a closed box more than 500 W are needed. The amp I used has exactly the right peak power to push the drivers 0.5 mm below the XMax, at least in the simulations. > Would need to think about it a bit more, but I wouldn't have thought you > have enough of them to hit 120dB at 10Hz though (at least not in closed > box form?). In fact they aren't enough, they reach something like 113.5 dB in free air (again, simulations, with all the eight drivers pushed to the limits). But then there's the 1/8 pi emission and the room gain (at 10 Hz you are well within the pressure region), so their able to reach the target in room. > It's quite fun looking for stuff which goes loud and deep enough to test > these woofers. I have 4x in IB form - distortion is incredibly low and > you can fire them up at 10Hz and hear.... Nothing at all... However, > after about 45 seconds you have a nasty headache and feel rather quesy > and it's time for some fresh air... I have double glazed windows and > they do flex quite a lot... Nevertheless, I still think that bass extension down to infrasonic frequencies is helpful for transient reproduction. For sure much more useful than trying to reproduce > 20 KHz frequencies. Well, I agree, 120 dB at 10 Hz is quite an exageration, but this is another story. I wanted something able to reproduce anything that the "ear" (including the skin) is able to "sense". May be I will never find a recording needing this kind of performances in all of my life. Nevertheless I wanted to be sure that if this ever happens my system is ready for it. :) So far I haven't even been able to find a record that makes the speaker cones moving in a visible way. :) > However, I last setup my system with the default DRC templates and these > target a declining target at 10Hz. If you remember we discussed > bottoming out at 10Hz hence the reason for the default targets. I guess Yep, I know. Another advantage of closed subs is that they self protect against excessive excursion at the lowest frequencies. I could even drop the infrasonic filter I used to include with my old vented sub. Of course not a big problem anyway, just a matter of adding a steep sub filter > with your system you are now pretty happy that this is difficult with > low of high spec drivers!! So I can probably get a lot more out of my > setup if I tune my targets a bit more. Also I suspect that my ECM 8000 > doesn't measure down <20Hz with decent flatness... You have to check. With my mic and my sub I can see the sweep emerge from the room noise since 5 Hz. > This is because of room modes? Because of the mouth position, which is about 1.5 m away from the closest corner. This caused an improper excitation of the room modes. It was already clear from simulations, but I hoped that the tilted ceiling would change things enough to avoid the problem. Unfortunately at such low frequencies it doesn't make much difference. The simulations where right, as many acoustics experts already warned me. Nevertheless I had to try before giving up. The hole in the wall was already there, I had to check before closing it. > Are you finding that having 6 drivers is giving you much better node > excitation in your current setup? They are 8, in two 4 elements array. BTW, didn't find much difference WRT my old sub with respect to this. I think that to get the benefits of multiple subs you have to place them much more spaced away in some key points of the room (e.g. Floyd Toole arrangement). > 160Hz seems > impressively high for such a widely dispersed driver array. Lyngdorf go > up to about 300+Hz on their demo setups though, so definitely it's a > workable design! Alwasy remember that you're not in free air. At this frequencies the room dominates. Talking about "dispersed array drivers" make little sense when you have four walls around at no more than 1 m from the drivers, and 2 of them about 20 cm away, like in my situation. In my case it worked so well simply because the response below 200 Hz was about 8 dB above the response at higher frequencies. So crossing at 160 Hz and reducing the sub level provided a reasonable equalization for this behaviour. Still quite far from what DRC can do though. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Ed W <li...@wi...> - 2008-04-21 15:42:57
|
Hi > Well, I agree, 120 dB at 10 Hz is quite an > exageration, but this is another story. I wanted something able to reproduce > anything that the "ear" (including the skin) is able to "sense". May be I will > never find a recording needing this kind of performances in all of my life. > Nevertheless I wanted to be sure that if this ever happens my system is ready > for it. :) So far I haven't even been able to find a record that makes the > speaker cones moving in a visible way. :) > You really need to plug in a big tv! All new films are now stretching the bass range. I used to have to search out films with the odd subsonic boom (eg Hitch hikers guide to the galaxy which has an 8Hz explosion - sounds like a gentle pop on most systems...). However, most new films are stretching the bass in extreme new ways. With something like transformers at moderate volume I get my sofa rattling and quite a lot of the room excited during most of the fight scenes... Very immersive I don't think music really stretches the bass all that much - certainly not instrumental music (excluding organ perhaps) Ed W |
From: Bernt R. <ber...@ly...> - 2008-04-21 14:10:42
|
Denis, I didn't interpret you as saying room tuning isn't important. I interpreted you as saying both are important. And I agree :-) I guess I forgot to mention that my room is heavily damped. Rt60 inside 0.3s below 100 hz. But I compromised with the "family acceptance factor" and made all the damping in the ceiling, so I still have side wall reflections. Short decay improves the signal noise ratio of the room and therefore the audio playback, and the result is a more dynamic, yet more quiet presentation. And short decay in the bass (the hard part ;-) ) does provide a more spacious bass reproduction and provides more breathing space for the rest of the bandwidth too. Not quite as in a large space, but somewhat in that direction. But I am not convinced that the same room treatment scheme is required for a corrected system as for an uncorrected systems. For instance I do not hear a significant difference (perhaps none at all) my system whether I put up some 60x120x5 cm rockwool acoustic panels with 5 cm air behind them to cancel the first side wall reflections - as long as it is corrected. I have a pragmatic view of room tuning vs dsp. Some rooms will never sound really good no matter how much dsp is applied. But there are other rooms & systems that responds very well - even very lively rooms. Uli's system sounds very very good even though there is lot of energy bouncing around. I see a lot of damping materials on the pictures. Did you consider using diffusers? I haven't experimented with diffusers myself but it seems to be a big hit in Sweeden and partly in Norway. Vennlig hilsen / kind regards Bernt -----Original Message----- From: drc...@li... [mailto:drc...@li...] On Behalf Of Denis Sbragion Sent: 21. april 2008 12:35 To: General discussion about DRC Subject: Re: [Drc-fir-users] My new system (slightly OT) Hello Bernt, On Mon, April 21, 2008 12:05, Bernt wrote: > IMO, nothing compares to a well-tuned system in a well-damped room. My room > has an acoustically bad geometry, but the combination of damping and room > correction works better than almost any room with long decay. ... sorry, I overlooked this. Don't get me wrong: I was not saying that the passive treatment isn't important. If I had to choose between an undamped room with DRC and a propelry damped room without DRC I would choose the second one, without the slightest esitation. The room still provides the greatest improvement, no question about this. What I was trying to say is that the common belief that with a properly damped room you can completely avoid equalization in my opinion is wrong. Even when everything is tuned down following closely everything needed to get a good listening room, equalization still has its role, and it isn't a marginal one. Furthermore now I'm convinced that to get the real best you can't simply go for minimum phase equalization, like the one provided by a good parametric equalizer. A complete mixed phase equalization is still needed, and the benefits are still worth the effort even in a properly tuned room, not only in the typical average listening room. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javao ne _______________________________________________ Drc-fir-users mailing list Drc...@li... https://lists.sourceforge.net/lists/listinfo/drc-fir-users |
From: Denis S. <d.s...@in...> - 2008-04-21 14:35:32
|
Hello Bernt, On Mon, April 21, 2008 16:10, Bernt Rønningsbakk wrote: > But I am not convinced that the same room treatment scheme is required for a > corrected system as for an uncorrected systems. For instance I do not hear a ... yep, I agree. Having a DSP at hand changes the perspective. > I see a lot of damping materials on the pictures. Did you consider using > diffusers? I haven't experimented with diffusers myself but it seems to be a > big hit in Sweeden and partly in Norway. Yes, the original plans included diffusers on the rear wall (i.e. wall behind the listener). So far I can't use them because temporarily there's the bed there. So I resorted to the solution you see in the pictures, i.e. absorbers about 1 m around the listening position with a free area, about twice the surface, around it. This prevents direct bounces from the rear wall but allows some "leakage" of sound from the back, mimicking somewhat the behaviour of a diffusive wall. It seems to work pretty well, even though it isn't the optimal solution. Furthemore, the absorbing material seems so much because it is placed almost completely on one side of the room, somewhat following the LEDE approach and to intercept first reflections. But the room is much taller on the other side, something like 4.3 m high, so the total surface occupied by damping material isn't that much, about 15-20% of the whole room internal surface (I don't remember the exact numbers). Else I would get an anechoic chamber instead of a listening room. :) Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Denis S. <d.s...@in...> - 2008-04-28 11:12:09
|
Hello, On Mon, April 21, 2008 10:46, Denis Sbragion wrote: > So now I decided to change direction. I will optimize the system in order to > let DRC do it best. Here you can find some preliminary results (30 Mb): > > http://servizi.infotecna.it/ListeningRoom/drc-graphs.pdf > > You will notice some weird things in those measures. It's because I took the > wrong files to do the graphs. further refinements, changed a bit the xover settings, changed the sub levels. Here are the final measurements: http://servizi.infotecna.it/ListeningRoom/Misure/ Not much difference, just some refinements and the absence of the anomlies I introduced by mistake the first time. Bye, -- Denis Sbragion InfoTecna Tel: +39 0362 805396, Fax: +39 0362 805404 URL: http://www.infotecna.it |
From: Gregory M. <gma...@gm...> - 2008-04-29 07:17:38
|
On Mon, Apr 21, 2008 at 4:46 AM, Denis Sbragion <d.s...@in...> wrote: > Hello, > > after six months of hard work, here are some pictures of my new system: > > http://servizi.infotecna.it/ListeningRoom/ > > This is not completely off topic for some simple reasons. Lets start from the > beginning. I noticed your speakers are on casters .. have you tried rolling them to the center and computing a crosstalk canceled set of filters for them? |
From: Mihail Z. <mih...@gm...> - 2008-04-30 02:35:50
|
On Tue, 29 Apr 2008 03:17:28 -0400 "Gregory Maxwell" <gma...@gm...> wrote: > On Mon, Apr 21, 2008 at 4:46 AM, Denis Sbragion <d.s...@in...> wrote: > > Hello, > > > > after six months of hard work, here are some pictures of my new system: > > > > http://servizi.infotecna.it/ListeningRoom/ > > > > This is not completely off topic for some simple reasons. Lets start from the > > beginning. > > I noticed your speakers are on casters .. have you tried rolling them > to the center and computing a crosstalk canceled set of filters for > them? Few weak ago, i try "hardware" crosstalk cancellation. I setup speaker like it recommended on ambiophonic.org and put barrier (matеress :)) betwen it ... And it change my life :) I want say big thanks for share you experience year ago: >Ah, well .. crosstalk cancellation is something that I've experimented >with for a long time. >At least in my experience, once you've used it.. it is hard to go back >to plane stereo. Personally I find that the transition from Stereo -> >cross talk canceled stereo is more substantial than the differences >between stereo and mono. But just like going from mono to stereo there >are additional challenges. Speaker placement is far more sensitive, >speaker uniformity more critical, listening position more limited... >Increased risks of comb filtering artifacts, etc.. >I think it's worth it: with a properly crosstalk canceled system the >speakers just vanish.. there is a transparency which I've never found >with regular stereo. Not the sort of muddy haze of sound that you get >from typical "hifi" stereo-widening, but clear localizable sound with >distinctive and accurate locations far beyond the speaker arc. As >one of my victims^wtest subjects once said: "It is very easy to >imagine the orchestra in front of you, as though reality were the >illusion". I have same impression, it realy great! Most impressive Pink Floyd - The Dark Side Of The Moon and King Crimson - The Power To Believe. "Hardware" crosstalk cancellation much less critical to listener position and gives great result with drc. And you absolutely right - i never can't return to ordinary stereo. |