If you're not opposed to performing all of the eq in the digital domain, then I would suggest just doing it all with DRC. You'll probably need to set PLMaxGain to the 3-4 level (I've never set it above 2).
Alan Jordan's DRC Designer includes a measurement function. Once you run it you will notice a script appear in the main folder called drcWrapperRecordSweep.bat which you can edit (if you'd like) for range of freq and time (I've gotten good results with sweeps as short as 10 seconds). You should try your luck with a single measurement taken from the sweet spot. Aim the mic up at about a 45 deg angle and try to make the loudest peak of the two sweeps hit just under -6db. I know this could be done in REW as well, but I've never done it that way....... perhaps xrk971 could offer some advice on that?
The erb config file is interesting as the window is quite short at 20hz, but notice it rams up very steeply from there (exponents of 1.85) and surpasses even the normal config file in the upper midrange and treble window length. The bass will appear lumpier with the erb file than with the minimal config file when the final response is graphically analyzed, but you would probably be surprised by how good it sounds.
I'd love to tell you to save your precious time and just use what I have found to work rather than experiment with all the possibilities but you know as well as I that it's the best way to learn. So, have fun and don't hesitate with any concerns or questions.
Alan Jordan's DRC Designer includes a measurement function. Once you run it you will notice a script appear in the main folder called drcWrapperRecordSweep.bat which you can edit (if you'd like) for range of freq and time (I've gotten good results with sweeps as short as 10 seconds). You should try your luck with a single measurement taken from the sweet spot. Aim the mic up at about a 45 deg angle and try to make the loudest peak of the two sweeps hit just under -6db. I know this could be done in REW as well, but I've never done it that way....... perhaps xrk971 could offer some advice on that?
The erb config file is interesting as the window is quite short at 20hz, but notice it rams up very steeply from there (exponents of 1.85) and surpasses even the normal config file in the upper midrange and treble window length. The bass will appear lumpier with the erb file than with the minimal config file when the final response is graphically analyzed, but you would probably be surprised by how good it sounds.
I'd love to tell you to save your precious time and just use what I have found to work rather than experiment with all the possibilities but you know as well as I that it's the best way to learn. So, have fun and don't hesitate with any concerns or questions.
The amount of control is ridiculous and it looks like one of the things Jordan's gui tries to do is prevent you from screwing around too much with parameters. Personally I don't bother with any of the available guis since I find it easy enough to just edit the config files and then generate filters with scripts. Two things I have experimented with are bypassing the (optional) ringing truncation stage and changing the peak limiting type to log weighted linear phase. The former might help flatten the freq response (I noticed this much more with my previous system which for some reason I was having trouble getting flat enough) and the latter may help improve the phase response in the bass region.
Unfortunately, I don't know how to make DRC Designer work with the newer version of DRC. Once you generate filters however, a couple of scripts will appear in the main folder which can be edited. You can drop the 3.2.1 folder into the DRC Designer folder and aim the scripts at the newer DRC folder (I think the config files are named slightly differently in the newer version well so this change needs to be reflected in the script as well) but you'll have to run them by clicking directly on them so no more DRC Designer. You should rename the scripts if you make changes to them because DRC Designer will re-write them if you use it to generate filters again. I could just share the scripts I use for 3.2.1 if that would make things easier.
Unfortunately, I don't know how to make DRC Designer work with the newer version of DRC. Once you generate filters however, a couple of scripts will appear in the main folder which can be edited. You can drop the 3.2.1 folder into the DRC Designer folder and aim the scripts at the newer DRC folder (I think the config files are named slightly differently in the newer version well so this change needs to be reflected in the script as well) but you'll have to run them by clicking directly on them so no more DRC Designer. You should rename the scripts if you make changes to them because DRC Designer will re-write them if you use it to generate filters again. I could just share the scripts I use for 3.2.1 if that would make things easier.
I'll find my way around it then, no problem (I think 
). I also have the older scripts from Alan. If I somehow don't succeed I know where to come for help.
First I need to finish a first reflection point damping panel. Then on to the measurements. I'll be taking my time to learn. Probably going to ask more questions like: How do you record square waves in REW?.
). I also have the older scripts from Alan. If I somehow don't succeed I know where to come for help. First I need to finish a first reflection point damping panel. Then on to the measurements. I'll be taking my time to learn. Probably going to ask more questions like: How do you record square waves in REW?
.
Yesterday I got my first afternoon of playing with both REW and DRC. Had some fun in the process but I need to get better measurements. I had the speakers way too loud after setting the right SPL with the REW calibration measurement. I guess the line array is a bit more efficient in the lower mid than the calibration test anticipated. I didn't have a lot of time and went on with those loud measurements. Lots of ambient noise during the day made it a bad moment to be testing.
First played with REW and it's Auto EQ function. Imaging tightened up and it really sounds quite well. Way better than the generic curve I made previously.
Next I played with DRCDesigner. At first I was really disappointed. The vocals were less focussed and the REQ PEQ was quite a bit better. I didn't trust the measurements I made with DRCDesigner so I exported the REQ impulse responses and converted them to 32 bit float. The trouble is not seeing what is happening, at least REW gives you that!
Using the results from that sounds quite promising. The focus was back and all in all a pleasing sound. Now I need to let is rest for a bit, having heard so many versions I can't tell witch one I like anymore. First I need to get proper measurements. That in itself is pretty hard to do because I live in a very busy street. Due to a blocked road parallel to ours there's much more traffic than normal. Measuring at night won't be an option. I'll try again next week or so and maybe use Audiolense for the measurement suite. That way I have the option to record directly to 32 bit float.
No pretty graphs yet, maybe next time. I'll first need some sane measurements before I come to any sort of conclusion (lol).
First played with REW and it's Auto EQ function. Imaging tightened up and it really sounds quite well. Way better than the generic curve I made previously.
Next I played with DRCDesigner. At first I was really disappointed. The vocals were less focussed and the REQ PEQ was quite a bit better. I didn't trust the measurements I made with DRCDesigner so I exported the REQ impulse responses and converted them to 32 bit float. The trouble is not seeing what is happening, at least REW gives you that!
Using the results from that sounds quite promising. The focus was back and all in all a pleasing sound. Now I need to let is rest for a bit, having heard so many versions I can't tell witch one I like anymore. First I need to get proper measurements. That in itself is pretty hard to do because I live in a very busy street. Due to a blocked road parallel to ours there's much more traffic than normal. Measuring at night won't be an option. I'll try again next week or so and maybe use Audiolense for the measurement suite. That way I have the option to record directly to 32 bit float.
No pretty graphs yet, maybe next time
. I'll first need some sane measurements before I come to any sort of conclusion (lol).
Yesterday I got my first afternoon of playing with both REW and DRC. Had some fun in the process but I need to get better measurements. I had the speakers way too loud after setting the right SPL with the REW calibration measurement. I guess the line array is a bit more efficient in the lower mid than the calibration test anticipated. I didn't have a lot of time and went on with those loud measurements. Lots of ambient noise during the day made it a bad moment to be testing.
First played with REW and it's Auto EQ function. Imaging tightened up and it really sounds quite well. Way better than the generic curve I made previously.
Next I played with DRCDesigner. At first I was really disappointed. The vocals were less focussed and the REQ PEQ was quite a bit better. I didn't trust the measurements I made with DRCDesigner so I exported the REQ impulse responses and converted them to 32 bit float. The trouble is not seeing what is happening, at least REW gives you that!
Using the results from that sounds quite promising. The focus was back and all in all a pleasing sound. Now I need to let is rest for a bit, having heard so many versions I can't tell witch one I like anymore. First I need to get proper measurements. That in itself is pretty hard to do because I live in a very busy street. Due to a blocked road parallel to ours there's much more traffic than normal. Measuring at night won't be an option. I'll try again next week or so and maybe use Audiolense for the measurement suite. That way I have the option to record directly to 32 bit float.
No pretty graphs yet, maybe next time
You wil need to do a lot of measurements a lot of different filtering and a lot of target curve changing before you come close to the best result.
The learning curve is pretty steep, a lot of different parameters to play with. No matter what software you use.
The only way to get through is to get you fingers dirty in the machineroom
But it is worth it (and you will probably never be 100% finished
)Koldby
Hahaha, I know! But it is rewarding work and I hear a lot of potential already. It's fun to browse trough all my old favourites in the mean time and discovering new ones.
As suspected, the first reflections of the back wall are quite easy to spot in the impulse. Need to figure out an elegant way to deal with that. Depth is still lacking while I gained width from the side damping panel behind the curtain. I guess I can cook up a few mobile panels to use for critical listening... Oh no! there I go again
As suspected, the first reflections of the back wall are quite easy to spot in the impulse. Need to figure out an elegant way to deal with that. Depth is still lacking while I gained width from the side damping panel behind the curtain. I guess I can cook up a few mobile panels to use for critical listening... Oh no! there I go again
Yes, but well .... First reflections and standing wave pattern in such a small space... I really think you need to look into diffusors and absorbers..... ouch, here we go again
So what was the problem exactly with the measurements done with DRC designer? How did they look compared to the ones made with REW?
I don't have experience using REW for measurements so I can only advise on the measurement tool used in conjunction with DRC designer. All that matters in the end is the SNR of the measurement and not the SPL of the system during the measurement. The protocall I use is:
---Set the system volume level to the loudest safe/comfortable level for use with a sine sweep of -6db (digital level). For this purpose you can use Audacity to create and play a sine sweep (create a fullrange "chirp" with a level of 0.5) or just run the DRC Designer measurement function starting with a low volume level and gradually increasing to the maximum level that you or the system can comfortably tolerate.
---Set the recording level (for me it's a mic preamp/usb interface) so that the peak recorded level of the loudest channel is no more than -6 db.
You should probably split the measurement script into two so that that you can measure one speaker at a time. This will cut in half the amount of time you need between cars driving by. I have found a 10 second sweep to be adequate even with my low-output system.
Once you have good measurements you can breathe a sigh of relief. From that point forward we can talk about correcting/viewing the response using DRC, Audacity and REW.
I don't have experience using REW for measurements so I can only advise on the measurement tool used in conjunction with DRC designer. All that matters in the end is the SNR of the measurement and not the SPL of the system during the measurement. The protocall I use is:
---Set the system volume level to the loudest safe/comfortable level for use with a sine sweep of -6db (digital level). For this purpose you can use Audacity to create and play a sine sweep (create a fullrange "chirp" with a level of 0.5) or just run the DRC Designer measurement function starting with a low volume level and gradually increasing to the maximum level that you or the system can comfortably tolerate.
---Set the recording level (for me it's a mic preamp/usb interface) so that the peak recorded level of the loudest channel is no more than -6 db.
You should probably split the measurement script into two so that that you can measure one speaker at a time. This will cut in half the amount of time you need between cars driving by. I have found a 10 second sweep to be adequate even with my low-output system.
Once you have good measurements you can breathe a sigh of relief. From that point forward we can talk about correcting/viewing the response using DRC, Audacity and REW.
Yes, but well .... First reflections and standing wave pattern in such a small space... I really think you need to look into diffusors and absorbers..... ouch, here we go again
First reflections? True, standing waves? Not that much of a problem. Having the array in the corners seems to work really well. No overpowering lows, no complaints yet from me. The arrays are way more smooth on the bottom end than my previous speakers with 15" drivers. They do still shake the floor when needed though!
.I'm being very picky with the results I'm getting. I think most people would be more than pleased to get to this level. I just know there's a little more to be had once I get the refection behind the arrays taken care of. Even if it is with movable panels.
No other obvious reflections in the first 20 ms. The side panel is doing it's job. It's a huge panel tucked away behind the curtain. The width extends beyond the speakers and the speakers disappear completely sonically. No small feat for such huge towers.
First reflections? True, standing waves? Not that much of a problem. Having the array in the corners seems to work really well. No overpowering lows, no complaints yet from me. The arrays are way more smooth on the bottom end than my previous speakers with 15" drivers. They do still shake the floor when needed though!
I'm being very picky with the results I'm getting. I think most people would be more than pleased to get to this level. I just know there's a little more to be had once I get the refection behind the arrays taken care of. Even if it is with movable panels.
No other obvious reflections in the first 20 ms. The side panel is doing it's job. It's a huge panel tucked away behind the curtain. The width extends beyond the speakers and the speakers disappear completely sonically. No small feat for such huge towers.
I reckon a line array excites room modes in a totally different way compared to a point source, so maybe that's giving you a great bonus here too
I reckon a line array excites room modes in a totally different way compared to a point source, so maybe that's giving you a great bonus here too
This is very true IMO, I have a set of smaller arrays (20drivers) and floor/ceiling reflections are not separated from the array itself in the time domain. The reflections virtually extend the physical array. In other words (simplified) they attenuate instead of distract.
Had traditional floor standers before and had problems with reflections, the arrays without any sound treatment to the room dramatically improved this issue. Treating the room since then have improved is even further.
Have a look at this:
The Murphy Corner-Line-Array Home Page
It is a very good thing to be able to put line arrays into the corners. One of my fellow audiophile freinds have made line arrys using the same Vifa drivers and have them in the corners just like Murphy´s . He only have 16 drivers pr.side and still have very extended lows without subwoofer an only minimal EQ.
The Murphy Corner-Line-Array Home Page
It is a very good thing to be able to put line arrays into the corners. One of my fellow audiophile freinds have made line arrys using the same Vifa drivers and have them in the corners just like Murphy´s . He only have 16 drivers pr.side and still have very extended lows without subwoofer an only minimal EQ.
Size matters....
Hi guys,
When it comes to room integration I believe that floor to ceiling (or at least 75% of floor to ceiling height) line arrays have similar advantages to Dr Geddes distributed sub woofer layout.
Below 80Hz, a few distributed subs will be vastly superior to a single sub.
Full range Line Arrays can extend these benefits even more if you use enough drivers and cover at least 75% of the floor to ceiling height.
You get a very even and natural bass response covering all or most of the room.
These benefits also extend much higher up in frequency with full range Line Arrays, for different reasons of course, the crucial low midrange (up to about 400Hz) is much easier to get right.
Also if you wall mount your arrays you kiss goodbye all baffle step worries and gain boundary reinforcement benefits....
By the way low end performance with line arrays using lots of small full range drivers is totally dependent on Sd.....Size matter!
I like the corner horns from Murphy, but for my listening requirements, they do not have enough depth, dynamics or detail in the bass, they need subs.
Even with the benefit of wall mounting / boundary reinforcement, I have found that the equivalent of a single 15 driver each side (approx 800Sd) is required to provide effortless bass at realistic SPL's. YMMV....!
Hope this helps.
Cheers
Derek.
Hi guys,
When it comes to room integration I believe that floor to ceiling (or at least 75% of floor to ceiling height) line arrays have similar advantages to Dr Geddes distributed sub woofer layout.
Below 80Hz, a few distributed subs will be vastly superior to a single sub.
Full range Line Arrays can extend these benefits even more if you use enough drivers and cover at least 75% of the floor to ceiling height.
You get a very even and natural bass response covering all or most of the room.
These benefits also extend much higher up in frequency with full range Line Arrays, for different reasons of course, the crucial low midrange (up to about 400Hz) is much easier to get right.
Also if you wall mount your arrays you kiss goodbye all baffle step worries and gain boundary reinforcement benefits....
By the way low end performance with line arrays using lots of small full range drivers is totally dependent on Sd.....Size matter!
I like the corner horns from Murphy, but for my listening requirements, they do not have enough depth, dynamics or detail in the bass, they need subs.
Even with the benefit of wall mounting / boundary reinforcement, I have found that the equivalent of a single 15 driver each side (approx 800Sd) is required to provide effortless bass at realistic SPL's. YMMV....!
Hope this helps.
Cheers
Derek.
This is very true IMO, I have a set of smaller arrays (20drivers) and floor/ceiling reflections are not separated from the array itself in the time domain. The reflections virtually extend the physical array. In other words (simplified) they attenuate instead of distract.
Had traditional floor standers before and had problems with reflections, the arrays without any sound treatment to the room dramatically improved this issue. Treating the room since then have improved is even further.
Better performance/value proposition than going with the Adytin iMagic yes?
Did you audition iMagic?
Do you have any idea how your speakers compare to iMagic?
Hi Derek,
With a Sd of 925 for the 25 driver line arrays we've got that covered
. Also well above the 75% floor to ceiling height. I agree that it seems to work quite similar to the Geddes multisub approach. With the advantage of spreading out the sources in the vertical direction as well. Geddes proposal spoke of a height difference as well, placing at least one sub at an elevated spot in the room.