The latest Stereophile review was for the multi-channel version of Dirac Live. My system is just the old 2-channel.
In the grand scheme of things, setup wasn't too painful, but I guess I wasn't expecting the computer world to bring its worst attributes into the audio world. Too many configuration options for default software installs to anticipate properly. Sticking to the subject of Dirac Live, the setup was pretty much as described in Ken Rockwell's excellent write-up Rockwell-Review.
Things to look out for, were the options. For example, the default is for the "sofa" setup, and if you notice mid calibration that you wanted to select the "chair" setup, you basically have to throw everything out and start over. Also, a bit of confusion about saving "projects" versus saving "filters". It eventually sorted out, but you'll probably lose some valuable test calibrations in the process of creating multiple filters, due to that confusion. Losing filters is no big deal, because you can create them simply by tugging and pulling at the orange curve. The time-consuming thing is the multiple mic setups, so you don't want to corrupt those files. After applying some discipline (saving files often, with different file names) things became reasonable and salvageable.
Kind of crazy and unexpected, after I constructed my filters, the default device in Dirac was for the JRiver v19, rather than my Dirac-modified sound card....I didn't realize that mattered until many other things were tried. Fix was a simple click to activate the correct choice, but jeez- why wasn't the default the same device that I calibrated for?
Another time-waster was in the coordination of JRiver settings with the Dirac settings. Didn't quite think clearly through this but when you think of it, your "audio device" way back in Windows Sound settings has options for upsampling to 96k. The Dirac also has options for upsampline to 96k, and JRiver also has these options. The trick here isn't to process the processed processor files, so that your computer isn't working over-overtime trying to convert your music stream. I don't know if any of that made sense, but the advice here is to use as many "automatic" settings as possible, so that you don't have multiple devices trying to convert your data stream.- it doesn't work, and instead of the encouraging "streaming" light appearing on your Dirac filter UI, you'll instead see the "disconnected" light, which does not help your attitude when troubleshooting. (anticipation of breakthrough improvement is painful--you don't want it to take too long)
JRiver has lots of variables, such as "kernal streaming" or "direct sound" or "WASAPI", without a lot of explanation of the differences or potential incompatibilities. Not much input from the web either. More trial and error to find the combo that lights the magic "streaming" light. Eventually found web reading suggesting that WASAPI bypasses all windows-based sound controls filters and equalizers, which sounds good to my "single wire, with gain" ethos.
I simplified my mic samples by purchasing a $20 clamp-on mic stand from a music supply (for kick drum mics). Using that setup I was easily able to clamp the mic to the back of my high-backed aeron office chair, which approximated my ear height, then wheel closer or further from the sweet spot and clamp to right shoulder or left shoulder position to get the necessary (9) room samples. I read on one of the threads somewhere that DRC doesn't really kick in unless the samples have at least a 36" distance between them, but I didn't really find that to apply in my nearfield experience, with my end corrected result sounding great with only about a 30" separation of the mic extremes.
Now, I'm wondering if I should get a computer for my home system, because I'm betting there are many more room effects in my listening room (where I am much farther from my speakers than in the office). That will be for another day. I guess there may be more tinkering in my future- for this type of acoustic clarity, especially regarding natural instrument reverb, its worth it.
In retrospect, I'm probably exaggerating the time and trouble involved. A full mic setup start-to-finish is probably a 15-20 minute exercise. Creating a filter, now that everything is familiar, takes a couple minutes tops. All that troubleshooting for device compatibility initially was probably 3-4 hours, because I wasn't sure if I had interface, software, processor, cabling or any other cause/effect until going through the trial-by-doing manly method without too much reference to actual instructions.
If I do go forward with DRC for my home listening room, with my experience gained in the office, I'd guess my time from download of software to full-on room-corrected listening to be less than a half hour. I'm tempted to do it, just to amortize my initial efforts at troubleshooting. If I want to really be punished, I think there may be a variation to setting up that is more specific for subwoofers. I'll need a bit of recovery time before I open up that option. I'm enjoying my current state way too much.
In the grand scheme of things, setup wasn't too painful, but I guess I wasn't expecting the computer world to bring its worst attributes into the audio world. Too many configuration options for default software installs to anticipate properly. Sticking to the subject of Dirac Live, the setup was pretty much as described in Ken Rockwell's excellent write-up Rockwell-Review.
Things to look out for, were the options. For example, the default is for the "sofa" setup, and if you notice mid calibration that you wanted to select the "chair" setup, you basically have to throw everything out and start over. Also, a bit of confusion about saving "projects" versus saving "filters". It eventually sorted out, but you'll probably lose some valuable test calibrations in the process of creating multiple filters, due to that confusion. Losing filters is no big deal, because you can create them simply by tugging and pulling at the orange curve. The time-consuming thing is the multiple mic setups, so you don't want to corrupt those files. After applying some discipline (saving files often, with different file names) things became reasonable and salvageable.
Kind of crazy and unexpected, after I constructed my filters, the default device in Dirac was for the JRiver v19, rather than my Dirac-modified sound card....I didn't realize that mattered until many other things were tried. Fix was a simple click to activate the correct choice, but jeez- why wasn't the default the same device that I calibrated for?
Another time-waster was in the coordination of JRiver settings with the Dirac settings. Didn't quite think clearly through this but when you think of it, your "audio device" way back in Windows Sound settings has options for upsampling to 96k. The Dirac also has options for upsampline to 96k, and JRiver also has these options. The trick here isn't to process the processed processor files, so that your computer isn't working over-overtime trying to convert your music stream. I don't know if any of that made sense, but the advice here is to use as many "automatic" settings as possible, so that you don't have multiple devices trying to convert your data stream.- it doesn't work, and instead of the encouraging "streaming" light appearing on your Dirac filter UI, you'll instead see the "disconnected" light, which does not help your attitude when troubleshooting. (anticipation of breakthrough improvement is painful--you don't want it to take too long)
JRiver has lots of variables, such as "kernal streaming" or "direct sound" or "WASAPI", without a lot of explanation of the differences or potential incompatibilities. Not much input from the web either. More trial and error to find the combo that lights the magic "streaming" light. Eventually found web reading suggesting that WASAPI bypasses all windows-based sound controls filters and equalizers, which sounds good to my "single wire, with gain" ethos.
I simplified my mic samples by purchasing a $20 clamp-on mic stand from a music supply (for kick drum mics). Using that setup I was easily able to clamp the mic to the back of my high-backed aeron office chair, which approximated my ear height, then wheel closer or further from the sweet spot and clamp to right shoulder or left shoulder position to get the necessary (9) room samples. I read on one of the threads somewhere that DRC doesn't really kick in unless the samples have at least a 36" distance between them, but I didn't really find that to apply in my nearfield experience, with my end corrected result sounding great with only about a 30" separation of the mic extremes.
Now, I'm wondering if I should get a computer for my home system, because I'm betting there are many more room effects in my listening room (where I am much farther from my speakers than in the office). That will be for another day. I guess there may be more tinkering in my future- for this type of acoustic clarity, especially regarding natural instrument reverb, its worth it.
In retrospect, I'm probably exaggerating the time and trouble involved. A full mic setup start-to-finish is probably a 15-20 minute exercise. Creating a filter, now that everything is familiar, takes a couple minutes tops. All that troubleshooting for device compatibility initially was probably 3-4 hours, because I wasn't sure if I had interface, software, processor, cabling or any other cause/effect until going through the trial-by-doing manly method without too much reference to actual instructions.
If I do go forward with DRC for my home listening room, with my experience gained in the office, I'd guess my time from download of software to full-on room-corrected listening to be less than a half hour. I'm tempted to do it, just to amortize my initial efforts at troubleshooting. If I want to really be punished, I think there may be a variation to setting up that is more specific for subwoofers. I'll need a bit of recovery time before I open up that option. I'm enjoying my current state way too much.