I’ve posted a couple of screenshots of the room corrections/equalizations that I’ve settled on, at least for the time being, at this link. One depicts the full spectrum, and one has an expanded view covering just up to 500 Hz. The room corrections I’ve settled on just extend up to 400 Hz, which just by coincidence happens to be the lower limit I had chosen for the speaker calibrations, for which I previously provided screenshots.
As you can see in the listings on the left side of the screenshots, I had made a total of 11 different measurements of each speaker, all in the vicinity of the listening position but at various distances and heights (all of them centered between the speakers). (“D” in the measurement position listings indicates distance from the plane of the speakers in feet, and “H” represents the measurement height in inches). Eventually I chose the measurements that were performed at a 12 foot distance and a 36 inch height to work with in deriving the corrections. The screenshots show those particular measurements, together with the equalization curves I’m now using.
The 12 foot distance corresponds just about exactly to my usual listening distance. The 36 inch height is about 2 inches above the mid-point between the two tweeters that are approximately centrally located on the vertical axis of the baffle area of my speakers. That height is also within a couple of inches or so of the height of my ears and my wife’s ears, when we are seated in the (quite low) Wassily chairs you can see in my system description photos.
You’ll note in the full spectrum screenshot that the upper treble rolls off significantly at the listening position. That occurs even though, as you can see in the speaker measurements I posted some time ago, the speakers themselves have a significant rise in frequency response in the top octave (the magnitude of that rise, remarkably, being similar to the magnitude of the dip at the listening position), and it occurs even though a mild dip in the 5 to 10 kHz area which the speakers exhibited in the close up measurements was pretty much compensated for in the speaker calibration process. But I’ve nevertheless chosen not to introduce any room correction equalizations in the treble region, after taking into account recommendations I had received from both Nyal Mellor of Acoustic Frontiers and Alan Langford of DEQX suggesting that it is generally best to leave the upper treble region alone, as well as my ongoing satisfaction with the tonal balance of my speakers in that region, as well as the impression I’ve formed over the years that a system which measures flat at the listening position throughout the treble region will sound too bright to most listeners.
Another point I found to be particularly striking, when I compared the measurements I had taken at the various distances, is that as might be expected the numerous peaks and valleys in frequency response occurring in the bass region as a result of room reflections will be located at frequencies that differ significantly depending on the specific measurement position. Meaning that basing a correction on a measurement performed at a distance that is a foot or two greater or less than the listening distance can result in an equalization boost being applied to a peak instead of a valley, or an equalization cut being applied to a valley instead of a peak. Particularly if one is aggressive in trying to correct for relatively narrow peaks or valleys.
Summing up my experience with the DEQX to this point, I would say that it has provided significant benefit to me as a result of its speaker calibration function, its room correction function, and as a preamp, and as a DAC. As I alluded to in an earlier post, speaker calibration seems to be especially beneficial in my system on recordings that are sonically mediocre, or worse. @Bombaywalla had said some time ago, either in this thread or in the “Sloped Baffle” thread, that that can be expected to be a consequence of improvements in time coherence, and my experience with the DEQX appears to confirm that.
So I am glad I chose to make the substantial investments of money and time that have been involved. The time factor presumably would have been considerably less if I had chosen to utilize the DEQXpert service, but while I have no way of knowing whether the results of doing that would have ended up being better than, worse than, or similar to what I have accomplished, I feel that the learning experience resulting from doing it all myself has been sufficiently valuable that I don’t regret not having utilized that service.
Next on my list, now that I have added the excellent and LOMC-capable Herron phono stage to my system (as I mentioned in an earlier post), is that in the next couple of weeks I’m planning to purchase an LOMC cartridge, to use in place of or as an alternate to my Soundsmith re-tipped vintage Grace F9E moving magnet. I’m planning on getting a Dynavector 17D3, in part because of the excellent reviews it has received over the years, but particularly because member @Rodman99999 had mentioned to me in a thread a while back that it is “magic” in the vintage Magnepan Unitrac tonearm he and I both use.
Thanks for reading. Best regards,
-- Al
As you can see in the listings on the left side of the screenshots, I had made a total of 11 different measurements of each speaker, all in the vicinity of the listening position but at various distances and heights (all of them centered between the speakers). (“D” in the measurement position listings indicates distance from the plane of the speakers in feet, and “H” represents the measurement height in inches). Eventually I chose the measurements that were performed at a 12 foot distance and a 36 inch height to work with in deriving the corrections. The screenshots show those particular measurements, together with the equalization curves I’m now using.
The 12 foot distance corresponds just about exactly to my usual listening distance. The 36 inch height is about 2 inches above the mid-point between the two tweeters that are approximately centrally located on the vertical axis of the baffle area of my speakers. That height is also within a couple of inches or so of the height of my ears and my wife’s ears, when we are seated in the (quite low) Wassily chairs you can see in my system description photos.
You’ll note in the full spectrum screenshot that the upper treble rolls off significantly at the listening position. That occurs even though, as you can see in the speaker measurements I posted some time ago, the speakers themselves have a significant rise in frequency response in the top octave (the magnitude of that rise, remarkably, being similar to the magnitude of the dip at the listening position), and it occurs even though a mild dip in the 5 to 10 kHz area which the speakers exhibited in the close up measurements was pretty much compensated for in the speaker calibration process. But I’ve nevertheless chosen not to introduce any room correction equalizations in the treble region, after taking into account recommendations I had received from both Nyal Mellor of Acoustic Frontiers and Alan Langford of DEQX suggesting that it is generally best to leave the upper treble region alone, as well as my ongoing satisfaction with the tonal balance of my speakers in that region, as well as the impression I’ve formed over the years that a system which measures flat at the listening position throughout the treble region will sound too bright to most listeners.
Another point I found to be particularly striking, when I compared the measurements I had taken at the various distances, is that as might be expected the numerous peaks and valleys in frequency response occurring in the bass region as a result of room reflections will be located at frequencies that differ significantly depending on the specific measurement position. Meaning that basing a correction on a measurement performed at a distance that is a foot or two greater or less than the listening distance can result in an equalization boost being applied to a peak instead of a valley, or an equalization cut being applied to a valley instead of a peak. Particularly if one is aggressive in trying to correct for relatively narrow peaks or valleys.
Summing up my experience with the DEQX to this point, I would say that it has provided significant benefit to me as a result of its speaker calibration function, its room correction function, and as a preamp, and as a DAC. As I alluded to in an earlier post, speaker calibration seems to be especially beneficial in my system on recordings that are sonically mediocre, or worse. @Bombaywalla had said some time ago, either in this thread or in the “Sloped Baffle” thread, that that can be expected to be a consequence of improvements in time coherence, and my experience with the DEQX appears to confirm that.
So I am glad I chose to make the substantial investments of money and time that have been involved. The time factor presumably would have been considerably less if I had chosen to utilize the DEQXpert service, but while I have no way of knowing whether the results of doing that would have ended up being better than, worse than, or similar to what I have accomplished, I feel that the learning experience resulting from doing it all myself has been sufficiently valuable that I don’t regret not having utilized that service.
Next on my list, now that I have added the excellent and LOMC-capable Herron phono stage to my system (as I mentioned in an earlier post), is that in the next couple of weeks I’m planning to purchase an LOMC cartridge, to use in place of or as an alternate to my Soundsmith re-tipped vintage Grace F9E moving magnet. I’m planning on getting a Dynavector 17D3, in part because of the excellent reviews it has received over the years, but particularly because member @Rodman99999 had mentioned to me in a thread a while back that it is “magic” in the vintage Magnepan Unitrac tonearm he and I both use.
Thanks for reading. Best regards,
-- Al