Some fascinating posts in this thread this week, it has certainly got me thinking about a few things. Although one thing I would say is that the thread has drifted a little onto how much power does a Devialet produce and how does it do it, and a little away from the "How much power do you really need" question.
This takes me back to the original video. One thing that is a fact is that the CI Precision amplifiers were routinely showing 400 to 750w peaks. OK - RebelMan was quick to dismiss the validity of the video, making the point that SPL reading were not measured. I can only agree that the video would have been more informative if this data was included, but personally I take more of an "it is what it is" view, we have the video, what can we make of it? If anyone finds something better, then please post it, meanwhile I will try to make the best of what we have. One point here is that we do not really know how accurate the CI Precision amp power readings are, but they are a serious audio company, and the CI Precisions are very expensive, this is not some gimmick on a toy. I do not know for sure, but I suspect that they are probably reasonably accurate. Does anyone know for sure?
So how loud were they playing in the video? Obviously this was not measured, but there are some clues. From 5:50 to 7:00 minutes you can clearly hear the voices of the attendees in the room over the music, furthermore, in the same time frame you can see three people approaching to within a foot of the speakers, so this does give us an idea,
For a bit of fun, I tried the "Laptev Sea" track at home, armed with a probably not very accurate iPhone app. At a loud but almost comfortable level, I was getting an average of 90dB, with peaks of about 97dB. I am sure that if recorded, you could hear me speak over the music at this level. As a guestimate, I would say the music in the video is at about this level, if anything a little quieter. I tried pushing the volume higher, and did get peaks of about 103dB. I did this for maybe two seconds. Trust me, I would NOT want to listen to this track at this kind of volume. I know about hearing damage due to sustained SPL levels, and even a couple of seconds at that volume was too much. It is a guess I know, but I am 99% as certain as I can be that the volumes in the video demo room were much lower. I would not have stuck my head one foot away from the Blades with that volume level, not a chance. For the record, I briefly ran the amps to about -7dB, and the clipping indicator was not activating. So how much power did this need? I do not know, but if someone knows a cleaver way of measuring this I am all ears. (If now maybe suffering temporary hearing loss)
So I am happy to speculate at maybe 90 to 95dB in the room with peaks, and defeatedly not significantly higher, that just would not be credible. Running as loud as I did for a very brief moment, people would have been leaving the room, I would not have stayed in there for more than a second.
When I first saw the video, my thoughts went to the Harbeths, I know they are not the most efficient speaker, and I presumed they must be a hopelessly difficult load to drive. A quick check on the 'net shows this presumption to be wrong. Stereophile have tested the 40.1's, and the opposite is true, they have a very amplifier friendly looking impedance curve, see link below:
https://www.stereophile.com/content/harb...asurements
The review is also interesting in that Art Dudley ran the Harbeths with a selection of low powered amplifiers, and found a preference when using a 25wpc amp. Now I am sure that if the CI Precisions in the video were swapped for a 25wpc amp, playing the same track in the same room at the same volume, the result would sound very different. Much speculation I know, but whatever anyone says, those amps were showing 750w peaks.
So what about the Laptev Sea track itself? This I think is a factor. As it happens, I have a copy on CD. So I thought I would try it in my car. My car has the standard stereo as provided by the manufacturer. I actually really like how it sounds, it is pretty decent, but every now and again I might get a particular track at a particular volume where I feel the need to turn the volume down by one click, it is just on the edge of distortion. So it sounds good, but is clearly marginal on power. Playing the Latev Sea track, and it was really struggling. So out of thousands of pieces of recorded music that I have played in that car, the Laptev Sea track comes out on top as one of the hardest to reproduce.
Checking in Roon, the album has a DR of +/- 11, the Laptev Sea track is +/- 3. (which shows you how album DR's can be misleading) iTunes indicates a volume compensation value, which for the Laptev Sea track is -0.9dB. What this tells anyone I am not sure.
For me, the exceptional thing about the track (subjectively) is the power and extent of very low bass. At the very lowest frequencies the Harbeths have an impedance of about 8 ohms. Others have mentioned that the Devialet in dual mono can run +/- 55v, giving 110v. If we assume say 100v (or running almost flat chat, as it otherwise might be known) and using ohms law, this gives a current of 100/8 = 12.5amps. Volts x amps = watts, so this is 100 x 12.5 = 1250 watts. OK, that is a lot, too big a number, but the Blades are a touch more efficient than the Harbeths, so for me the 750w number, as a transient peak looks entirely credible. In other words, I could not stay in the Room with the Blades at almost full volume, with the slightly less efficient Harbeth's and a bigger room, I might get nearer to full volume, but on balance, the 750w figure does look credible, and indeed it is the number on the front of the amp. OK, a lot of assumptions here and the "math" is crude compared to what is a very complex and dynamic system, but it is starting to make some sense to me, maybe in some situations you might need a touch more power than you might think.
As a bit of trivia, listening on the Blades to the recording of the Harbeths playing Laptev Sea, everything sounded clear, no obvious distortions, and the low bass comes through nicely. Obviously the track sounds better played direct on the Blades and the Harbeth's would undoubtedly sound much better listening to them direct in the room, but those Harbeths still sound decent when recorded, I am guessing they are impressive when heard for real. (I have heard other Harbeths, but never the 40.1)
As a final point, I remain fascinated by all this. It is easy do some simple calculations, I want to listen at a maximum of 95dB, my speakers produce 93dB for 1w/m, 10 watts is fine etc., but there is a lot more going on here I think, and to be honest, I think I need to learn more to fully understand all this. I mean, Mr Art Dudley was happy running his 40.1's with a 25 wpc amp, and I am sure they sounded great, the video I think does show a real world situation, albeit a rather extreme one, and the speakers are taking fully 30 times that much power, it boggles the mind.
This takes me back to the original video. One thing that is a fact is that the CI Precision amplifiers were routinely showing 400 to 750w peaks. OK - RebelMan was quick to dismiss the validity of the video, making the point that SPL reading were not measured. I can only agree that the video would have been more informative if this data was included, but personally I take more of an "it is what it is" view, we have the video, what can we make of it? If anyone finds something better, then please post it, meanwhile I will try to make the best of what we have. One point here is that we do not really know how accurate the CI Precision amp power readings are, but they are a serious audio company, and the CI Precisions are very expensive, this is not some gimmick on a toy. I do not know for sure, but I suspect that they are probably reasonably accurate. Does anyone know for sure?
So how loud were they playing in the video? Obviously this was not measured, but there are some clues. From 5:50 to 7:00 minutes you can clearly hear the voices of the attendees in the room over the music, furthermore, in the same time frame you can see three people approaching to within a foot of the speakers, so this does give us an idea,
For a bit of fun, I tried the "Laptev Sea" track at home, armed with a probably not very accurate iPhone app. At a loud but almost comfortable level, I was getting an average of 90dB, with peaks of about 97dB. I am sure that if recorded, you could hear me speak over the music at this level. As a guestimate, I would say the music in the video is at about this level, if anything a little quieter. I tried pushing the volume higher, and did get peaks of about 103dB. I did this for maybe two seconds. Trust me, I would NOT want to listen to this track at this kind of volume. I know about hearing damage due to sustained SPL levels, and even a couple of seconds at that volume was too much. It is a guess I know, but I am 99% as certain as I can be that the volumes in the video demo room were much lower. I would not have stuck my head one foot away from the Blades with that volume level, not a chance. For the record, I briefly ran the amps to about -7dB, and the clipping indicator was not activating. So how much power did this need? I do not know, but if someone knows a cleaver way of measuring this I am all ears. (If now maybe suffering temporary hearing loss)
So I am happy to speculate at maybe 90 to 95dB in the room with peaks, and defeatedly not significantly higher, that just would not be credible. Running as loud as I did for a very brief moment, people would have been leaving the room, I would not have stayed in there for more than a second.
When I first saw the video, my thoughts went to the Harbeths, I know they are not the most efficient speaker, and I presumed they must be a hopelessly difficult load to drive. A quick check on the 'net shows this presumption to be wrong. Stereophile have tested the 40.1's, and the opposite is true, they have a very amplifier friendly looking impedance curve, see link below:
https://www.stereophile.com/content/harb...asurements
The review is also interesting in that Art Dudley ran the Harbeths with a selection of low powered amplifiers, and found a preference when using a 25wpc amp. Now I am sure that if the CI Precisions in the video were swapped for a 25wpc amp, playing the same track in the same room at the same volume, the result would sound very different. Much speculation I know, but whatever anyone says, those amps were showing 750w peaks.
So what about the Laptev Sea track itself? This I think is a factor. As it happens, I have a copy on CD. So I thought I would try it in my car. My car has the standard stereo as provided by the manufacturer. I actually really like how it sounds, it is pretty decent, but every now and again I might get a particular track at a particular volume where I feel the need to turn the volume down by one click, it is just on the edge of distortion. So it sounds good, but is clearly marginal on power. Playing the Latev Sea track, and it was really struggling. So out of thousands of pieces of recorded music that I have played in that car, the Laptev Sea track comes out on top as one of the hardest to reproduce.
Checking in Roon, the album has a DR of +/- 11, the Laptev Sea track is +/- 3. (which shows you how album DR's can be misleading) iTunes indicates a volume compensation value, which for the Laptev Sea track is -0.9dB. What this tells anyone I am not sure.
For me, the exceptional thing about the track (subjectively) is the power and extent of very low bass. At the very lowest frequencies the Harbeths have an impedance of about 8 ohms. Others have mentioned that the Devialet in dual mono can run +/- 55v, giving 110v. If we assume say 100v (or running almost flat chat, as it otherwise might be known) and using ohms law, this gives a current of 100/8 = 12.5amps. Volts x amps = watts, so this is 100 x 12.5 = 1250 watts. OK, that is a lot, too big a number, but the Blades are a touch more efficient than the Harbeths, so for me the 750w number, as a transient peak looks entirely credible. In other words, I could not stay in the Room with the Blades at almost full volume, with the slightly less efficient Harbeth's and a bigger room, I might get nearer to full volume, but on balance, the 750w figure does look credible, and indeed it is the number on the front of the amp. OK, a lot of assumptions here and the "math" is crude compared to what is a very complex and dynamic system, but it is starting to make some sense to me, maybe in some situations you might need a touch more power than you might think.
As a bit of trivia, listening on the Blades to the recording of the Harbeths playing Laptev Sea, everything sounded clear, no obvious distortions, and the low bass comes through nicely. Obviously the track sounds better played direct on the Blades and the Harbeth's would undoubtedly sound much better listening to them direct in the room, but those Harbeths still sound decent when recorded, I am guessing they are impressive when heard for real. (I have heard other Harbeths, but never the 40.1)
As a final point, I remain fascinated by all this. It is easy do some simple calculations, I want to listen at a maximum of 95dB, my speakers produce 93dB for 1w/m, 10 watts is fine etc., but there is a lot more going on here I think, and to be honest, I think I need to learn more to fully understand all this. I mean, Mr Art Dudley was happy running his 40.1's with a 25 wpc amp, and I am sure they sounded great, the video I think does show a real world situation, albeit a rather extreme one, and the speakers are taking fully 30 times that much power, it boggles the mind.
1000 Pro - KEF Blade - iFi Zen Stream - Mutec REF10 - MC3+USB - Pro-Ject Signature 12
