Dynaudio (Evoke) + ExpertPro 220

[David A]

Possibly @Jean-Marie is referring the cross-over?

That's what I wondered, and why I asked, but the problem with that if what SAM is primarily designed to do is to correct phase errors then a speaker which has had its bass phase response corrected by the crossover has no need of SAM. There is one other problem and that is that I have never seen a test report for a speaker that has perfect phase response in the bass.

If @Jean-Marie was referring to the crossover he would be correct, you would expect to see no difference between SAM off and SAM at 0% if phase correction is the only thing happening at 0% and that's what Devialet's statements suggest but @Delija has Dynaudios and @jvoki has Magicos and the test reports I have seen for  several models of speaker from both manufacturers have always shown phase errors at low bass frequencies so neither of those speakers is going to fall into Jean Marie's "Speaker A" group hence we should expect to find a difference between SAM off and SAM on at 0% in their test results.

I cannot think of a speaker with a cone bass driver that does not have phase errors below 150 Hz. If there is a speaker that does not have phase errors in that area then it probably doesn't have a cone woofer so it's will use something like  a ribbon or electrostatic panel (only other bass driver technologies I can think of) and Devialet say they can't make SAM profiles for those types of drivers so we would not be able to test their performance with SAM.

[Delija]

Magicos are sealed and C60s are vented speakers, so it could make a difference in SAM profiles for these two speakers. Besides, the differences in measurements made by @jvoki with SAM 0% and SAM OFF are below 25 Hz - maybe some of them are 'masked" on C60s in my room, because I have pretty strong room modes around 20-22 Hz combined with room gain and also vented speakers, even quite large, hardly have any response below 20Hz - BR is always tuned higher.

I wish I could repeat my measurements (I should also order UMIK-1 and switch from XTZ Room Analyzer to REW), but I don't have C60s any more and my current speakers don't have SAM profile.

BTW, I don't like what SAM 0% is doing to Magicos - it behaves like 24 dB/oct high-pass filter. It is below 25 Hz - but why would anyone want that? If you want subsonic filter - then you use subsonic filter. Or you explicitly turn on high-pass filter in Devialet configurator. Even with SAM 50% there is some "filtering" - according to measurements posted by @jvoki.

[Jean-Marie]

Possibly @Jean-Marie is referring the cross-over?

That's what I wondered, and why I asked, but the problem with that if what SAM is primarily designed to do is to correct phase errors then a speaker which has had its bass phase response corrected by the crossover has no need of SAM. There is one other problem and that is that I have never seen a test report for a speaker that has perfect phase response in the bass.

If @Jean-Marie was referring to the crossover he would be correct, you would expect to see no difference between SAM off and SAM at 0% if phase correction is the only thing happening at 0% and that's what Devialet's statements suggest but @Delija has Dynaudios and @jvoki has Magicos and the test reports I have seen for  several models of speaker from both manufacturers have always shown phase errors at low bass frequencies so neither of those speakers is going to fall into Jean Marie's "Speaker A" group hence we should expect to find a difference between SAM off and SAM on at 0% in their test results.

I cannot think of a speaker with a cone bass driver that does not have phase errors below 150 Hz. If there is a speaker that does not have phase errors in that area then it probably doesn't have a cone woofer so it's will use something like  a ribbon or electrostatic panel (only other bass driver technologies I can think of) and Devialet say they can't make SAM profiles for those types of drivers so we would not be able to test their performance with SAM.

yes I was referring to the cross over as @thumb5 guessed.

I'm not pretending that there exists cone based speakers with zero phase defect, especially in the low frequencies, but I would greatly expect that the amount depends vastly of the type of load, type of speaker, etc.... As I stated, this was a though experiment pushing the the extreme to 'demonstrate' that the effect of SAM at 0% was certainly very much speaker dependent.

Jean-Marie

[David A]

@Jean-Marie ,

I accept that it's a thought experiment but take a look at the impedance plots in Stereophile speaker reviews. There are usually huge swings in impedance in the below 150 Hz range, regardless of speaker type so, as a thought experiment on my part, I find it hard to believe that there are all that many speakers which are going to require so little phase correction in that region that the effect of SAM at 0% would be indistinguishable from the result with SAM off, PROVIDED the right thing is measured and that it is measured in a way which can reveal the difference.

I make those 2 provisions because of my experience with SAM at 0% and at very low levels (below 25% or so). The main thing I notice with SAM at that level isn't a change in bass levels but rather greater bass clarity and definition and I'm not certain that frequency response is what we need to measure in order to demonstrate that. My second proviso about measuring things in the right way relates to the fact that measuring the same thing in different ways can reveal different things which are important, as can easily be seen when you compare a measurement of total THD which is a simple % and compare it to a measurement of THD showing the spectrum of the distortion which reveals significantly different information and which tends to be much more indicative of the sort of sound to expect from that amount of distortion.

[Confused]

@Confused ,

Thanks for pointing out that you can play test signals from Roon and measure the results in REW. I found your comments on the results of your experiments quite interesting, especially in relation to SAM. Would it be possible for you at some stage to post some of your results with SAM, especially some frequency response plots made with SAM at different settings, including one plot with SAM actually turned off and another with SAM on at 0%. That would definitely be off topic for this thread and would definitely deserve a head of it's own but that kind of information would be extremely useful for me and I suspect for many others here as well.

David - See below.  The Green line is SAM 100%, the blue line SAM 0%, the orange line SAM off.  

   

[Jean-Marie]

@Jean-Marie ,

I accept that it's a thought experiment but take a look at the impedance plots in Stereophile speaker reviews. There are usually huge swings in impedance in the below 150 Hz range, regardless of speaker type so, as a thought experiment on my part, I find it hard to believe that there are all that many speakers which are going to require so little phase correction in that region that the effect of SAM at 0% would be indistinguishable from the result with SAM off, PROVIDED the right thing is measured and that it is measured in a way which can reveal the difference.

I make those 2 provisions because of my experience with SAM at 0% and at very low levels (below 25% or so). The main thing I notice with SAM at that level isn't a change in bass levels but rather greater bass clarity and definition and I'm not certain that frequency response is what we need to measure in order to demonstrate that. My second proviso about measuring things in the right way relates to the fact that measuring the same thing in different ways can reveal different things which are important, as can easily be seen when you compare a measurement of total THD which is a simple % and compare it to a measurement of THD showing the spectrum of the distortion which reveals significantly different information and which tends to be much more indicative of the sort of sound to expect from that amount of distortion.

@David A

We are in violent agreement, especially the fact that the frequency response within the listening room is probably one of the least significant measurements to study the impact of phase correction. 

I am not surprised that SAM at 0% gives you more bass clarity. 
I’m using it at 100%, and believe it or not, but on some pieces it results in less bass than without SAM because SAM is correcting some kind of what I would call overshooting of the speaker for certain frequencies. 

Jean-Marie

[Confused]

@Jean-Marie ,

I accept that it's a thought experiment but take a look at the impedance plots in Stereophile speaker reviews. There are usually huge swings in impedance in the below 150 Hz range, regardless of speaker type so, as a thought experiment on my part, I find it hard to believe that there are all that many speakers which are going to require so little phase correction in that region that the effect of SAM at 0% would be indistinguishable from the result with SAM off, PROVIDED the right thing is measured and that it is measured in a way which can reveal the difference.

I make those 2 provisions because of my experience with SAM at 0% and at very low levels (below 25% or so). The main thing I notice with SAM at that level isn't a change in bass levels but rather greater bass clarity and definition and I'm not certain that frequency response is what we need to measure in order to demonstrate that. My second proviso about measuring things in the right way relates to the fact that measuring the same thing in different ways can reveal different things which are important, as can easily be seen when you compare a measurement of total THD which is a simple % and compare it to a measurement of THD showing the spectrum of the distortion which reveals significantly different information and which tends to be much more indicative of the sort of sound to expect from that amount of distortion.

@David A

We are in violent agreement, especially the fact that the frequency response within the listening room is probably one of the least significant measurements to study the impact of phase correction. 

I am not surprised that SAM at 0% gives you more bass clarity. 
I’m using it at 100%, and believe it or not, but on some pieces it results in less bass than without SAM because SAM is correcting some kind of what I would call overshooting of the speaker for certain frequencies. 

Jean-Marie

One thing I have just noticed is that we (based on a not very representative sample of only three!) seem to be getting different results.  In my case, SAM 0% appears to give a slight increase in bass versus SAM off (much as @David A expected and Floyd Toole predicted) @jvoki (previous page) gets the opposite result, more bass with SAM off versus SAM 0%.  Then we have @Delija who reports that SAM 0% and SAM off yield the same result, in terms of FR at least.

So three results, and all different.  Interesting, but I am not at all sure what to make of this!

[David A]

@Confused ,

I should have responded to your earlier post by now but the weekend got in the way and I've been thinking, probably thinking too much because I don't know enough about what's going on to hope for much result that will convince anyone, but I have come up with something.

So far we have 3 sets of results, 2 with REW and one with XYZ Room Analyser. I don't know how much the difference in software makes so I'll just say what I see from the 2 REW results.

@jvoki 's result shows frequency response with SAM On at 0% dropping from around 25 Hz but also stronger below 25 Hz with SAM at 50 and 100% than with SAM off. Looking at Devialet's info for SAM with the Magico S7 it shows extension to 26 Hz with SAM off and to 17 Hz with SAM on but doesn't say what the SAM setting is. In some ways jvoki's result matches what Devialet are saying. The increasing gain with higher levels of SAM starts at around 50 Hz and increases as frequency drops. The gain really starts about an octave above where the divergence between SAM On and OFF starts.

Your results with the Blades are actually similar to jvoki's in some ways. Devialet give an extension of 24.6 Hz without SAM and 14.4 with SAM. Your results with SAM Off and at 0% start diverging more at around 24 Hz and your boost with SAM at 100 % starts a bit above 50 Hz, a little more than an octave above where the divergence between SAM Off and at 0% starts. In both your case and jvoki's, the divergence between SAM Off and at 0% starts at around the frequency Devialet give for your speakers without SAM. What's really different between your result and jvoki's is what happens below the frequency where SAM Off and at 0% starts to occur.

That leads me to wonder whether the difference is due to the nature of the phase error being corrected at that frequency. In other words, I think the result both of you got is probably in line with Devialet's published claims about your speakers, pretty minimal as the information for Devialet's claims is given that it just gives us 2 frequencies, no description of what is going on at the before frequency when SAM is switched on, and no plots which show actual frequency response.

@Delija reports that measurements with SAM off and SAM on for his Dyns is the same and from your results and jvoki's that what I would expect to see down to the claimed extension frequency without SAM. Below that what your results and jvoki's show is that the amount of divergence differs with different speakers and Delija reports no difference with his which is closer to your result than to jvoki's.

All 3 of you have presented your results with plots with different scales and overall frequency ranges which makes comparing the actual results tricky but when I really started analysing them, I ended up thinking that they may not be all that different to each other for the reasons given above.

I think we'd all like to see a lot more information from Devialet to tell us what's going on.