ELAC DF63 Floor standing Speaker Review

[Henryk]

I have read BOTH Amir's and Erin's reviews and compared their measurements.

I am not feeling it it a "Pass" as you say, but more curious as to why the large differences found in measurements.
When two well respected guys come to fairly different conclusions, I tend to want to know why, and if the speaker was damaged or broken in any way, or maybe quality control issues etc.

Of course its easy to slag off any product easily, but takes some work to figure the "Why" of the differences.

Seeing as how the speaker is a mass-market item, couldn't you just poll some owners on other forums/sites for their experience?
I mean, it's ELAC; They must sell thousands of these sets!

Amir and Erin aren't the only trustworthy audiophiles in the world.

Yes, another trade-off in the world of high-fidelity.

So true, huh?
Seems like you gotta spend quite a bit to not feel like you're 'settling'.

I feel like the Q7/Q11 Meta is in a real sweet price/performance spot though.

 

[Sokel]

Ok, so I've done the math to work out what the dB SPL levels were at 1 metre for the speaker during Amir's noticing of the tweeter issue, which he noted at - 10dB, - 5dB, and intolerable issue at 0dB on his DAC/amp. I have to say I got ChatGPT to work this out after I gave it sensitivity specs of speaker and voltage output level of Amirs DAC/amp when he did his review of that DAC/amp when he had it set at - 16.5dB. I sanity checked the output of ChatGPT and I think it's OK. Following is couple of screenshots showing the results as copying and pasting wouldn't give correct format for some reason:

View attachment 474907
View attachment 474906
Apologies that images so large, don't know why couldn't be cut & pasted text from ChatGPT without insert of strange characters. But anyway, using Amir's descriptors this means at 98.3dB it became audible, at 103.3dB it was quite noticeable, and at 108.3dB it was almost intolerable. Note that this is at 1m for one speaker and the dB levels are the dB at 1kHz if you had played a 0dBFS (full scale) sine wave at those amp/DAC settings with this speaker, so by no means was he actually exposed to those levels because the music would be recorded below 0dBFS and he would be sitting further away than 1m. This instead is just to give an idea of how his amp/DAC "system" was setup in terms of max potential SPL output at the different points that he noticed tweeter issues with this speaker. I find this useful for reference because I measure my system in my room using REW at my listening point and know what the 0dBFS output level is for each of my different amp settings that I use - so it gives me or you a way of listening to Amir's track where he identified tweeter issues and you can say to yourself "yes that's the level of stress that the speakers were exposed to when Amir identified the issues", plus for curiosity it enables you to check you don't have same issues on your system. I worked out for instance that my speakers were at 101dB for 1m when I listened to Amir's track yesterday and I didn't detect tweeter issues in my JBL 308p speakers for instance, albeit this is a bit less than some of the testing Amir did in his listening review but does still sit in the range where he started to notice issues. (I hope all the math and assumptions are correct in here, I think it's ok).

EDIT: if you read this post within the first 20min of me posting it then have another look as I changed a few subtle things that change the meaning quite a lot.

I maybe missed it but I don't see the recorded level of the track (at least the combined spectrum at it at a busy point) which can never-ever reach 0dB.
I may have a look at it's line-level voltage and come back with some rough numbers.

 

[Robbo99999]

I maybe missed it but I don't see the recorded level of the track (at least the combined spectrum at it at a busy point) which can never-ever reach 0dB.
I may have a look at it's line-level voltage and come back with some rough numbers.

Maybe you misunderstood the point of the exercise, it's to understand Amir's system setup so you can reproduce that on your own speakers when listening to same track. It's just understanding the dB SPL (at 0dBFS for sine sweeps at 1kHz) that would play through that model of speaker with Amir's 0dB, -5dB, and -10dB on his DAC/amp where he noticed tweeter issues on that track. It's so you can setup your own system to the same speaker stress level as that which Amir was using. It means you have to measure your speaker system at different amp settings (at 0dBFS for sine sweeps) at your listening point and you work the SPL back mathematically to xdB at 1m for 1 speaker. It's a bit convoluted, but it enables people to (within some margin of error) to stress their speakers to the same level as Amir was during his noticing of the tweeter issue. It's got nothing to do with trying to quantify the SPL at Amir's listening point, and indeed has nothing to do with trying to quantify the SPL of the music playback at 1m even, it's just about setting up your speakers to the same stress level that Amir noticed the tweeter issue - it's immaterial what the recording level is of the track, as long as different versions of the track such as Youtube vs whatever source Amir was using doesn't have significantly different recording levels for that track.

 

[Sokel]

Ok, so I've done the math to work out what the dB SPL levels were at 1 metre for the speaker during Amir's noticing of the tweeter issue, which he noted at - 10dB, - 5dB, and intolerable issue at 0dB on his DAC/amp. I have to say I got ChatGPT to work this out after I gave it sensitivity specs of speaker and voltage output level of Amirs DAC/amp when he did his review of that DAC/amp when he had it set at - 16.5dB. I sanity checked the output of ChatGPT and I think it's OK. Following is couple of screenshots showing the results as copying and pasting wouldn't give correct format for some reason:

View attachment 474907
View attachment 474906
Apologies that images so large, don't know why couldn't be cut & pasted text from ChatGPT without insert of strange characters. But anyway, using Amir's descriptors this means at 98.3dB it became audible, at 103.3dB it was quite noticeable, and at 108.3dB it was almost intolerable. Note that this is at 1m for one speaker and the dB levels are the dB at 1kHz if you had played a 0dBFS (full scale) sine wave at those amp/DAC settings with this speaker, so by no means was he actually exposed to those levels because the music would be recorded below 0dBFS and he would be sitting further away than 1m. This instead is just to give an idea of how his amp/DAC "system" was setup in terms of max potential SPL output at the different points that he noticed tweeter issues with this speaker. I find this useful for reference because I measure my system in my room using REW at my listening point and know what the 0dBFS output level is for each of my different amp settings that I use - so it gives me or you a way of listening to Amir's track where he identified tweeter issues and you can say to yourself "yes that's the level of stress that the speakers were exposed to when Amir identified the issues", plus for curiosity it enables you to check you don't have same issues on your system. I worked out for instance that my speakers were at 101dB for 1m when I listened to Amir's track yesterday and I didn't detect tweeter issues in my JBL 308p speakers for instance, albeit this is a bit less than some of the testing Amir did in his listening review but does still sit in the range where he started to notice issues. (I hope all the math and assumptions are correct in here, I think it's ok).

EDIT: if you read this post within the first 20min of me posting it then have another look as I changed a few subtle things that change the meaning quite a lot.

Ok, I have some rough numbers, what you see is calibrated to the last mV:

That's a given busy moment of the electric sum of the track (the previous ones I posted were acoustic ones) , the most I saw coming out of it was 330mV.
Now, consider that this is an unbalanced connection and we also have to subtract -4dB for intersample overs, etc, the works.

Actual max output at 0dB after all that is 1.572V .

Now you have some numbers to work with.

 

[Robbo99999]

Ok, I have some rough numbers, what you see is calibrated to the last mV:

View attachment 474978

That's a given busy moment of the electric sum of the track (the previous ones I posted were acoustic ones) , the most I saw coming out of it was 330mV.
Now, consider that this is an unbalanced connection and we also have to subtract -4dB for intersample overs, etc, the works.

Actual max output at 0dB after all that is 1.572V .

Now you have some numbers to work with.

(You've missed the point of the exercise, see my previous post)

 

[Sokel]

(You've missed the point of the exercise, see my previous post)

I probably did, yes, thought that the exercise was to determine Amir's SPL level at the time that the speaker was falling apart.

 

[Salt]

Two different approaches to answer a question, and the results are not far apart of each other ... isn't science a wonderful playground ... ??

 

[Robbo99999]

I probably did, yes, thought that the exercise was to determine Amir's SPL level at the time that the speaker was falling apart.

Well, it was to determine the dB SPL level at 1m for one speaker during a 0dBFS sine sweep at 1kHz point during the different amp levels that he noticed the subjective differences of "audible", "noticeable", "intolerable" that corresponded to -10dB, -5dB, and 0dB on his DAC/amp settings. It's just so that people can theoretically set up their systems to those corresponding levels by changing their amp settings and running sine sweeps at 0dBFS, noting the dB SPL at 1kHz at their listening position and then you can work it back mathematically to xdB at 1m for one speaker. That way you can run your system at the same level that Amir experienced his issues. As long as the version of the track you listen to is the same one in terms of recording level as the one that Amir listened to then you can guarantee within a certain margin of error that your speakers are outputting the same SPL per speaker at 1m, and thereby you're stressing them to the same level at the different points that Amir experienced the tweeter issues. It's just to see if your system (whatever it is) has the same problems, it's a curiosity thing.

The calculated levels I came up with were as follows that Amir had his setup set to (apart from the -16.5dB, ignore that) for a 0dBFS sine wave at 1kHz at 1m for one speaker:

Then you'd just marry your system to those values and then listen to the track and see if you can identify tweeter issues, and then be happy that you don't have any hopefully. I tested my system up to 101dB at 1m and didn't notice any tweeter issues which would fall into the level of speaker stress where he had started noticing it. (Don't get carried away with the actual dB number I'm showing, it's a calibration point for a 0dBFS sine wave at 1kHz at 1m for one speaker, the music doesn't play that loud as the recording level is lower, it's literally just so you can calibrate your system to the same level)

 

[Mnyb]

If someone want to play "Fading Sun" in a way that highs are at 90dB SPL has to very well know that this 30-ish Hz low it has it may have to reach in excess of 110dB, you can see it in my chart.
Make sure that room and speaker can do it, although I very much doubt that anyone in the right mind would try that with the thread's speaker.

Understands Why amir uses this a sub bass test . Hardly ever heard such bass from any record, my dual subs did some actual subwoofing very dynamic piece , did not hear any kind of issues with the bells, in the midrange/treble .I get if this kind off bass melts lesser speakers . but no speakers should have problems with the rest of the track ? Hence I voted poor .

 

[Thomas_A]

Understands Why amir uses this a sub bass test . Hardly ever heard such bass from any record, my dual subs did some actual subwoofing very dynamic piece , did not hear any kind of issues with the bells, in the midrange/treble .I get if this kind off bass melts lesser speakers . but no speakers should have problems with the rest of the track ? Hence I voted poor .

Same here. Been using that song for quite some time to "test" speaker and headphones. It can be cranked up quite some. Another one which is a bit more busy;

 

[beeface]

Why are people reserving thread slots?

I'm reserving a thread slot on page 9. watch this space

 

[Putter]

Seeing as how the speaker is a mass-market item, couldn't you just poll some owners on other forums/sites for their experience?
I mean, it's ELAC; They must sell thousands of these sets!

Amir and Erin aren't the only trustworthy audiophiles in the world.



So true, huh?
Seems like you gotta spend quite a bit to not feel like you're 'settling'.

I feel like the Q7/Q11 Meta is in a real sweet price/performance spot though.

The point isn't that they are audiophiles whatever that may mean, but that they are one of very few with sophisticated measurement equipment, aka the Klippel.

 

[beagleman]

Since I own a few speakers that both Amir and Erin have measured and have also heard a few more they have done, its a matter of comparing MY subjective impressions between the Two of them and finding in relation to the measurements, for the most part I tend to agree with Amir a bit more than Erin, but still value BOTH guys contributions and greatly value their measurements and time they commit to doing this.

I am for the most part, able to ((at maybe 85% accuracy about)) tell if I am going to like or dislike most speakers now based on comparing their measurements to ones I own or have heard and their respective measurements.

This one here is the FIRST, I have reservations/confusion about, as the subjective and measurements are fairly at odds with each review.

 

[Henryk]

The point isn't that they are audiophiles whatever that may mean, but that they are one of very few with sophisticated measurement equipment, aka the Klippel.

But Amir emphatically stated that the issue in question is quite easily audible.
I mentioned "audiophiles" because I'm referring to someone who is a trained listener.

To confirm what Amir is saying, all we need is another trained listener; Not another Klippel measurement.
No?

 

[beagleman]

But Amir emphatically stated that the issue in question is quite easily audible.
I mentioned "audiophiles" because I'm referring to someone who is a trained listener.

To confirm what Amir is saying, all we need is another trained listener; Not another Klippel measurement.
No?

But there is one.

Erin did a review of this recently, and his subjective and objective were somewhat at odds. Not starting some debate about it, but usually things tend to align a lot closer is all.

 

[Henryk]

But there is one.

Erin did a review of this recently, and his subjective and objective were somewhat at odds. Not starting some debate about it, but usually things tend to align a lot closer is all.

I hear ya, 100%

All I'm saying is that the input of another trained listener would go a long way in clearing the matter up since the trait in question is clearly audible.
It would be much easier to get another trustworthy subjective data point, than another trustworthy objective one (another Klippel).

 

[dfuller]

Try this and tell me, even more complex : Casino royale by Derivakat

Crazy to think I went to college with her.

 

[Power Pop 23]

ELAC was willing to loan a DF63 to Erin's Audio Corner for measurement.

Perhaps the discussions in this thread will incentivize ELAC to contact Amir to discuss his review of the DF63.

 

[daniboun]

Same here. Been using that song for quite some time to "test" speaker and headphones. It can be cranked up quite some. Another one which is a bit more busy;

Nice song, added it to my playlist. Very good test and pleasant BTW !
Check this one too : It doesn't go as low but it's the kind of experimental music that's super cool to listen to on good speakers.

 

[Power Pop 23]

I hear ya, 100%

All I'm saying is that the input of another trained listener would go a long way in clearing the matter

Or @amirm could offer ASR members an annual 'Fantasy Camp' for listener training, meet and greets with select audio luminaries, and listening to well-reviewed gear in treated listening spaces - somewhere warm and dry in late January might appeal to Amir.... perhaps allowing him to explore new routes and locales in his mobile command center.