• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Dynaudio Core 47 Review (Professional Monitor)

audio2920

Active Member
Joined
May 21, 2021
Messages
233
Likes
288
Sorry for derailing this thread, it wasn't my intent... I'll stop after this reply. Shan't be offended if anyone wants to delete...

Besides, there is the ultimate gotcha: you cannot put the Dolby logo on your film unless the soundtrack is mixed at a Dolby certified room. That fact makes all other arguments mute.

I find that pre-mixing at reference level gives the best translation for when we final mix. That's all. I don't think I'm alone in that.

It's only the final mix that needs to be done in a certified room if you want a Dolby license (heck, I've seen people get away with pretty much just the Printmaster these days...) but that doesn't mean we don't *want* to be able to work at reference level prior/post that part of the process.

How can there be an editorial process on a scripted film???
Not sure how to take this? Yes, films are edited. Both picture and sound get edited.
 

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
Sorry for derailing this thread, it wasn't my intent... I'll stop after this reply. Shan't be offended if anyone wants to delete...
My attempt at pulling this wayward thread back on topic to the Core 47 is this: based on the strengths and weaknesses of these measurements, it appears that one specific use case for the Core 47 is multi-channel mixing on-axis where wall reflections are mostly absorbed or diffused and all spatial cues come from the on-axis speakers. However, its power specs suggest that this would be great for midfield or reference level mixing right? You'd be wrong because measurements show things falling apart at 96dB:

Screenshot 2021-07-27 095544.png



So this ends up performing best as a nearfield speaker at 86dB from 1 meter, on-axis in a well treated room for mixing purposes only. If you require require wall reflections for soundstaging or spatial cues, don't get this speaker.
 

audio2920

Active Member
Joined
May 21, 2021
Messages
233
Likes
288
My attempt at pulling this wayward thread back on topic to the Core 47 is this: based on the strengths and weaknesses of these measurements, it appears that one specific use case for the Core 47 is multi-channel mixing on-axis where wall reflections are mostly absorbed or diffused and all spatial cues come from the on-axis speakers. However, its power specs suggest that this would be great for midfield or reference level mixing right? You'd be wrong because measurements show things falling apart at 96dB

Exactly my thought process, and then, that is disappointing to see at 96dB for sure!

Much appreciated @Spocko
 

829

New Member
Joined
Jul 27, 2021
Messages
3
Likes
4
Hi there!

I'd just like to comment on a few things here quickly. D's Jupiter measurement chamber is not fully anechoic, but semi-anechoic. Not sure which measurement error(s) this introduces but I'm quite sure their engineers are aware of it. Yes, D's philosophy of (not) publishing measurements is a bit strange - it seems they are not sure about which measurements correlate to the sound they are aiming (and tuning) for... Btw., they probably could fix at least on-axis frequency response easily as they provide firmware updates to their DSP engine! They also could shut Jupiter down, unless it provides better/more consistent data than a Klippel NFS (does it have ANY advantage over it?), get a Klippel NFS instead [older price list attached ^^] and transform Jupiter's room into a state-of-the-art laboratory for immersive audio; and listening room for better pro loudspeaker designs.

Furthermore, I'd not say that measurements aren't providing all required information and hence that loudspeakers could sound differently than they measure (e.g., 708P). I think the main issue still is translation of measurements. Everything is there - but we don't interpret it well enough. Research on psychoacoustics is still ongoing. If we take the 708P and compare it to the (flatter measuring) 8351B, we see a much cleaner waterfall plot below 1 kHz with the 708P, i.e., decay time is a lot lower [actually, throughout the whole spectrum]. This and the higher SPL capabilities, especially with deep bass frequencies made the 708P more attractive to me than the 8351B. Certainly, the 8351B has a clear advantage once vertical dispersion get as important as horizontal. A side not to @amirm, thank you for your great effort! I just find it a bit hard to compare some data as there seems to be inconsistency on how, and sometimes on which, measurements get published. May I suggest to fix that by establishing your own personal checklist and thus following the same procedure(s) with every review, all over again? Same order of plots, same sizes, same ... everything! :) If time permits, older threads might be adjusted; I guess RAW data is backed-up safely.

While reading through this forum I noticed, there seems to be a huge misunderstanding about which benefits proper acoustic room treatment offers. It definitely is not just about avoidance (or not) of ceiling/floor/wall reflections according to the Klippel early reflection and in-room response predictions. I don't want to go into details here; but all manufacturers of our best-reviewed loudspeakers do actually suggest (in their documentation) to use proper acoustic room treatment with their products [and they do know, why]. This includes HARMAN (JBL Pro, Revel etc.), Neumann and Genelec. On the other hand, proper acoustic room treatment should not be an excuse to disregard off-axis frequency response (D's excuse), as it cannot fix a loudspeakers' flaws.

Finally, are the Core loudspeakers bad products? I absolutely don't think so as they offer some unique features - but probably the 47 is the least attractive one. The Core Sub on the other hand looks quite promising, especially regarding the qualities D seems to have focused on, and there should be no directivity errors here. ;)

Guess what, it is much more painful to find well produced music than well engineered loudspeakers!
 

Attachments

  • KLIPPEL Analyzer System_Example Packages_NFS (2020_09).pdf
    76.9 KB · Views: 185

829

New Member
Joined
Jul 27, 2021
Messages
3
Likes
4
However, its power specs suggest that this would be great for midfield or reference level mixing right? You'd be wrong because measurements show things falling apart at 96dB:
Speaking of psychoacoustics; the line drawn at 50 dB on the 96 dB SPL graph is relatively useless, as it does NOT provide the threshold of audibility. Our hearing sensitivity is non-linear and worst in the bass region.

50 dB relative to the 96 dB SPL target just provides a delta of 46 dB, which indicates a relative THD of 0.5 %. The worst case on the 96 dB SPL measurements, in the bass region, is found probably at around 135 Hz and here we are still below(!) 3.8 % THD. I am not aware of any (trained) listener who can repeatedly distinguish such a low distortion at 96 dB target SPL and roughly 135 Hz, while music is being reproduced - so yes, the 47 is perfectly suited for higher playback levels. The shown (but still below) 1.7 % THD at around 2 kHz is more of a concern as this is closest to our most sensitive hearing range; but still perfectly acceptable.

It is a shame that the measurements for loudspeakers stop at 96 dB SPL reference @ 1 meter; where headphones are tortured up to 114 dB SPL. This makes sense in terms of available headroom as on contrary to loudspeakers, there simply exists no headphone that nails target frequency response and hence, EQ is a must!

An additional SPL measurement for loudspeakers, similar to what the German magazine Sound & Recording provides would certainly give a better idea of the SPL handling capabilities [with their targets set to a still conservative 10 % THD reference in the bass region and 3 % THD above; rather than using a fixed target SPL].
 
Last edited:

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,337
Likes
5,053
So this ends up performing best as a nearfield speaker at 86dB from 1 meter, on-axis in a well treated room for mixing purposes only. If you require require wall reflections for soundstaging or spatial cues, don't get this speaker.
I should clarify something here: In a pair (as you'd use this), 86dB per speaker ends up being more like 95dB (after room gain and doubling of volume from two speakers). That's way louder than I mix - I average around 83dB SPL between two speakers. It's legitimately tiring to listen long term louder than that.
 

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
2,999
Location
Southern California
I should clarify something here: In a pair (as you'd use this), 86dB per speaker ends up being more like 95dB (after room gain and doubling of volume from two speakers). That's way louder than I mix - I average around 83dB SPL between two speakers. It's legitimately tiring to listen long term louder than that.
Ooops, thank you for that correction, duly noted! So in LCR configuration the 86dB performance is more than enough for mixing purposes I would assume.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
They also could shut Jupiter down, unless it provides better/more consistent data than a Klippel NFS (does it have ANY advantage over it?),
Depending on how fast it spins, it could generate measurements much faster than Klippel NFS. I have to measure 1000+ times. They could measure 72 points which is just 16 times that arc. They could be finished in a few minutes vs a few hours for me. I can however generate far higher resolution data as I can go down to 1 degree spatial accuracy and they are stuck at 5 degrees or whatever the arc is configured at.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,388
Location
Seattle Area
A side not to @amirm, thank you for your great effort! I just find it a bit hard to compare some data as there seems to be inconsistency on how, and sometimes on which, measurements get published. May I suggest to fix that by establishing your own personal checklist and thus following the same procedure(s) with every review, all over again? Same order of plots, same sizes, same ... everything! :) If time permits, older threads might be adjusted; I guess RAW data is backed-up safely.
Thanks for kind words. On consistency, there is a lot of it. When there is not, is the tools. For example, until I upgraded to the new software from Klippel a week ago, there was no way to capture contour plots in fixed size. So I had to resort to screen capture and such which was both tedious and varied from graph to graph. Most of this is fixed with the new software but still are some holes like CSD graph.

I have also evolved the test over time so if you go back to very early reviews, you find a different set although the core spin graphs have been constant from day one.

The raw data is backed up in multiple places and computers/servers. I do plan to go even further in the future with that to assure longevity. Fortunately I export the spin information and a number of members have built databases on them so there is good redundancy there.
 

audio2920

Active Member
Joined
May 21, 2021
Messages
233
Likes
288
On the subject of distortion in the Cores, firstly I wonder how much worse that gets if you go up another 6dB or whatever it can take...

Now, this might sound stupid, but for a long time I've had a theory (backed up by no real science) that when I mix on monitors with very low distortion I find it harder to maintain a consistent level throughout the working day. When I play the mix back, I have to globally correct for my own drift.

I think this doesn't happen to me so much with higher distortion monitoring, and my current theory is that the brain interprets distortion as loudness, so, once you're subconsciously tuned in to that distortion, the brain kinda use that as a reference for "how loud is this in absolute terms?". Remove the distortion, and I lose that reference point a bit. The mix is still balanced relatively within itself, but it's harder judge exact loudness with absolute consistency across a 10+ hour day. When I play it back the next day (for example) I think, "this all needs to come up a dB here, down a dB there..."

(Or, on occasion, I sit there thinking "my meters are all in the red, and we're at reference level, but it just doesn't seem very loud!" probably because it's playing cleaner than I'm used to.)

Maybe I'm talking rubbish, it just "feels" like that's what happens. Of course there's a million other variables and I don't often get to hear multiple main monitors in the same studio, on the same day, on the same mix... But I wonder if part of the reason I like older Dynaudios, weirdly, is that they aren't super clean!?

So that then leads me to a question I can't believe I'm asking, but I wonder if a manufacturer would ever deliberately leave some distortion in to make their speaker more "exciting" or "edgy" sounding, and give me that a bit of "this is distorting, therefore it's loud"......?? Seems unlikely, but, I dunno.
 

heflys20

Active Member
Joined
Jul 24, 2021
Messages
115
Likes
98
Intentionally leave distortion to a $2500 speaker with DSP that is supposed to be "breathtakingly accurate?" Hopefully not. Lol. Measured high distortion would be considered a negative for potential buyers who haven't heard the product.
 
Last edited:

audio2920

Active Member
Joined
May 21, 2021
Messages
233
Likes
288
Intentionally leave distortion to a $2500 speaker with DSP that is supposed to be "breathtakingly accurate?" Hopefully not. Lol. Measured high distortion would be considered a negative for potential buyers who haven't heard the product.

haha! No, you're right, it wouldn't sell. Even I (with my stupid theory) would steer clear. I do think there may be an argument for super clean speakers actually not being that exciting to listen to loud, but no, that's not what you're after when you're paying for something described as "accurate".....
 

dfuller

Major Contributor
Joined
Apr 26, 2020
Messages
3,337
Likes
5,053
Speakers with low distortion make it very easy to creep level way, way up.
 

audio2920

Active Member
Joined
May 21, 2021
Messages
233
Likes
288
Speakers with low distortion make it very easy to creep level way, way up.

I guess that's a better summary of what I'm saying! And, I guess that matters a bit less in music mixing than it does for film/TV, because it'll largely come out in the wash at some point. Mastering at the latest, so long as you haven't messed up the dynamics too much as a result.
 
Joined
Jul 21, 2021
Messages
20
Likes
36
Actually Confidence series employ a waveguide
I'm aware of that, but it's used only in the models where the tweeter is located below woofers. Confidence 20, which is a two-way speaker, doesn't have a proper waveguide.

I got the same answer from them, when I asked for measurments of my Focus XD 600 speakers. I also asked them for measurements of the Core series. They do not publish them, but they do publish measurements from the Lyd series. Now that is not consistent.
Maybe LYDs are developed under a different leadership? The funny thing is that LYDs also measure quite well compared to Core, and they must be aware of that, because they publish the measurements for LYD, but not for Core.

Regarding Focus 60 XD (I assume you meant that model), those were my dream speakers not too long ago (Focus 20 XD as well). I have a feeling that Dynaudio's active speakers must measure better than passive ones... Yet they all have this outdated design, so I don't feel very confident about buying them anymore.

I'm sure you're enjoying them a lot though...
 

soundwave

Member
Joined
Jun 28, 2018
Messages
18
Likes
22
I'm aware of that, but it's used only in the models where the tweeter is located below woofers. Confidence 20, which is a two-way speaker, doesn't have a proper waveguide.


Maybe LYDs are developed under a different leadership? The funny thing is that LYDs also measure quite well compared to Core, and they must be aware of that, because they publish the measurements for LYD, but not for Core.

Regarding Focus 60 XD (I assume you meant that model), those were my dream speakers not too long ago (Focus 20 XD as well). I have a feeling that Dynaudio's active speakers must measure better than passive ones... Yet they all have this outdated design, so I don't feel very confident about buying them anymore.

I'm sure you're enjoying them a lot though...
No, I meant the Focus 600 XD. They are the frontrunner of the 60 XD. With the software update they sound identical. Only the finish is different. The only measurements I found of the 600 XD are these:

1627574645675.png


But I don`t know how reliable these measurements are.
I use them with a mini DSP and Dirac anyway. They go extremely low (18 hz) Very tight bass, because of the sealed construction.
 

RoA

Active Member
Joined
Jun 24, 2021
Messages
125
Likes
65
It's just like how people don't like the NS10's sound but many producers/engineers love them

The reason they 'love' them is not because they sound good, it's because it's a known quantity, its familiar. Some say they sound so shite that if you can listen to a bad recording through NS10's you can pretty much tolerate it anywhere else
 
Top Bottom