• 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!

PMC Twenty.21 Bookshelf Speaker Review

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
But all mono tracks recorded go through EQ when they are mixed over 2 channels, till they sound right to the creators, the correction will be automatically baked into the human operators feedback loop.

This. What the research has shown is that humans tend to prefer a certain sonic balance. Producers and engineers are humans. If something sounds too bright to them, then they adjust things.
 

Soniclife

Major Contributor
Forum Donor
Joined
Apr 13, 2017
Messages
4,499
Likes
5,417
Location
UK
Yes it is indeed addressing the point. If you accept the research that a perfect linear speaker evaluated in mono is the preferred one, you can mimic this sound in a stereo setup by applying slight EQ in the 1-10 kHz region. This will of course be a compromise, since a full compensation will apply to the center phantom image.
Again this does not address your claim that this is not done in the mixing stage, the only place this can be properly addressed, as it needs to be applied selectively to center images, not to hard panned tracks.
 

Thomas_A

Major Contributor
Forum Donor
Joined
Jun 20, 2019
Messages
3,422
Likes
2,406
Location
Sweden
Again this does not address your claim that this is not done in the mixing stage, the only place this can be properly addressed, as it needs to be applied selectively to center images, not to hard panned tracks.

If it would, then the research evaluating speaker response (in mono) has come to the wrong conclusions. If compensated for already at the source, e.g. for centered voices, the frequency response of an ideal centered mono speaker should mimic the curve presented in Shirley et al. So far I have not seen any evidence of this. In my own experience, such response, i.e. dip 1-2 kHz and peak 2-5 kHz, sound harsh and bright when used in stereo. An inverted curve, as e.g. can be seen in few reviewed speakers here on the forum, sounds both smoother and more natural, IMO.

https://www.audiosciencereview.com/...m-technical-picture-upload-thread.1452/page-2
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
Recordings aren't made just by listening on monitors in the studio. Headphones are used. A variety of speakers are used. I don't think the goal is ever to produce a recording that is "flat." It seems to me that the audio research done over the years has essentially found that a neutral, well-behaved speaker will sound the best in "most" environments. Then, we'll have to EQ from there to achieve the maximum audio nirvana.

Really? Where? All the studios I've ever been in track on main monitors and mix on mains/secondaries with a 'reality check' on nasty things like NS10s. Headphones are used by the artist(s) for tracking so as to maintain separation, not by the recording engineers.


S.
 

Listen first

Member
Joined
Jul 2, 2020
Messages
6
Likes
7
I am astonished that someone could say that and also say this

The first para quoted defies basic logic. You don't even need any specialist knowledge to see that.
There is significant evidence that we do perceive low frequencies as being quieter than mid-range frequencies. Whatever you determine that this "equal loudness" contour looks like if you flip it upside down you get a frequency response of the human hearing. Impose that frequency response on your measurement microphone and now you want to tune the loudspeaker differently to see a flat response on the analyser. Invariably you will end up with a bass tilt on the loudspeaker which is very common.
 

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
Really? Where? All the studios I've ever been in track on main monitors and mix on mains/secondaries with a 'reality check' on nasty things like NS10s. Headphones are used by the artist(s) for tracking so as to maintain separation, not by the recording engineers.


S.

I don't mean they are used in a "studio" sense, but I'm pretty sure tracks are listened to on all sorts of different equipment while things are being put together. I've read lots of anecdotal stories about the recording of classic albums talking about tracks being played for different audiences on different equipment like boom boxes and car stereos and the like by artists and producers. If something sounded messed up in that environment, I'd assume the issues would be investigated.

I guess what I'm saying is that recording/sound engineers have had many many years of experience with producing recordings in studios and then hearing what those recordings sound like "out in the wild." Presumably, they have learned how to end up with a recording that sounds good to people outside of their studio environment...although it's far from a perfect science and there's obviously a wide range of skill levels and expertise involved.

Also, now that I think about it I've seen lots of documentaries that have shown engineers in studios doing their thing and most of the working environments don't appear to be that far removed from a normal living environment. They have sound damping materials on the walls and what not, but there is furniture in the room and big consoles full of knobs and sliders...there's lots of reflective surfaces. They are pretty sonically controlled, but they certainly aren't anechoic chambers.
 
Last edited:
D

Deleted member 16502

Guest
That would make no sense to me, because you hear everything with the those dips. So if you hear a real non-recorded sound you hear it with those dips, and to create that with a loudspeaker you would need a perfectly flat response to reproduce it exactly the same. It wont ever be exactly the same, because the transducer is different....but it most definitely wont even be the same if the response is not flat. Maybe my train of thought is wrong though...

Good point, one would indeed again need a linear speaker response to find what one is used to. But is it conceivable that there are circumstances where you prefer something else than neutral compared to hearing the actual source? Assume your hearing has a dip in the frequency range of spoken word, would you then not prefer a speaker peaking in the same region for better understanding of voice?
 

Koeitje

Major Contributor
Joined
Oct 10, 2019
Messages
2,292
Likes
3,880
Good point, one would indeed again need a linear speaker response to find what one is used to. But is it conceivable that there are circumstances where you prefer something else than neutral compared to hearing the actual source? Assume your hearing has a dip in the frequency range of spoken word, would you then not prefer a speaker peaking in the same region for better understanding of voice?
Its basically Toole's circle of confusion: mastering and recording might not have been done on a neutral setup. So if you play that back through a neutral speaker you wont hear it the way the mix/master was meant to be heard.
 

infinitesymphony

Major Contributor
Joined
Nov 21, 2018
Messages
1,072
Likes
1,806
The pink noise curve is often used as a rough mixing and mastering target:

image-1.png


This curve results in a mix that sounds less "peaky" overall, especially on smaller speakers that can't reproduce bass well, and most people find that it sounds more "analog" or "vinyl-like." You'll find that this curve gets adjusted to suit certain genres. Electronic music tends to have less of a slope, resulting in slightly increased mid and significantly increased treble response. Metal music might leave the curve alone except for the treble where it's bumped up for emphasis or tone shaping. And so on.

Only translation matters in the end, unless you're working on a project where everyone has identical equipment and rooms.
 

Sgt. Ear Ache

Major Contributor
Joined
Jun 18, 2019
Messages
1,894
Likes
4,150
Location
Winnipeg Canada
Yeah, well with pink noise what you usually want to shoot for is a fairly flat line out to about 2500hz, then a slope ending maybe 10dbs down at 20khz. Something like that anyway, lol...
 
Last edited:

ROOSKIE

Major Contributor
Joined
Feb 27, 2020
Messages
1,915
Likes
3,394
Location
Minneapolis
Yeah, well with pink noise what you usually want to shoot for is a fairly flat line out to about 2500hz, then a slope ending maybe 10dbs down at 20khz. Something like that anyway, lol...
There are many, many "house curves".
I typically use flat from 120-1k, then a drop of about 5-6fb from 1k to 10k. Bass is shelved up about 5-10db(depending on playback volume, less when loud) from 120hrz down.
What is interesting is how good this "house curve sounds to me.
I have many speakers and one speaker set that is designed for flat in room response, it is very bright yet still sounds good to me due to various other qualities and the fact that good music can be involving despite flaws. When EQ'd for a downward sloping curve this speaker sounds much better than its natural condition. I suggest anyone who has this PMC, or another speaker that has "poor" response but they still like, try EQ.
Just give it a shot and see why folks on this forum find the PMC to be substantially under engineered and disappointing for $2k each. (To be clear I have never heard this model, only other speakers that measure poorly)
As well this speaker measures so poorly in other parameters it just seems vastly and unfairly expensive by any contemporary standard. 25 years ago sure, now? No way, not cool.
 

tonapo

Active Member
Joined
Oct 14, 2019
Messages
151
Likes
78
I suggest anyone who has this PMC, or another speaker that has "poor" response but they still like, try EQ.
Just give it a shot and see why folks on this forum find the PMC to be substantially under engineered and disappointing for $2k each. (To be clear I have never heard this model, only other speakers that measure poorly)

I have the newer PMC Twenty5 22 and I posted earlier about my situation. Its a great looking speaker but I didnt like the sound very much although I put that down to the room, so much so, I moved my kit from one room to another to try and get a better experience. Just after that, I also did what is being suggested by @ROOSKIE , where I measured the room and had Homeaudiofidelity.com generate some filters which are currently in use, loaded into HQ Player. As I have begun to understand a bit more about the subject I am now tempted to have another go myself in REW over the weekend.

My measurements in my current room are on the left, and on the right are the modified results - these graphs are supplied from homeaudiofidelity.com. Happy for comments, although I might not understand them!!
1593858684018.png
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
The pink noise curve is often used as a rough mixing and mastering target:

image-1.png


This curve results in a mix that sounds less "peaky" overall, especially on smaller speakers that can't reproduce bass well, and most people find that it sounds more "analog" or "vinyl-like." You'll find that this curve gets adjusted to suit certain genres. Electronic music tends to have less of a slope, resulting in slightly increased mid and significantly increased treble response. Metal music might leave the curve alone except for the treble where it's bumped up for emphasis or tone shaping. And so on.

Only translation matters in the end, unless you're working on a project where everyone has identical equipment and rooms.

This is the spectrum of Tracy Chapman's "Fast Car":

NTDodtz.png
 

tuga

Major Contributor
Joined
Feb 5, 2020
Messages
3,984
Likes
4,281
Location
Oxford, England
And that of Beethoven 9th's second movement (Suitner, Staatskapelle Berlin - Denon/VEB Deutsche Schallplatten):

EvKQAy2.png
 
Top Bottom