I would like to put it even more drastical:
Binaural recording is not suited for listening over speakers.
IMO one has to decide, whether the head related transfer
function (HRTF) comes in at the recording venue or at
home.
If we discuss the "sane limits" of speaker parameters,
we might already have enough difficulties in defining
those limits in terms of frequency response and polar
dispersion for "usual" stereo recordings .
To load the speaker (!) with compensational duties
due to binaural recording will lead us far astray from
ever finding some agreement on desired
speaker properties.
Kind Regards
As several point out, the room is an important determinant in what is heard; in my case, a very live living room with 4 or 5 dual slap echoes was the starting point. Wool wall to wall carpeting, and covering about 20% (?) of the wall area with wool carpeting helped a great deal, as did Duette honeycomb shades for the floor to ceiling glass wall.
Other factors; baffle step issues, and edge diffraction peaks. (A personal bete noire). Any ringing will be far more audible than a simple frequency response would indicate, which is why I prefer waterfall measurements to frequency response.
With those issues subdued to my satisfaction, the house curve I prefer is:
400 Hz +89
1 kHz +87
3 kHz +85
10 kHz +85.
And I wound up with about 1/2 db bump from 100 to 200Hz. All 1 meter, mic height midway between mid and tweeter, relatively short gate, typically 5 ms above 500 Hz.
Personal reference; mostly symphonic, primarily Davies Hall, San Francisco, but also small groups locally. For me, live music is important for timbral accuracy, and for spectral balance, and usually has no correlation with stereo staging.
Other factors; baffle step issues, and edge diffraction peaks. (A personal bete noire). Any ringing will be far more audible than a simple frequency response would indicate, which is why I prefer waterfall measurements to frequency response.
With those issues subdued to my satisfaction, the house curve I prefer is:
400 Hz +89
1 kHz +87
3 kHz +85
10 kHz +85.
And I wound up with about 1/2 db bump from 100 to 200Hz. All 1 meter, mic height midway between mid and tweeter, relatively short gate, typically 5 ms above 500 Hz.
Personal reference; mostly symphonic, primarily Davies Hall, San Francisco, but also small groups locally. For me, live music is important for timbral accuracy, and for spectral balance, and usually has no correlation with stereo staging.
Recently, I was away from home for four weeks and thus "unlearned" my dipole speakers, which had a flat FR then. After I returned I could immediately hear a remaining issue that needs fixing. In other words there is some shrieking especially obvious when playing female voices (e.g. Kate Bush) loud.
I can confirm that there is some blooming of the radiation pattern because I also use conventional dome tweeters for my dipole so the power response can't be perfect (see measurements on my projects page). And my initial reaction to the Orion shelving down filter also was "SL evens out the power response". But...the speakers I used before (closed box two way, 8"+1", xover @ ~1.7KHz, LR4) in the very same (pretty "live sounding") room at the very same place with the very same music created the same shrieking. I didn't pay too much attention to it back then because I was blaming the room. But I would say it was at a comparable magnitude.
So two speakers virtualy as different as they can be with the same issue while all the rest remained constant...hmmm...
Right now I have a down shelving filter applied to my dipole. Starting at only 6KHz (that is beyond the blooming) it slopes down 4dB and yes, there is a noteworthy improvement.
I find the tool that jlo posted on page 3 pretty interesting (and it confirms what has been said in this thread and my observation above): It indicates that potentially every real (non-ideal) speaker, which always poses a somewhat uneven power response/varying directivity could potentially benefit from a down sloping FR. Only if directivity remains constant, no such equalization is required presumed the room has a constant RT60.
So yes, fixing the directivity problem seems the best approach but nobody of us has a constant directivity louspeaker
And as long tweeters beam, equalization might be the only choice to at least cure the symptoms provided it creates a better outcome overall. And if it really does then I don't really care if I equalize to -2dB @20KHz or -4dB.
Oliver
I can confirm that there is some blooming of the radiation pattern because I also use conventional dome tweeters for my dipole so the power response can't be perfect (see measurements on my projects page). And my initial reaction to the Orion shelving down filter also was "SL evens out the power response". But...the speakers I used before (closed box two way, 8"+1", xover @ ~1.7KHz, LR4) in the very same (pretty "live sounding") room at the very same place with the very same music created the same shrieking. I didn't pay too much attention to it back then because I was blaming the room. But I would say it was at a comparable magnitude.
So two speakers virtualy as different as they can be with the same issue while all the rest remained constant...hmmm...
Right now I have a down shelving filter applied to my dipole. Starting at only 6KHz (that is beyond the blooming) it slopes down 4dB and yes, there is a noteworthy improvement.
I find the tool that jlo posted on page 3 pretty interesting (and it confirms what has been said in this thread and my observation above): It indicates that potentially every real (non-ideal) speaker, which always poses a somewhat uneven power response/varying directivity could potentially benefit from a down sloping FR. Only if directivity remains constant, no such equalization is required presumed the room has a constant RT60.
So yes, fixing the directivity problem seems the best approach but nobody of us has a constant directivity louspeaker
And as long tweeters beam, equalization might be the only choice to at least cure the symptoms provided it creates a better outcome overall. And if it really does then I don't really care if I equalize to -2dB @20KHz or -4dB.Agreed, it is an illusion, whose quality we're after.
If we can answer that easily AND have the proper products (i.e. drivers) handy then our hobby becomes obsolete and we need to find a new one
Oliver
Hi Dennis,
As you know, I have listened to the Orion in my home and I do agree, the original Orion was a pretty decent speaker. I never though it had the midrange of the NaO II but both speakers suffer from the polar bloom above the above the crossover point. With the NaO II it's maybe a little less severe because of the higher crossover point and narrower baffle. And I have introduced a couple of mods to the NaO II, but ultimately they were to improve sensitivity by moving some of the panel eq from the passive x-o into the active circuits. It wasn't to change the voicing of the speaker. I even played around with the on axis response to reduce the power radiated just above the crossover by splitting the C3 cap in the HP section of the passive crossover to a cap, in parallel with, a resistor in series with a cap. This reduces the on axis response in the 2 to 4kHz range. I offered this as a "suggested mods" to experiment with but never an official update. I have also encouraged builders to tweak the passive crossover if they think they can improve the sound to their liking (a benefit of a passive crossover).
An externally hosted image should be here but it was not working when we last tested it.
To be totally honest, while I have had differences of option with SL over the years, I think most people realize that I do have great respect for his knowledge and technical astuteness. Being critical and/or argumentative does not imply disrespect. But some how this whole 3.2 thing just makes me sad. I feel let down. It is possible that some may see these updates as the "master has an epiphany". I see it as more recognition of deficiencies which are now being addressed. But I believe they are being addressed in a manor with involves a great deal of compromise which is ultimately a result of being locked in to the physical construction of the speaker. I just feel that as a leader in this field, if SL were not locked into the Orion format he might well take a different approach. He has been successful in marketing the Orion and in doing that perhaps he has backed himslef into a corner. Congradualtions on the financial success, but that's not my objective. The 3.2 seems to me to really be a redesign more than an update. Due to my experience with the NaO II and the NaO Note, which I think is along the same progression with regard to the polar problems both the NaO II and Orion share (and I think ultimately the reason behind the 3.2 update), I think that the clean sheet of paper approach is the better way to go.
I could be way off base on this. I certainly can not read SL's mind so I don't know what he is thinking. But my mind just takes me to thinking that if the Orion was not marketed as, "the last speaker you will ever own" and having grown such a loyal following for the Orion, SL might be taking a different path. I mean to come back and say, sorry guys, the Orion was the next to last..... Instead we have the Orion active crossover is was, apparently, the next to last Orion crossover you will ever own, or will it be?
And again, I haven't read SL's entire page on the 3... series updates but some of what I have read just sounds like a little kid who just found out that tomorrow he is going to Disney Land and other parts just sound like attempts to justify the changes with somewhat related science. Frankly, that is in part why I haven't read it completely. I don't agree with what I read and I turn away. Maybe it is silly, but it leaves me empty and I just feel sad.
A lot of this is related to my experience in developing the NaO Note. As a result, I just can't buy the flat on axis response isn't right argument. Everything else about the 3.2 falls apart from there. As I have discussed in my earlier posts when eqed to the same, ruler flat on axis response the NaO II and Note sound different. And, in a given room, after on axis response (FR) what do you have left but polar response (PR) and radiated power (RP)? Given a target for radiated power vs frequency, fixing one of PR or FR then defines the other. It like FR = RP/PR, you have one equation relating 3 variables. Specify 2 and the 3rd one drops out.
Believe me the NaO II - Note comparison has me sitting back too though. I really didn't expect the differences I hear. I am some what perplexed and really couldn't say which speaker I prefer. As an example of the differences consider something like the recording of the drum kit on the first cut of the Sheffield Labs Set-up and Test Disk. On the NaO II the transients are very sharp with an electrostatic like impact and clarity. On the Note that are more subdued but the tones of the drum skins are richer and more robust, brushed cymbals shimmer. I like the way the NaO II sounds but when I listen to the Note I keep saying to myself that it sure sounds a lot like a drum kit in my room. It obviously has little to do with on axis response. And then there are some recording I like better on the Note, with others I prefer the NaO II. There is so much we simple don't have any control over. That takes me back to the live reference thing. Go out and buy 1/2 a dozen good recordings of the same symphony preformed in different concert halls by different orchestras, or in the same hall and the same orchestra but recorded by different engineers or mixed on different speakers. They all sound different. If I voice my speaker to sound more real for one of the recordings do you really think that voicing will carry over for the other 5? It ain't gonna happen.
Anyway, while I am not in flat ain't right camp, if the 3.2 update make the Orion sound better to those who make the effort and go to the expense to implement it, then that is a good thing regardless of my beliefs. I'll keep designing to flat and continue to provide the user the control to adjust the treble/bass levels as he/she prefers because we certainly all have individual preferences and differing rooms to deal with. Good as any speaker may be, its true worth can be measured only by the enjoyment it brings when listening to recorded music.
Last edited:
Kate Bush's voice shrieks at low volume. It just starts to hurt at higher volumes.
In my dipole line array i could get around blooming
of the radiation pattern, because i use fullrange drivers
there.
But a different problem occured, since the fullrangers
have no symmetric radiation into the front and rear
halfroom, which shows drastically above 4Khz.
I missed upper presence and brilliance (sparkle), but
if i added a few db's above the presence range to the
front radiation, the result was too bright
- which is the wrong word -
adding "sparkle" was accompanied by getting nasty
and it took a while until i accepted the reason why.
If there is a discontinuity in radiation pattern and energy
resonse - in case of my dipol 08 array there was a lack in
rear radiation above 4 Khz - you cannot trade energy
response for on axis response arbitrarily.
So i made a rear tweeter panel using 2 tweeters with
small vertical and even smaller horizontal offset, to
mimic the missing rear radiation pattern to some
degree.
A single tweeter was tested before and showed the
direction was right, but did not match perfectly.
The augmenting tweeters fill up the rear radiation
from upper presence upwards and give a "smooth"
compensation of the rear sound, which was too dull
before.
After that "radiation pattern fixing" it was no problem
to adjust presence and brilliance without the speaker
keeling over into "nasty".
I also could not judge a speaker using Kate Bush recordings.
It is a very 'produced' sound imo and i am not able to tell,
which ingredients of that screachy sound are due to her voice
and which are due to recording/microphone and 'special effects'.
It seems a case of actress and sound engineer working together
in a congenial manner.
It is a very 'produced' sound imo and i am not able to tell,
which ingredients of that screachy sound are due to her voice
and which are due to recording/microphone and 'special effects'.
It seems a case of actress and sound engineer working together
in a congenial manner.
Last edited:
Of course, some of us have directivity more constant than others.So yes, fixing the directivity problem seems the best approach but nobody of us has a constant directivity loudspeaker
[Or pretend to, at least....
]Flex Your PCD Mettle: - Page 23 - Techtalk at Parts-Express.com
It's unfortunate that the only directivity map we have of Orion was measured by a competitor:
Attachments
Last edited:
Interesting. I'm of the almost completely opposite opinion.
For me a "very good" system sounds like real people playing real music on real instruments in a real space. Be it recorded in a concert hall, studio or bedroom. Systems that please my ear sound like I'm listening to the "real thing." Sure, a lot of recordings are complete studio constructs, but the voices and instruments, often even the spaces can sound real.
Now I do own some recordings that are a complete fabrication. Switched on Bach or other W. Carlos recordings would be a classic example. The sounds only ever existed in an electronic form. Carlos was not trying to bring us the sound of the playback equipment. Just one example - and why I say "almost."
But anything that started as real sound usually doesn't lose its voice and timbre just because it was recorded and (over) engineered in a studio or edited in Pro Tools. Most of the recordings I listen to truly benefit from a more accurate or "real" sound. The "live event" might have been really live (I like those) or cut together from parts and pieces. But even those can sound surprisingly real.
A different point of view or taste, I suppose.
So yes, fixing the directivity problem seems the best approach but nobody of us has a constant directivity louspeaker
Oliver
Well, I do. At least from 1kHz up.
I liked the Pluto better, too. It has it's faults, of course, but does seem better balanced and more fun to listen to than the Orion. Remarkable little speaker.
I have to agree with Pano here. On a good system, even studio recorded music sounds real - by real, I mean real voices, real spaces and real instruments. That feeling of being there is unmistakable, and you don't have to go looking for it. Its just there.
Yea, you can't talk about speakers w/o talking about recording. I'll see ra7's assertion and raise it--I actually think studio albums have a tendency to sound better than live recordings and I'm far from alone on this. That's part of the advantage of recording in a studio. I do tend to prefer the live "performance" however but generally not d/t SQ.
Dan
Dan
I have to chuckle from time to time when I read posters saying they want their speakers to reveal the music just as the recording engineer heard it.
Could someone tell me just how can anybody buying CD's who wasn't at the recording session ever know how it sounded to the recording engineer?
No, no; it's how we IMAGINE the engineer heard it.
Still, not knowing the production details, the presumption that an accurate reproduction system provides the best approximation of the intent is reasonable, and implicitly, that means flat.
[Intent, however, DOES vary....
]
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.