Elias,
obviously you have looked into this in some more detail than I did. Is there a way for me to get a better understanding without attending university courses?
Well, I don`t need the "natural" effect (stereo isn`t natural either) - "convincing" would be enough. But it looks like "listening to stereo while turning ones head" hasn't been a subject of scientific scrutinity lately.
When I turn my head 45° to either side, I still hear the sound sources on the stereo stage - albeit less well defined as at 0°. They definitely have not been relocated to the loudspeaker position(s). Seems to be individual too. 🙄
Rudolf
It works...
Yet another update/change of mind ...
I think the DSS filter actually works. And I've been using/experimenting more since November last year. I view it now as a "tone control" somewhat like a simplified baxandall.
It makes things very easy as normally I'd just measure flat, then apply the DSS and adjusting the attenuation from 1db to 3db. I found that 3db is too much and there is an Orion owner who found the same:
oriongateway.org/orion-3-2-treble-shelf/
The implementation with DSP (MiniDSP) is simply a centre frequency if 1.5khz, Q=1, and attenuation of -1 to -3db, adjusting by ear. 1db makes a lot of difference and need to be done over time (e.g. a week of listening with -1db, then -2db, etc.).
I have done this with my dipoles (with waveguides) and also the Econowave variant. Same result/finding.
Will bump again if I find anything else interesting...
Yet another update/change of mind ...
I think the DSS filter actually works. And I've been using/experimenting more since November last year. I view it now as a "tone control" somewhat like a simplified baxandall.
It makes things very easy as normally I'd just measure flat, then apply the DSS and adjusting the attenuation from 1db to 3db. I found that 3db is too much and there is an Orion owner who found the same:
oriongateway.org/orion-3-2-treble-shelf/
The implementation with DSP (MiniDSP) is simply a centre frequency if 1.5khz, Q=1, and attenuation of -1 to -3db, adjusting by ear. 1db makes a lot of difference and need to be done over time (e.g. a week of listening with -1db, then -2db, etc.).
I have done this with my dipoles (with waveguides) and also the Econowave variant. Same result/finding.
Will bump again if I find anything else interesting...
I have found very similar results to what you have suggested.
In my experience with various DDS level settings, I've also found that while a certain attenuation (i.e. ~-2-3db) sounds very realistic at a given volume level (i.e. realistic levels that are fairly high... something like ~95dba), the same attenuation level is fairly lifeless when listening at a relatively low listening level. At the lower listening level (ie. ~80dba) an attenuation of ~-1-2db seems much more realistic to me. Further, I would say that even very small adjustments to the DSS are quite noticeable.
I'm still convinced, at least on average, that some DSS is a definite improvement over flat unless one is only listening at very low levels most often (~ -70-75dba) in which case a flat response is definitely sweet and engaging but not very realistic IMO.... and the same flat response becomes fairly harsh to my ears at higher listening levels.
In my experience, variations in the mix at the studio (often at the direction of the artists involved) are often more significant than 1-2db... many artists I know are hearing impaired at higher frequencies and thus it does not seem unreasonable to me that many would mix with a relatively high levels in the 2khz+ region and often those same artists would likely prefer "flat" speakers for their listening pleasure.
cheers,
Josh
Thanks Rob! That helps me make decisions.
Dave, you know as soon as you say a foot of fiberglass won't absorb down to 100Hz, someone has got to show you up.😛 The grey line here has no absorption, the blue line has a foot of fiberglass--80 inches by 50 inches. Ethan Winer(professional in the field who posts on many forums--esp recording forums Rigid fiberglass density tests) says that more area works better than more depth. I'd agree with him based on everything I've done so far. I'd bet you are right when it comes to a more normal sized panel, but I'm getting significant absorption down to 70 Hz and even a little below 60.
Dan
I've experienced very similar results by using cheap acoustic ceiling tiles to damp the reflected sound on the front wall, behind my dipoles and also on the rear wall.
Definitely a big improvement and also very cheap.
I think I'm going to make picture frames with the acoustic tiles to improve WAF. 🙂
Fellas,
I'm afraid the whole premise of this thread is incorrect. (At least as far as the magnitude of the DSS filter.)
The (overall) net change to the Orion ASP EQ setting regarding this "DSS" shelf filter also includes an update (increase) to the tweeter level setting for the Orion system. (If you're not a Linkwitz owner and don't have access to the full revision update on his website then you don't have all the information.)
Evaluating another speaker system that exhibits a "flat" response and then applying a -3.2db shelf filter is not the same thing and is a complete apples/oranges comparison.
I dropped my subscription to this thread and didn't notice. Sorry about that.
Cheers,
Dave.
I'm afraid the whole premise of this thread is incorrect. (At least as far as the magnitude of the DSS filter.)
The (overall) net change to the Orion ASP EQ setting regarding this "DSS" shelf filter also includes an update (increase) to the tweeter level setting for the Orion system. (If you're not a Linkwitz owner and don't have access to the full revision update on his website then you don't have all the information.)
Evaluating another speaker system that exhibits a "flat" response and then applying a -3.2db shelf filter is not the same thing and is a complete apples/oranges comparison.
I dropped my subscription to this thread and didn't notice. Sorry about that.
Cheers,
Dave.
Last edited:
Quoting SL:
"We therefore left the tweeter at its flat level setting and started to use shelving lowpass filters to shape the response. Immediately we found great improvement, but it took us many hours of listening"
So I too start with flat, then apply DSS.
This also conforms with the explanation from SL in regards to HRTF
Do you have an on-axis plot of the Orions with and without DSS?
"We therefore left the tweeter at its flat level setting and started to use shelving lowpass filters to shape the response. Immediately we found great improvement, but it took us many hours of listening"
So I too start with flat, then apply DSS.
This also conforms with the explanation from SL in regards to HRTF
Do you have an on-axis plot of the Orions with and without DSS?
Last edited:
Yes, I'm familiar with his description of the process evolution, but you're missing the point. 🙂
Did you look at the plot? Do you see a -3.2 db relative difference in the upper range of the two tweeter responses?
As I said, there is a concurrent ASP tweeter (leg) level increase that is also part of the Revision 3 modifications. So, you just can't apply a -3.2db (1.8khz) SLP EQ to any speaker and evaluate the "DSS" filter. It's dedicated to the Orion system and only with other modifications as well. Do you see what I mean?
Cheers,
Dave.
Did you look at the plot? Do you see a -3.2 db relative difference in the upper range of the two tweeter responses?
As I said, there is a concurrent ASP tweeter (leg) level increase that is also part of the Revision 3 modifications. So, you just can't apply a -3.2db (1.8khz) SLP EQ to any speaker and evaluate the "DSS" filter. It's dedicated to the Orion system and only with other modifications as well. Do you see what I mean?
Cheers,
Dave.
Last edited:
Hi Gainphile,
ORION-3 3.2.1 (scroll down if required)
Also read DB's summary.
The 3.2.1 eq goes nearly all the way through. I could imagine it is now a monotonically falling response that ends with the DSS, which is IMO, however the most important change, perceptually.
ORION-3 3.2.1 (scroll down if required)
Also read DB's summary.
The 3.2.1 eq goes nearly all the way through. I could imagine it is now a monotonically falling response that ends with the DSS, which is IMO, however the most important change, perceptually.
First . . . I consider what's happening above 10kHz (and probably above 5kHz) as pretty much inconsequential (regarding the effect being discussed). Second (and this especially in light of SL's comment regarding how small a change is necessary to be audible) . . . what we are hearing is *substantially* influenced by room effects that can swamp the equalization changes in magnitude . . . I can measure greater changes in my listening room simply by opening or closing one set of curtains. I am not at all comfortable with this Quixotic pursuit of a one-size-fits-all "perfect" equalization curve . . . maybe that representation is "good marketing" for the designer/manufacturer, but it doesn't represent my reality . . .
Deward,
I understand your point. 🙂
My point was addressing the misunderstanding (that I noticed) that a "DSS" EQ is a defined -3.2db 1.8khz shelving filter that could be applied to many different speakers. It is not.
SL's description of this on his webpage is confusing, so I can understand how folks (like Gainphile) got confused.
Just trying to clarify. Maybe I've muddied the issue even further? If I did, sorry.
Cheers,
Dave.
I understand your point. 🙂
My point was addressing the misunderstanding (that I noticed) that a "DSS" EQ is a defined -3.2db 1.8khz shelving filter that could be applied to many different speakers. It is not.
SL's description of this on his webpage is confusing, so I can understand how folks (like Gainphile) got confused.
Just trying to clarify. Maybe I've muddied the issue even further? If I did, sorry.
Cheers,
Dave.
A lot of hifi-wannabes talks of a flat frequense-response as if it was the holy grail or the answer to everything in sound-reproduction but it shure isn`t. Dynamic capasity, explosivity, specially in the lower end plays a big part in how a asystem appears.
That is not what SL said. Read his presentation here:
http://linkwitzlab.com/Presentations/BAF-Freq-response-requirements.pdf
Clearly his explanations are not Orion specific. Slide #10 and #11 are the most relevant.
The -3.2db DSS may not be generally applied to any speakers, but even if you have an Orion it may not apply to the room.
Orion 3.2 Treble Shelf Filter | Orion Gateway
Hence that's why I think:
- The concept is valid and universal, that high frequency energy needs to be attenuated, and the shelving filter works. Flat is not correct. Why? I don't know. It may be due to HRTF as SL explained, or simple power radiation issues as John K advised.
- The parameter of shelving (center frequency, attenuation level) may be speaker & room dependent.
In simplest term, it's the old "Tone Control". If you look at Baxandall's treble transfer function they look eeriely similar.
Note that I have investigated these with Dipole and Constant directivity box speakers.
You are hearing what I am hearing, so I guess being upside down doesn't effect the ears . . .
There is a benefit of more blood flow to brain/ear apparatus. Remember that BATS generally have excellent hearing. This is due to being upside down most of their lives.
😛
Gainphile,
In it's current iteration, the acoustic response of the Orion is "essentially" flat. It slopes down slightly, but it DOESN'T mimic the electrical response of a -3.2db SLP. In fact, if you disengaged that portion (and just that portion) of the (latest revision) Orion ASP the upper range of the acoustic response would be significantly ELEVATED.
The -3.2db shelving filter is part of the modification, BUT there's also a +2.9db level increase in the tweeter leg. So (relative to pre-Orion 3 configurations) the "tilt" of the acoustic response has only been changed 0.3db. It's a bit more than that because the midrange levels have been increased slightly.....but no where near 3.2db. (You can visually see that in the plot I posted.)
I was looking back through this long, wandering thread and posts #161/282 caught my attention. When you implemented the 3.3db shelving filter (to a flat acoustic response) and "immediately did not like it," it's not surprising. You applied a correction that is MUCH more drastic than the Orion 3 modification creates.
Unfortunately, much of this thread went off on a tangent based on incorrect information that you supplied. 🙂
Cheers,
Dave.
In it's current iteration, the acoustic response of the Orion is "essentially" flat. It slopes down slightly, but it DOESN'T mimic the electrical response of a -3.2db SLP. In fact, if you disengaged that portion (and just that portion) of the (latest revision) Orion ASP the upper range of the acoustic response would be significantly ELEVATED.
The -3.2db shelving filter is part of the modification, BUT there's also a +2.9db level increase in the tweeter leg. So (relative to pre-Orion 3 configurations) the "tilt" of the acoustic response has only been changed 0.3db. It's a bit more than that because the midrange levels have been increased slightly.....but no where near 3.2db. (You can visually see that in the plot I posted.)
I was looking back through this long, wandering thread and posts #161/282 caught my attention. When you implemented the 3.3db shelving filter (to a flat acoustic response) and "immediately did not like it," it's not surprising. You applied a correction that is MUCH more drastic than the Orion 3 modification creates.
Unfortunately, much of this thread went off on a tangent based on incorrect information that you supplied. 🙂
Cheers,
Dave.
- Status
- Not open for further replies.