Not really a problem, the shuffler doesn't take away from imaging at all. It merely gives your brain more time to process it.
The first signal hitting the ears remains the same. I still have the seamless stage without being able to directly point at the speakers on a (whole) lot of songs. There's always going to be exceptions.
Judging by the various level and quantity of work that is available, there are a lot of compromises made right there. But when done right it can paint a pretty compelling 3D picture. Not all recordings are created equally of coarse...
Take the song: Ashes to Ashes from David Bowie. A pop song (obviously) that does not try to mimic a recording of a band, but actually is a symphony of sounds and voices. But it works! There's a lot of freedom taken with the recording process, but it gets across.
I participated in almost every test hosted by X. It was a similar discussion that started these tests. "Can we measure what we hear" being the most important question surrounding that discussion. But how many people look past the IR that their measurement suite shows? How many look what happens at certain frequencies, filtering out the rest of the IR? What do we need to look for?
If we could first could agree on what that "state-of-the-art" represents....
Should we listen to Toole? Geddes? Linkwitz? Dunlavy? Danley? etc. etc...
I tried to listen to them all. And as far as I can tell up till now, they all make good and valid points. I still value phase more than most on that list but that's beside the point.
Or are we looking at the wrong list of names and should we look at:
Wilson Audio
Bowers and Wilkins
Scaena
Bang and Olufson
Magico
and many many more
I even tried to look at every one of those (and many more) to see what they are doing. But there seems to be a huge lack of information on the more expensive speakers, with a few exceptions...
Now I don't "think" I've got it all right, but I did get it better by using measurements. Not ever forgetting my ears, as they were also in heavy use.
To me, personally, it was quite remarkable that I could hear what was/went wrong with the earlier Rephase 2 phase shuffler. To me it means that we CAN hear all those tiny differences.
I haven't even started on judging amplifiers yet, I'm pretty sure a lot would disproof of the amp I'm using.
So Bob, tell me, what is the current-state-of-the-art?
Last edited:
What is it, that you sent from that iPhone?
Apologies, I really enjoy reading this thread because of the attention to detail and the quest to find the measurement techniques to tease out why certain things have a disproportionate impact on the sound we hear....
As I'm reading this on my phone surrounded by little people, sometimes the phone gets mashed and apparently it sends these blank messages - I don't know how to delete them, so please take them as a sign of interest and approval!
I will
...Honored to have you as a regular reader
. I was worried something I said scared you away .
.....I was worried something I said scared you away
Glad the happy ending story, was some curious too what happened.
Sent from my Desktop using optical mouse
My system of coarse... (just kidding).So Bob, tell me, what is the current-state-of-the-art?
I meant that in a general way. A "state of the art" system would be different for each venue or room. How a speaker interacts acoustically with the listening room is in my opinion the major weak link in most good systems. But it would use equipment and speakers that were recognized as being competitive with whatever anyone wants to call "the best". The best these days is pretty good, and hard to beat. That was my point.
I would bet that your speakers compete well with anybodies idea of "state of the art". My 3 way OB's aren't perfect in the room I'm in (the room is just barely big enough), but they do have quite a bit of magic to my ears. When the program material is high end, they mesmerize and astound me. I can't help but giggle with a nerd grin. I may have ruined my hobby of speaker building. I may not be able to top these babies.
Agreed. But most of that agreement is confined to the first 20-25ms. That is to say that most agree early reflective energy is more damaging than helpful. But after that, its a crap shoot. There is almost no consensus on what the acoustic response should be after that first (20-25ms).
One could argue that the quality of the source material (original recording) is the most important aspect of all. A great recording in a acoustically poor room probably sounds better than a poor recording in a great acoustic space. That says a lot to me.
Unfortunately, us end users have little control over this other than seeking out the best mastering of a particular recording available.
Last edited:
Wouldn't that be near impossible? Every room is different, even if we clean up the early 20-25 ms. So what room is ideal? Should we have a different room for different genres? If we look at venues and their purposes there are a lot of different preferences for pop/rock vs classical vs Jazz/blues etc...
(I know, a very broad statement, but I do not think we can define one good room)
A studio monitoring setup searches for neutrality and a clear sound to be able to judge. In listening we can search for that sweet room, fitting our choice of music. But I don't think there's a one size or shape fits all here...
It was my intended point to imply that sound after 20-25ms is more about subjective preference than acoustics. So we are in agreement I think.
In fact, a smooth and spectrally even decay is probably the most agreed aspect (if only aspect) of late energy. The slope and rate of that decay has a lot to do with what makes rooms sound different and that is the part that I think is mostly subjective preference.
Since 99% of us listen in small rooms, once early reflections are under control, the task becomes how big we want our small room to sound like. At least, this is where I am at.
Last edited:
I'll quote myself from the Phantom Center thread, still experimenting with processing to delay the comb pattern at the ear. The Choueiri IR wasn't giving me what I wanted, so I decided to make my own:
Remember, any Stereo setup with care taken to remove early reflections will get the top 7 pictures at the ear to some degree. That's an inherent flaw of the stereo triangle. Head shading will help, but measurements from Toole, among others, show this inter aural combing effect to be real.
The second set uses a delayed band passed copy of the original signal and mixes it in at low level with EQ on both the direct sound as well as that band passed injected (L+R) copy.
I still need to check a few things (how this influences my side signal chain for one). It might be best to apply it to both mid and side, I can't tell yet.
Did not get a chance to try it yet, this is just a theoretical example.
I've played around with Choueiri's IR and that one basically does the same thing. But it has a fixed EQ and isn't as clean in early waterfall plots as I'd like.
So I decided to make my own using lots of (linear) EQ, a delayed band passed (L+R) signal and lots of fooling around between JRiver and REW.
Here's a comparison:
I'll show early waterfall plots of my newest shuffler (I haven't tried this one "live" yet, but a raw pré version showed promise enough to spend some time on it)
First the pure dirac case (as any Stereo setup would exhibit):
Early waterfall of direct signal to ear:
(0.1 ms rise time)
(0.2 ms rise time)
(0.3 ms rise time)
After which we'd get the inter aural comb pattern (0.270 ms has passed)
(0.1 ms rise time of L + 0.270 ms delayed R)
(0.2 ms rise time of L + 0.270 ms delayed R)
(0.3 ms rise time of L + 0.270 ms delayed R)
(0.4 ms rise time of L + 0.270 ms delayed R)
After a lot of work and searching variables I came up with this:
Early waterfall of direct signal to ear:
(0.1 ms rise time)
(0.2 ms rise time)
(0.3 ms rise time)
After which we'd get the inter aural comb pattern (0.270 ms has passed)
(0.1 ms rise time of L + 0.270 ms delayed R)
(0.2 ms rise time of L + 0.270 ms delayed R)
(0.3 ms rise time of L + 0.270 ms delayed R)
(0.4 ms rise time of L + 0.270 ms delayed R)
With a bit more time I might be able to get an even flatter response at the ear, though I think this should do fine for preliminary testing. It seems to work just fine to get that longer "time frame" (for our brain) of the stereo signal at the ear with relatively flat FR balance.
I can translate this to a FIR filter if anyone wants to run some tests.
This one has some extra (very mild) EQ as it's my center signal from the mid/side processing chain. I still need to check/update my side signal to match.
Remember, any Stereo setup with care taken to remove early reflections will get the top 7 pictures at the ear to some degree. That's an inherent flaw of the stereo triangle. Head shading will help, but measurements from Toole, among others, show this inter aural combing effect to be real.
The second set uses a delayed band passed copy of the original signal and mixes it in at low level with EQ on both the direct sound as well as that band passed injected (L+R) copy.
I still need to check a few things (how this influences my side signal chain for one). It might be best to apply it to both mid and side, I can't tell yet.
Did not get a chance to try it yet, this is just a theoretical example.
My understanding of what your talking about here is limited. I have read enough to know there is some substance to the phenomena (the 2.6K dip and so on). I have no idea whether its worth fixing or not or if its that noticeable once it is fixed. Your the pioneer on this topic and so I look forward to your listening perceptions once you think you got it tackled.
As for me, I dont have the tools to address this issue even if I wanted to.
As for me, I dont have the tools to address this issue even if I wanted to.
As soon as I find a working recipe, I could convolve you a song to try it.
Maybe 3 versions, one plain, one with the cross talk eliminated (eh, more like delayed) and a third with some minor mid/side EQ added. You can pick the favorite, your setup should be excellent for this kind of test.
Maybe 3 versions, one plain, one with the cross talk eliminated (eh, more like delayed) and a third with some minor mid/side EQ added. You can pick the favorite, your setup should be excellent for this kind of test.
I've heard it mentioned many times that eliminating reflections during the first 20mS is very important. That's the delay range where the Phase Shifter, Flanger and Choruser guitar sound effect pedals do their thing. I don't disagree, but I don't know why reflections after 20mS are any less damaging. At 20mS you'd sense a "doubling" effect. From 50mS - 150mS intelligibility is poor. With any delayed signal mixing back in with the direct signal, you get comb filter effects.
There's the question of what is the perfect listening room (?). I tend to agree with Linkwitz on that, where he basically says a regular living room that will have many random reflections may be about as good as it gets. The only way to minimize the damage created by one reflection is to have enough other random delay reflections that they all largely fill in each others cancellations at the listening position. Different delays means cancellations happening at different frequencies.
Resonance is a different thing. Corners and parallel walls that ring have start up and decay times that aren't zero. This ringing will vary depending on the duration of a musical note or drum beat. Optimizing for transients doesn't do any good for sustained sounds. Optimizing for sustained sounds leaves transients to be a problem.
So perhaps the best listening room would have no parallel surfaces, absorbtive material in all corners, but otherwise be relatively reflective in a random way. If intelligibility is important, you wouldn't want to have any of the reflections involve delays between 50mS - 150mS (according to David Griesinger, formerly of Lexicon), so distances over about 25 feet could do damage to that. Walls that are beyond 25 feet should probably be thoroughly damped in a way where the damping is substantially effective down to at least 100HZ.
There's the question of what is the perfect listening room (?). I tend to agree with Linkwitz on that, where he basically says a regular living room that will have many random reflections may be about as good as it gets. The only way to minimize the damage created by one reflection is to have enough other random delay reflections that they all largely fill in each others cancellations at the listening position. Different delays means cancellations happening at different frequencies.
Resonance is a different thing. Corners and parallel walls that ring have start up and decay times that aren't zero. This ringing will vary depending on the duration of a musical note or drum beat. Optimizing for transients doesn't do any good for sustained sounds. Optimizing for sustained sounds leaves transients to be a problem.
So perhaps the best listening room would have no parallel surfaces, absorbtive material in all corners, but otherwise be relatively reflective in a random way. If intelligibility is important, you wouldn't want to have any of the reflections involve delays between 50mS - 150mS (according to David Griesinger, formerly of Lexicon), so distances over about 25 feet could do damage to that. Walls that are beyond 25 feet should probably be thoroughly damped in a way where the damping is substantially effective down to at least 100HZ.
its recommended -10 to -15db within 20ms at listening position. if you acheive more or less that target, your brain will be able to distinguish between the source and the room and will be able to tune out the room much more efficiently.
early reflections are always detrimental for well explained reasons
secondary reflections are often okay and actualy desired iif brought back to the LP enough reduced.
There's the question of what is the perfect listening room (?). I tend to agree with Linkwitz on that, where he basically says a regular living room that will have many random reflections may be about as good as it gets. The only way to minimize the damage created by one reflection is to have enough other random delay reflections that they all largely fill in each others cancellations at the listening position. Different delays means cancellations happening at different frequencies.
the only way to minimize the damage done by early reflection is by absorbing them so they are within -10 to -15db at 20ms at the listening position.
Resonance is a different thing. Corners and parallel walls that ring have start up and decay times that aren't zero. This ringing will vary depending on the duration of a musical note or drum beat. Optimizing for transients doesn't do any good for sustained sounds. Optimizing for sustained sounds leaves transients to be a problem.
room resonance and modal ringing is basically saying the same thing, no? you want to have even decay at all frequency. when you are able to have even decay (ETC) more or less, you will have natural decay and attack.
So perhaps the best listening room would have no parallel surfaces, absorbtive material in all corners, but otherwise be relatively reflective in a random way. If intelligibility is important, you wouldn't want to have any of the reflections involve delays between 50mS - 150mS (according to David Griesinger, formerly of Lexicon), so distances over about 25 feet could do damage to that. Walls that are beyond 25 feet should probably be thoroughly damped in a way where the damping is substantially effective down to at least 100HZ
even with no parallel surface, youll have early reflections. unparallel but symetrical walls can be good especially for bass, but for FR over 500hz, you still need to deal with early reflections.
what all room need is a free reflective zone and yes deep bass traps in all corners (and making sure you front the bass traps with reflective material so you dont make your room too dead by overdamping high frequencies compared to LF which would = uneven ETC FR). and dont forget proper speaker placement which need the speakers at least 4-5 feet away from the back wall and a suitable listening position well away from walls. thats a good start. for more, follow Jim adventures lol
early reflections are always detrimental for well explained reasons
secondary reflections are often okay and actualy desired iif brought back to the LP enough reduced.
There's the question of what is the perfect listening room (?). I tend to agree with Linkwitz on that, where he basically says a regular living room that will have many random reflections may be about as good as it gets. The only way to minimize the damage created by one reflection is to have enough other random delay reflections that they all largely fill in each others cancellations at the listening position. Different delays means cancellations happening at different frequencies.
the only way to minimize the damage done by early reflection is by absorbing them so they are within -10 to -15db at 20ms at the listening position.
Resonance is a different thing. Corners and parallel walls that ring have start up and decay times that aren't zero. This ringing will vary depending on the duration of a musical note or drum beat. Optimizing for transients doesn't do any good for sustained sounds. Optimizing for sustained sounds leaves transients to be a problem.
room resonance and modal ringing is basically saying the same thing, no? you want to have even decay at all frequency. when you are able to have even decay (ETC) more or less, you will have natural decay and attack.
So perhaps the best listening room would have no parallel surfaces, absorbtive material in all corners, but otherwise be relatively reflective in a random way. If intelligibility is important, you wouldn't want to have any of the reflections involve delays between 50mS - 150mS (according to David Griesinger, formerly of Lexicon), so distances over about 25 feet could do damage to that. Walls that are beyond 25 feet should probably be thoroughly damped in a way where the damping is substantially effective down to at least 100HZ
even with no parallel surface, youll have early reflections. unparallel but symetrical walls can be good especially for bass, but for FR over 500hz, you still need to deal with early reflections.
what all room need is a free reflective zone and yes deep bass traps in all corners (and making sure you front the bass traps with reflective material so you dont make your room too dead by overdamping high frequencies compared to LF which would = uneven ETC FR). and dont forget proper speaker placement which need the speakers at least 4-5 feet away from the back wall and a suitable listening position well away from walls. thats a good start. for more, follow Jim adventures lol
Last edited:
If you've done the work to avoid/absorb or diffuse the first reflections so that it meets the criterion youknowyou mentions the later reflections will most probably be lower in level already (at least most of them were in my case). Having a Haas kicker firing from side/behind further helps mask any reflections there.
It does work to create a virtual tail that hides most of the real room you're in.
I know I should have at least one bass trap (I have none) to improve my setup but I've managed to work around that by shifting energy to the other speaker that does not suffer from a room mode. I believe the Arrays help soften the room modes due to their height alone. I have one big anomaly in my left channel at ~70 Hz, but still can get even bass by having the right work just a little harder there.
As I've said countless times, in a family room it's hard enough to get "permission" to put up any damping panels. I made a promise not to add any more panels and will stick to that. But I know I could ....
Lol, I just caught that part.... Bob, if you haven't done so already, check the thread in jim1961's signature... I believe/think Jim has a pretty good handle on the room part! And that's a big understatement...
Back to the regular content;
I have multiple options to try my new IA comb delayer. I can mix it back in to the center only or I can de-convolve it to an IR to run it on the sides as well. I'll try it on the phantom center alone for starters. In an experiment which led to this newly created IA comb delayer I had it on both mid and side and it definitely worked on the mid content (very crude FR at that time) but seemed to let the hard panned sounds stick to the speaker locations. Not completely, like the very first shuffler, but enough to notice it immediately. Though it might work different with a FIR IR as it would no longer be a (L+R) mix like I have setup now. Do you fellows always run out of time like me when doing experiments like this? The days are just too short, and sleepless nights from puzzling about this all doesn't help either!
To bad I can't blast the music out loud in the middle of the night. Though, thinking about it, that would probably keep me up every night!
It does work to create a virtual tail that hides most of the real room you're in.
I know I should have at least one bass trap (I have none) to improve my setup but I've managed to work around that by shifting energy to the other speaker that does not suffer from a room mode. I believe the Arrays help soften the room modes due to their height alone. I have one big anomaly in my left channel at ~70 Hz, but still can get even bass by having the right work just a little harder there.
As I've said countless times, in a family room it's hard enough to get "permission" to put up any damping panels. I made a promise not to add any more panels and will stick to that. But I know I could ....
Lol, I just caught that part.... Bob, if you haven't done so already, check the thread in jim1961's signature... I believe/think Jim has a pretty good handle on the room part! And that's a big understatement...
Back to the regular content;
I have multiple options to try my new IA comb delayer. I can mix it back in to the center only or I can de-convolve it to an IR to run it on the sides as well. I'll try it on the phantom center alone for starters. In an experiment which led to this newly created IA comb delayer I had it on both mid and side and it definitely worked on the mid content (very crude FR at that time) but seemed to let the hard panned sounds stick to the speaker locations. Not completely, like the very first shuffler, but enough to notice it immediately. Though it might work different with a FIR IR as it would no longer be a (L+R) mix like I have setup now. Do you fellows always run out of time like me when doing experiments like this? The days are just too short, and sleepless nights from puzzling about this all doesn't help either!
To bad I can't blast the music out loud in the middle of the night. Though, thinking about it, that would probably keep me up every night!
Last edited:
To "clean" the first 20mS, you'd have to locate the speakers roughly 10 feet from any reflective surfaces... Not real practical in most living rooms, especially if there's a wife. With my open baffle speakers I find that they sound much better and image quite well as long as they are at least 3 feet (6mS) away from any walls. 20mS might be better but 6mS seems to be a "transition" distance, from my experience. Where the improvement becomes particularly audible. Linkwitz agrees with this number as well.
I get by with my baffles 50 cm away from the wall, a few big absorption panels and some DSP. It's pretty clean in that first 20 ms, even when you look a bit further than the standard IR representation. The plots show it as it cleans up FR and Phase plus the early waterfall plots. Could I do better? Perhaps, but not without losing other valuable things, like my lovely family .
I haven't seen plots from an OB taken at the listening position, so I cannot compare. I've seen Synergies doing very well though.
.I haven't seen plots from an OB taken at the listening position, so I cannot compare. I've seen Synergies doing very well though.
Not true. Absorb at boundary or redirect away from the LP.
This, plus spectrally the case from at least 250hz and up.
it means nothing to say 20mS or 6mS. how many db down at the listening position in 20 ms is what matters.
To "clean" the first 20mS, you'd have to locate the speakers roughly 10 feet from any reflective surfaces...
this is only true if you dont have treatment. your speakers can be closer to a wall with treatment and with proper treatment, at the listening position, you will reach -10 to -15 db within 20ms.
furthermore, what you are describing is about bass frequencies. when your speakers are too close to the back wall, closer then 3 feet, you get very bad bass cancelation.
however, even if you place your speakers 10 feet away from that back wall, you still need to reduce the reflection coming from the side walls, the wall behing the LP and the ceiling and the floor. all those resonacen must be -10 to -15db within 20ms at the listening position.
when it comes to SBIR, if you have well treated room, if your speakers are at least 5 feet away from the back wall, this will bring the first null high enough so its not too big a problem.
https://www.gearslutz.com/board/studio-building-acoustics/584207-sbir.html
20mS might be better but 6mS seems to be a "transition" distance, from my experience
early reflections are detrimental and those are all reflections that are not -10 to -15db within 20ms IIRC.