Hi Marcus..
Starting from LEDE and RFZ and ending with : Genelec Oy - DSP Monitoring Systems I really think that your answer could be not so obvious..
And, if I remember correctly, it’s curios that you point out about mastering process when you where the first linking about a not really defined mastering standardisation..
Audio Musings by Sean Olive: October 2009
http://www.glidemagazine.com/hiddentrack/wp-content/uploads/2009/09/bob_ludwig_1.jpg
Hmm I dunno what is that a ficus?
I guess most of the newer big names swear by room treatment but I am not so sure it isn't just as detrimental to the mix as "calibration" used to be.
Dan
I have to reiterate that there is no use in trying to analyze bad data. I told you how to get good data and you will just have to bite the bullit and do what I said to do - get a PC and HolmImpulse. Otherwise I see no reason to comment on what is most lkely just noise in the sense of trying to make any decisions about what the data means.
I have to reiterate that there is no use in trying to analyze bad data. I told you how to get good data and you will just have to bite the bullit and do what I said to do - get a PC and HolmImpulse. Otherwise I see no reason to comment on what is most lkely just noise in the sense of trying to make any decisions about what the data means.
Anyone know what is actually happening when you add the gating in REW? Seems you can't do it prior to making the graph. I know that the graphs look smoother and the bass starts dropping off, but is it actually a gated measurement or a simulation?
Dr. Geddes, I've got a PC, it's just a PITA to use for me. It's old and only speaks Japanese(long story). I didn't have much time today to do these, but I thought they should be useful--at least more so than before. Oh well, another day. I think I have to make more mistakes than most people before I get it.
Thanks,
Dan
Dr. Geddes, I've got a PC, it's just a PITA to use for me. It's old and only speaks Japanese(long story). I didn't have much time today to do these, but I thought they should be useful--at least more so than before. Oh well, another day. I think I have to make more mistakes than most people before I get it.
Thanks,
Dan
Last edited:
Dan
Certainly not more than most - just about typical. Most people don't understand how complicated this stuff can get.
What should I set the gating at? 1ms, 2ms,......10ms? Then for 300Hz and under it's best to just do it in their final destination not gated. Correct? I think I'm going to botch this one as well.
For the final install: Is there a best way to reduce floor and ceiling reflections other than eliminating the floor and ceiling? That would not make for a happy wife.
I'm starting to feel like I should send you a check. You have been extremely helpful.
Dan
For the final install: Is there a best way to reduce floor and ceiling reflections other than eliminating the floor and ceiling? That would not make for a happy wife.
I'm starting to feel like I should send you a check. You have been extremely helpful.
Dan
Last edited:
Thats exactly what HolmImpulse does so well - it sets the window for you. You can learn alot from that program. Save the check and buy a PC - they are as cheap right now as they have ever been. You almost can't pass them up.
Dan,
you never do a 'gated' measurement. Gating is always done afterwards. What you measure is an impulse response like this:
It shows the inital impulse response from 0 ms onward. Before it has decayed completely, it is superimposed by the first reflection (just after 2 ms). You want to 'gate' your impulse response in a way that the first reflection is just excluded. If your measurement program does not let you work with the impulse response diagram it is not worth to bother about.
you never do a 'gated' measurement. Gating is always done afterwards. What you measure is an impulse response like this:
An externally hosted image should be here but it was not working when we last tested it.
It shows the inital impulse response from 0 ms onward. Before it has decayed completely, it is superimposed by the first reflection (just after 2 ms). You want to 'gate' your impulse response in a way that the first reflection is just excluded. If your measurement program does not let you work with the impulse response diagram it is not worth to bother about.
Hello Jaco
The answer is indeed not obvious. The best you can do is mimic the acoustics of the typical control room - whatever that might be. We have to accept the fact that there is no adequate standard in music production hence there is no speaker-room system that is optimal for reproducing each and every recording.
Best, Markus
The answer is indeed not obvious. The best you can do is mimic the acoustics of the typical control room - whatever that might be. We have to accept the fact that there is no adequate standard in music production hence there is no speaker-room system that is optimal for reproducing each and every recording.
Best, Markus
Not sure we have to accept that. Earl and Markus have you taken a listen to Linkwitz's ideas about this? I know his designs are basically polar opposites to your own and a little anecdotal for your taste but I think between you 2 you might have stumbled onto a minimum prerequisite for accurate reproduction in a real room - basically controlling the polar response through the crossover region and an overall well behaved off axis frequency response on the horizontal axis.
I think Linkwitz's ideas presented in 2007 AES should be examined closer (more controlled) and maybe expanded to include controlled directivity and constant directivity.
I think Linkwitz's ideas presented in 2007 AES should be examined closer (more controlled) and maybe expanded to include controlled directivity and constant directivity.
Linkwitz (like Toole, like Geddes) claim that there is a "sweet spot" - the optimal direct-reflecting ratio, the optimal number, angle, delay, spectrum and level of reflections. I doubt that. We don't have enough knowledge to scientifically back up such a claim.
Not really: what Linkwitz's Orion does is to generate a strong lateral reflection that increases spaciousness. Olive showed that his trained listeners prefer that. But what are we talking about? Preference or reference? Without (re)production standards we're trapped discussing preference because there is no reference that can be objectively quantified.
Best, Markus
Well basically what I am looking for are speakers that operate independent of room reflections. Where basically the perceived FR is closer to the measured anechoic or free field response because the brain is actively removing or ignoring the late reflections in the room. Put subjective differences aside for the time being because it's unclear which is actually the most accurate in terms of reproduction. Basically I feel if it's at all possible that his claims are true then that needs to be tested because if it is true that we could be making much better recordings and have them translate to the home in more of a 1:1 fashion.
Currently if you were to look at the mixing room and mastering room as a mirror image of the playback room in reality the two images wouldn't correlate very well. The mixing room would have an abundance of treatment while the playback has none. There are a mix match of polar responses in a simple set of 2 ways let alone comparing two different types of different speakers which don't match up with mixing vs playback. Anyway if there are speakers which effectively let you blank out the rooms effects on the sound then finding out which ones actually work would be a giant step forward for accessible sound reproduction. Right now all we have is subjective opinions and anecdotal evidence which don't exactly hold up to close scrutiny. I think Linkwitz's model has the least holes in it and is the most accessible one.
Currently if you were to look at the mixing room and mastering room as a mirror image of the playback room in reality the two images wouldn't correlate very well. The mixing room would have an abundance of treatment while the playback has none. There are a mix match of polar responses in a simple set of 2 ways let alone comparing two different types of different speakers which don't match up with mixing vs playback. Anyway if there are speakers which effectively let you blank out the rooms effects on the sound then finding out which ones actually work would be a giant step forward for accessible sound reproduction. Right now all we have is subjective opinions and anecdotal evidence which don't exactly hold up to close scrutiny. I think Linkwitz's model has the least holes in it and is the most accessible one.
Shoot! REW will do what Dr. Geddes told me to do then. Still the way HolmImpulse works makes the most sense. With REW you'd have to make a lot more sweeps to get your full FR at optimal time window and you couldn't have them on one graph. To me none of this was obvious, but I'm learning.
I do some recording of myself playing keyed and stringed instruments. I like to check my masters mixes in the worst environment ever--the moving car. I like to make them on my Shure cans. I think more people today probably listen to music more in their cars then at home and thus the need to squash dynamics. So the quiet parts don't get lost in the road noise. If you record for maximum dynamics, you loose a lot while on the road and my little VW is quieter than any other car I've owned. If I only owned a Lexus or a Benz. Come to think of it, I'd prefer a Bugatti--ah! Road noise.
At risk of sounding like more of a moron, I've had some thoughts on how to reduce the VERs. Why not build a large ottoman with a thin wooden frame and filled with acoustic foam and cover in grill cloth. It should absorb a lot and still be a good place to rest your feet. As far as the ceiling goes, you could place a lot of thick acoustic material and make a false ceiling (like what Americans call a drop ceiling) again with grill cloth. I know there are people who do custom grill cloths for guitar amps that would even look incredible though pricey. You can get to look pretty much like anything you want.
Thanks again guys!
Dan
I do some recording of myself playing keyed and stringed instruments. I like to check my masters mixes in the worst environment ever--the moving car. I like to make them on my Shure cans. I think more people today probably listen to music more in their cars then at home and thus the need to squash dynamics. So the quiet parts don't get lost in the road noise. If you record for maximum dynamics, you loose a lot while on the road and my little VW is quieter than any other car I've owned. If I only owned a Lexus or a Benz. Come to think of it, I'd prefer a Bugatti--ah! Road noise.
At risk of sounding like more of a moron, I've had some thoughts on how to reduce the VERs. Why not build a large ottoman with a thin wooden frame and filled with acoustic foam and cover in grill cloth. It should absorb a lot and still be a good place to rest your feet. As far as the ceiling goes, you could place a lot of thick acoustic material and make a false ceiling (like what Americans call a drop ceiling) again with grill cloth. I know there are people who do custom grill cloths for guitar amps that would even look incredible though pricey. You can get to look pretty much like anything you want.
Thanks again guys!
Dan
We could go back and forth with papers all day long. But these papers in my opinion are outdated do to limitations in the transducers back then. I think a lot of old papers and conclusions need to be redone with more consideration for the polar pattern of the speakers. You must be careful not to be led astray by measurements that may or may not be audible.
I'm not sure we're talking about the same thing. These papers are not "outdated". It's valid data. Olive determined thresholds for a single lateral reflection in 3 different environments. See Olive et al. (1989), “The detection of reflections in typical rooms”, Journal of the Audio Engineering Society, S.539
Best, Markus
I have a paper or papers from Wireless World in the 1950s that support constant directivity. It's amazing how little is new. Advanced materials help to realize known old ideas, and computers help make it easier to speed up mathematical optimization, but engineers back then did very well without them. It's clear thinking that is the holy grail.
Last edited:
http://www.linkwitzlab.com/AES-SFO'08/accurate-stereo-rooms.pdf
zSHARE - Linkwitz AESUK_lecture_0711.mp3
Okay there are others which come to the same conclusions which predate this presentation. It is admittedly by him a bit anecdotal. But let me see if I can sum it up without injecting my own slant on the issue.
An Omni - not a dipole mainly because it is too hard to build a true dipole - is somewhat benign to the interaction of late reflections in a room and the perceived frequency response once you get past a certain threshold of around 6ms.
His life's work was on the Orion and the Pluto was actually an experiment with the aims of making a speaker that should sound completely different and excite entirely different modes in the room. He wanted to contrast the dipole with the omni since they are physically two different things but also have certain things in common - minimum diffraction effects and a frequency independent polar response.
The frequency independent polar response is where I believe Earl parts ways with his philosophy. Although if you look online subjective impressions are virtually identical in terms of how the in room response is perceived.
Back to SL
To his surprise they sound nearly identical in terms of Frequency Response despite measuring very different. There are some subjective differences in terms of the depth of the sound stage but it is unclear which sound stage is projecting the most accurate image.
It is also unclear to him if surround sound is more accurate than stereo although he subjectively prefers stereo. IMO this could be mainly the speaker placement and decoder being used which I have specific ideas about. And I believe this also if done right can further lessen the rooms influence.
zSHARE - Linkwitz AESUK_lecture_0711.mp3
Okay there are others which come to the same conclusions which predate this presentation. It is admittedly by him a bit anecdotal. But let me see if I can sum it up without injecting my own slant on the issue.
An Omni - not a dipole mainly because it is too hard to build a true dipole - is somewhat benign to the interaction of late reflections in a room and the perceived frequency response once you get past a certain threshold of around 6ms.
His life's work was on the Orion and the Pluto was actually an experiment with the aims of making a speaker that should sound completely different and excite entirely different modes in the room. He wanted to contrast the dipole with the omni since they are physically two different things but also have certain things in common - minimum diffraction effects and a frequency independent polar response.
The frequency independent polar response is where I believe Earl parts ways with his philosophy. Although if you look online subjective impressions are virtually identical in terms of how the in room response is perceived.
Back to SL
To his surprise they sound nearly identical in terms of Frequency Response despite measuring very different. There are some subjective differences in terms of the depth of the sound stage but it is unclear which sound stage is projecting the most accurate image.
It is also unclear to him if surround sound is more accurate than stereo although he subjectively prefers stereo. IMO this could be mainly the speaker placement and decoder being used which I have specific ideas about. And I believe this also if done right can further lessen the rooms influence.
OK, I did a smidge of research and found this: (distance=(time*speed)/2). Using that if I were to measure 1 meter away and placed my gate at 6msec, my graph should be accurate from 167 Hz on up d/t this equation (1/gate time). 5msec would make it useful from 200Hz, etc... Is that correct, or have I screwed this up? And yes, I still know I should just use Holm, but I like to understand all the little reasons why as best I can. It's the only way I'll get it through my thick skull.
Dan
Dan
- Status
- Not open for further replies.