Salem Oregon conference center studio

#16

Here is today's fresh REW ring of the control room. The desk is fully populated and two video monitors have been added since the last ring. Still, there are NO RFZ left and right panels (first reflection). Any help with this is fully appreciated.

CR_room_ring_16May2021.mdat: (5.53 MiB) Downloaded 996 times

CR_room_ring_16May2021.mdat: (5.53 MiB) Downloaded 996 times

Also any advice with how to deal with the live room acoustics is also appreciated.

Thanks for your encouragements.
-Ron

#17

I've had a quick look. Initial thoughts;

You have some reflections in the first 20ms of the impulse response that require dealing with. They are present in both L and R sweeps. Within the first 20ms or so these should be less than -20 dB. Above that and they will interfere noticably with the direct sound from the speakers reaching your ears at the mix position.

You can figure out where they are coming from using Stuart's "string trick" and deal with them accordingly. He describes the method in this post. Note, it only makes sense to do this one speaker at a time. Don't use impulse charts for L + R to do this. viewtopic.php?p=1373#p1373

Another thing that immediately struck me is the weedy bass response from 40Hz to 95Hz. Any idea what might be causing this? there doesn't seem to be anything out of sorts with the phase in that region. Are you missing a sub woofer or something

Or is a low shelf cutoff engaged?

Live room acoustics are less stringent than those for the control room. There is no single recipe for success; it depends on what you are recording. An intimate, whispered vocal requires a different setup than booming 80s rock drums. In my experience;

- To some extent, if it sounds good it is good.
- Variables acoustics are a huge boon. The practical man on a budget's access to this is via gobos, one side absorbant, the other reflective.
- You discover things about the room over time. The good spots for certain instruments, and so on.
- A reflective floor is essential.
- Pay attention to side and ceiling reflections that are detrimental, especially if you have a low ceiling.
- Bass trapping can be very useful to reduce the influence of room modes causing some notes from bass & tenor instruments to stand out more than others.

I wouldn't sweat the live room stuff too much. If you get the fundamentals correct - decent bass trapping, reflective floor, sort out nasty comb filtering causing reflections - you can alter things pretty easily as you figure the room out.

Cheers!
Jennifer

#18

I've had a quick look. Initial thoughts;

Hey Jennifer! I owe you a debt of gratitude for taking the time to look at my REW data and ESPECIALLY for alerting me to Stuart’s “String Trick” method. It got me out of a stalemate big-time! THANK YOU!

I hadn’t understood anything about impulse data and didn’t even know about the 20ms / -20dB rule (and now understand Stuart’s handle). That was EXACTLY what I needed to hear! I went to work on it and identified the reflection sources. This string method is so simple that it’s genius!! :lol:

I went ahead and set up temporary materials and retested.

Here is an Excel spreadsheet listing the points of concern with original readings and after-treatment readings.

Impulse_analysis_25May2021.xlsx: (528.99 KiB) Downloaded 1077 times

Impulse_analysis_25May2021.xlsx: (528.99 KiB) Downloaded 1077 times

Example: Before treatment – left speaker

Example: After treatment – left speaker

I was so jazzed to see how well this method nailed these places! :yahoo:

Here is the new REW data after treatment.

CR_room_ring_25Feb2021.mdat: (5.53 MiB) Downloaded 1026 times

CR_room_ring_25Feb2021.mdat: (5.53 MiB) Downloaded 1026 times

I think I can improve the borderline reflections from the cloud by tilting it further. It is only tilted about 4.7 degrees if my memory serves me correctly.

You have some reflections in the first 20ms of the impulse response that require dealing with. They are present in both L and R sweeps. Within the first 20ms or so these should be less than -20 dB. Above that and they will interfere noticably with the direct sound from the speakers reaching your ears at the mix position.

I should first point out that the left and right first reflection RFZ panels were not present in the data you looked at, so that isn’t a big surprise. I just didn’t know how to test for it, but now I do. I had been wondering if the wall/ceiling corners would need treatment and this test answered my questions well. The interesting thing is that the wall/ceiling corner above the right RFZ panel is 90 degrees….,

…but the Left side is at 109 degrees (sloping ceiling) with a heat pump in the area. The left ceiling didn’t present any reflection problems and I can now see why.

Question: The spikes seen at 180 µs, to 1080µs on either side are traced to the desktop. Would I be correct assuming that the human ear cannot detect that short of a delay and I can disregard it? If my memory is correct, I thought I had heard in the past that anything below 2ms is undetectable. There isn’t much I can do for these reflections unless I want to cover the desktop with Knauf Ecose!. I actually did that as an experiment and it really worked. It’s added into the data I just posted today.

Another thing that immediately struck me is the weedy bass response from 40Hz to 95Hz. Any idea what might be causing this? there doesn't seem to be anything out of sorts with the phase in that region. Are you missing a sub woofer or something Or is a low shelf cutoff engaged?

By “weedy” I believe you mean low level, right? I have no sub woofer and there is no low shelf cutoff engaged. I have been ringing the room at each step of the way while adding acoustics and have noticed this frequency range dropped as soon as the console desk was installed. I’m sure that is the culprit but don’t have any ideas of what to do about that. Suggestions? Is actually RAISING the bass shelf on the monitors a stupid idea? They are at center now, of course.

They got a “consultant” in recently to look at our three live rooms. He didn’t seem to be interested in talking about data – only had advice to cover as much of the walls and ceiling as possible with 2” of Knauf and Guliford of Maine fabric (we already have the corner bass traps in). I’m very skeptical about that approach! First of all, it would be more expensive and ugly in my opinion. It reminds me of an adage about young girls trying makeup: “If a little is good, then a LOT would be better!”

I am picturing more like spreading multiple panels in attractive shapes around the walls and ceiling. What do you think? I would like to have something to counterpoint with here. Do people actually RING live rooms using REW?

I still can’t get over how quickly these reflections were dealt with as a result of your advice. Thanks so much!

Cheers,
Ron

#19

Fantastic work with the string test to find those problem areas! Nicely done.

Regarding the live room, I very much disagree with the idea of covering as much as possible with 2" of knauf and then covering that in fabric as that would make for a room that's very dead in the high frequencies but not in the lower. Live rooms don't need to be as dead as control rooms unless you're making 1980's disco music! I would advocate looking at a live/dead end approach, so that you have maybe one wall that is quite heavily treated (but 4" panels rather than 2") so that you have something for the singer to back up against (ie the mic points at this dead wall), but the other end of the room a bit less treated so that you have some reflections to add energy to your drum sound etc.

Perhaps the best move would be to make some 2' x 4' panels that are basically sheets of 4" knauf, with a wood surround and covered in fabric (you can see similar mounted to the ceiling in pictures of Jennifer's studio). You can then experiment with where to position these to get the sound you want. You might want a few on the ceiling and a few on the walls, to taste, basically.

You could REW the live room to look for problem frequencies, but otherwise I think it is more about how it sounds on recordings made in there...

Cheers
Gareth

#20

Guit-picker wrote:I hadn’t understood anything about impulse data and didn’t even know about the 20ms / -20dB rule (and now understand Stuart’s handle).

Well spotted, I had forgotten that is what it meant!

Excellent analysis. I really like the way you annotated that impulse chart, I will adopt that method myself in the future.

Question: The spikes seen at 180 µs, to 1080µs on either side are traced to the desktop. Would I be correct assuming that the human ear cannot detect that short of a delay and I can disregard it? If my memory is correct, I thought I had heard in the past that anything below 2ms is undetectable. There isn’t much I can do for these reflections unless I want to cover the desktop with Knauf Ecose!. I actually did that as an experiment and it really worked.

I tested this a while ago in another producer's room with a large desk, with and without insulation on the desk, and measured significant comb filtering in the midrange frequencies due to reflections of that type from the desk.

It doesn't bother some people, who find the ergonomic benefit of the large desk exceeds the acoustic penalty. I have heard it said that your ears are not a measurement mic (correct) and that a moving human head will "cancel out" the perception this comb filtering as it changes as you move. I have no idea how true this is.

Personally, it does bother me, so I have a very small desk.

Another thing that immediately struck me is the weedy bass response from 40Hz to 95Hz. Any idea what might be causing this? there doesn't seem to be anything out of sorts with the phase in that region. Are you missing a sub woofer or something Or is a low shelf cutoff engaged?

By “weedy” I believe you mean low level, right? I have no sub woofer and there is no low shelf cutoff engaged. I have been ringing the room at each step of the way while adding acoustics and have noticed this frequency range dropped as soon as the console desk was installed. I’m sure that is the culprit but don’t have any ideas of what to do about that. Suggestions? Is actually RAISING the bass shelf on the monitors a stupid idea? They are at center now, of course.

Weedy is UK slang for weak. I have just done a search and see it has other meanings in US slang :lol:

I'd try to figure out what is causing it before boosting the low shelf to compensate. You say it was when the console desk was installed. An acoustic or signal cause then?

They got a “consultant” in recently to look at our three live rooms. He didn’t seem to be interested in talking about data – only had advice to cover as much of the walls and ceiling as possible with 2” of Knauf and Guliford of Maine fabric (we already have the corner bass traps in). I’m very skeptical about that approach!

I agree with Gareth. This is in general a terrible idea, and due to excessive high frequency suck with little low frequency effect will give you a boxy lifeless padded cell of a room. By bringing many gobos from room 2 into my control room I can make the control room approach this in terms of high frequency suck. As Gareth says, it's useful if you want dead sounding recordings. But this is an unflattering sound for most instruments. Avoid.

Cheers,
Jennifer

#21

endorka wrote:
Guit-picker wrote:I hadn’t understood anything about impulse data and didn’t even know about the 20ms / -20dB rule (and now understand Stuart’s handle).
Well spotted, I had forgotten that is what it meant!

Excellent analysis. I really like the way you annotated that impulse chart, I will adopt that method myself in the future.

Question: The spikes seen at 180 µs, to 1080µs on either side are traced to the desktop. Would I be correct assuming that the human ear cannot detect that short of a delay and I can disregard it? If my memory is correct, I thought I had heard in the past that anything below 2ms is undetectable. There isn’t much I can do for these reflections unless I want to cover the desktop with Knauf Ecose!. I actually did that as an experiment and it really worked.
I tested this a while ago in another producer's room with a large desk, with and without insulation on the desk, and measured significant comb filtering in the midrange frequencies due to reflections of that type from the desk.

It doesn't bother some people, who find the ergonomic benefit of the large desk exceeds the acoustic penalty. I have heard it said that your ears are not a measurement mic (correct) and that a moving human head will "cancel out" the perception this comb filtering as it changes as you move. I have no idea how true this is.r

AHA! Another great forum moment... this is exactly what I experienced this week when testing my room. As I moved my head up and down above the desk (which is currently empty so just a flat piece of wood) I heard comb filtering. I hadn't realised it was comb filtering, I called it phasing on my own thread, but of course they are one and the same sometimes.

Maybe for both of us there's a case for moving to a small desk and then separate rack furniture out of earshot...

Cheers
Gareth

#22

garethmetcalf wrote:Source of the postAHA! Another great forum moment... this is exactly what I experienced this week when testing my room. As I moved my head up and down above the desk (which is currently empty so just a flat piece of wood) I heard comb filtering. I hadn't realised it was comb filtering, I called it phasing on my own thread, but of course they are one and the same sometimes.

"This magic day when super-science
Mingles with the bright stuff of dreams" - Countdown by Rush

garethmetcalf wrote:Maybe for both of us there's a case for moving to a small desk and then separate rack furniture out of earshot...

Works for me. A photo of my building site of a studio from a few weeks ago. The desk has enough space for a computer keyboard, trackball, Presounus Faderport, and TC Electronic BMC-2 monitor controller (out of sight behind keyboard). The rack has 3 x 8 channel mic preamps and headphone amps for musicians - 6 channels. A few people have described it as minimal, but it really is not; out of earshot, as you say, is more like it.

#23

Thanks for the replies! So I am still assuming impulse reflections aren't perceivable in the ranges below 2ms - especially below 1ms, but I didn't think about comb filtering. Again, I don't know what I can do about the desk reflections. I may have to just let it be. Also, the weak bass response below 95Hz is still a mystery. Open to any ideas on that....

Thanks for pointing out good reasons why not to have a totally dead live room! I don't recall ANY examples on these forums of people who have done that.
Cheers,
Ron

#24

endorka, In your photo, where are your speakers? (Just curious)

#25

Guit-picker wrote:Source of the post endorka, In your photo, where are your speakers? (Just curious)

At that time, in the other room to keep them away from dust :-)

They are back now in the approximate position they will go when I build the soffits.

#26

Hey Ron. All I can say is: WOW! :shock:

I'm impressed! Amazing progress there. Some really REALLY great workmanship, all over. Good planning, and good execution.

Where are you right now? Did you meet the impossible deadline? What still needs doing?

- Stuart -

#27

And a brief response from something you asked about a while back:

Question: The spikes seen at 180 µs, to 1080µs on either side are traced to the desktop. Would I be correct assuming that the human ear cannot detect that short of a delay and I can disregard it? If my memory is correct, I thought I had heard in the past that anything below 2ms is undetectable.

Actually, "detectable" isn't really the right way of looking at. Human hearing apparatus (two ears and a brain), can only pick out individual echos down to about 20 ms, or maybe 30 ms depending on the conditions and the person. In other words, if you hear a sound and its reflection after 40 ms, you can hear it as two distinct sounds. But below 20ms, you can't do that any more. Your ears and brain aren't able to "detect" that as two sounds. Instead, you hear them as one single sound that seems to be coming from a different direction than where it really came, and with a different frequency response curve from what it really was.

That's the problem, of course! And the reason why you do not want any reflections in the first 20ms: Because it messes up your ability to determine directionality and frequency. Not good for a studio! The exact effect depends on many factors, such as the actual time delay (Eg. is it 12.3ms, or 12.5ms? That's different....), the difference in level between the direct sound and the reflection, the frequency content of the sound itself, the person, head movement, etc.

In general, the effect is to make the sound field seem wider, a bit closer and slightly more shifted to treble.... but that's just a VERY general thing! Don't quote me on that! It can also go the other way entirely, especially for short time differences... there are so many factors in play. This is used to good effect in concert halls and theaters, where side wall reflections are a GOOD thing, as it makes the music sound more enveloping, brighter, more pleasant. And that is fine for concert halls... but NOT for studios! You do NOT want to "color" your perception of the sound like that! "wider" is great for the audience listing to an orchestra in the hall, but not so great for the mix engineer hearing the same orchestra in the control room. It would lead the engineer to make poor mix decisions.

So, the goal is get rid of as many early reflections as possible, or at least keep them down in level. Even very early ones, if possible. Of course, it isn't always possible, and you can get to the point where you are trying to kill reflections that aren't even under your control at all, and some that aren't even reflections! (eg, artifacts from the speaker itself). So the basic idea is just to do your best to get rid of the big ones, and the ones further out in time, then move on to other issues. You seem to have done a good job on that!

- Stuart -

#28

One other thing: There's a better way to look at impulse graphs when you are doing my string trick:

That's from one of your MDATS, for the left speaker. To get that, go to the "impulse" tab, set the graph Y-axis units to "%" instead of "dBFS", zoom in so you get only about 0 to 25% on Y axis, and whatever you need on X-axis (usually around 30ms is fine), and Bob's your uncle! That makes it easier to see individual reflections.

Aim to get all your big peaks down below the 10% line, and you should be fine. Under 8% is better, if you can.... under 5% would be awesome, but is really hard to do!

- Stuart -

#29

Hey Stuart, welcome back! So glad you're on the mend! I was getting worried!

Hey Ron. All I can say is: WOW! I'm impressed! Amazing progress there. Some really REALLY great workmanship, all over. Good planning, and good execution.

Well, thank you Stuart! I can't express how much that means to me!

Where are you right now? Did you meet the impossible deadline? What still needs doing?
- Stuart -

Well that depends on WHICH impossible deadline you are referring to! :lol:

There was the May 15th announcement that I heard about second hand, then there was the May 31st slip, then there was the July 4th deadline at the end of our "Family vacation/fellowship" gathering for NW USA (this week). I've been pouring myself out on it, knowing that it was insurmountable - even a couple of all-nighters. One good thing was that I got an expanded team and my good friend to act as "Project Manager", so I am not wearing as many "hats".

In the meantime, Salem Oregon hit the hottest temperature in the ENTIRE USA this week (117 degrees F - go figure! Brutal! Good thing the studio has two AC systems!), highest of all time for the state! That made the glue dry too fast for our production of acoustic panels. The leadership got a reality check and will schedule a ribbon-cutting sometime soon after July 16th. Funny- that's when I was guessing we'd be ready in the first place! I have to hold back from saying, "I told you so!" :roll:

Yesterday, just enough panels for the small live room (LR1) arrived, so we are about to begin mounting them. I will be proudly posting pics and data, etc as soon as I can, you can be sure!

Endorka said:

Another thing that immediately struck me is the weedy bass response from 40Hz to 95Hz.

Stuart,
One question nagging me that I want to continue working on AFTER this ribbon cutting thing blows over is the weak bass that started right after the console desk was installed. I am at a loss as to what to do. The only thing I can think of is to adjust the bass shelf on my Adam Ax7 speakers. Do you have any better ideas on that?
Cheers,
-Ron-

#30

Guit-picker wrote:Source of the post One question nagging me that I want to continue working on AFTER this ribbon cutting thing blows over is the weak bass that started right after the console desk was installed. I am at a loss as to what to do. The only thing I can think of is to adjust the bass shelf on my Adam Ax7 speakers. Do you have any better ideas on that?

Sure! We can take a look at that. Maybe you could do the "walking mic test", but you could limit that to just the region around the desk: a couple of feet each way from the mix position. Send me the MDAT (it wil be big!) and I'll take a look to see if there's anything obvious. Then we can figure out a plan.

Here's how: The "walking mic" test, using REW

In the mean time, be careful how you mix in there! You'll likely find yourself subconsciously boosting the bass a bit too much, to compensate for the "missing" low end... which will make the mixes a little bass-heavy in other locations.

Furniture sure can affect the response of the room. It happens quite often, actually. I have my suspicions as to the cause in your case, but I don't want to rain on your parade... yet! After the ribbon-cutting celebration thing, then I'll tell you...

- Stuart -

Soundman2020 - Studio Design Forum