kcatthedog
Temp
Super Helpful Dude
Posts: 16,082 
Member is Online
|
Post by kcatthedog on Sept 5, 2015 7:13:06 GMT -6
I did a freq sweep of my room ( some what treated) using the sonarworks ref 3 calibration software and its sdc mike. here is the graph of the L and R channels;  Could you help me interpret it ? - I have a null at 100 - 150-250 is a little high - 400-700 maybe a little too much broad band absorption Currently I have: - 2x4 x4 cloud above mix position and on either side, 3 in total and 1 cloud further back in my dead end - I have floor to ceiling 2x4x4 panels in the right angles of each corner behind my monitors ( 4 in total) - 2 2x4x4 panels staggered along opposite sidewalls behind my mix chair - larger 6x7x 4 inch panel across most of back wall my monitors are mounted on separate weighted stands Remedies : - build super chunk corner bass traps behind the monitors instead of what I have now ? - put wood slats over back panel helmholtz and tune for 150-250 ? - for null , experiment with shifting monitor position ? Other ? Thanks. |
|
|
|
Post by mulmany on Sept 5, 2015 8:15:51 GMT -6
Can that do a waterfall plot?
Another option would be trying to move some more panels to the front wall.
|
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 5, 2015 8:22:36 GMT -6
no but I have Room eq wizard installed and I can get one from it. I was curios to compare their two different sweeps but with the same mike interface ?  I was thinking that my corner panels could be moved flat against front wall and then I cut put proper corner bass traps in place ? |
|
|
|
Post by mulmany on Sept 5, 2015 8:39:33 GMT -6
Your plot looks alot like mine in the low end. I need to do a complete redesign, but have not had the time or capital to do so.
The front wall absorption is very important with non flush mount speakers.
|
|
|
|
Post by swurveman on Sept 6, 2015 15:09:51 GMT -6
Sorry if this is a tangent, but I thought I'd share my experience:
About a year ago I used REW to analyze my room, but only used the mix position for positioning my mic. Doing that, I got a pretty flat response. Last weak I did 16 different measurements across a 2' x 2' area-with the mix position in the center- using the ARC 2 software. My room response curve from the ARC 2 software was drastically different, with a big low end bump and a large midrange valley.
My conclusion-and I'd welcome anybody to dispute this if I'm drawing the wrong conclusion- is that you have to do a lot of measurements to get a sense of your room response. Either that, or there's a big disparity between ARC 2 and REW. I probably should go back and do the exact same 16 measurements with REW and see what it says.
|
|
|
|
Post by mulmany on Sept 6, 2015 20:42:42 GMT -6
Room EQ correction is a balancing act. The ARC software was originally designed for movie theaters, as a tool to speed up system optimization/EQ. It is an averaging algorithm that tries to EQ all points as close to the primary location as possible. Its a difficult task in the best of circumstances, and really should only be used when traditional means have been exhausted.
|
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 7, 2015 4:49:11 GMT -6
Morning. DSP is a tool with strengths and weaknesses. The first graph (above) from sonarworks takes 24 captures mapping your mix position so is a composite. This is where you mix from so you want the best sound in the space. I think the map is to accommodate a sweet spot dependent on your monitors. So here is a waterfall , new sweep from today but this is from REW not Sonarworks. Room difference is I moved my corner floor to ceiling panels to centre of front wall, so its about 7x7x4 inches and my monitor stands are about a foot in front of that panel but there are no corner traps in front wall now. So I wouldn't compare the two graphs. I bought some more roxal insulation to make some corner traps but haven't installed yet. Thoughts ? btw GIK has a good video tutorial about using REW which is shareware and they only ask for a contribution. If you have a flat sdc its very easy to install and run, if you have been curious about how your room plots ? Article from GIK about understanding waterfall charts: www.gikacoustics.com/understanding-decay-times/ |
|
|
|
Post by mulmany on Sept 7, 2015 10:43:41 GMT -6
You deffinately have a few problem spots. Have you run your room dimensions through bob golds room mode calculator? It would be interesting if the peeks and valleys line up with the mode calculations.
I would only be concerned with the dip at 70hz, build up at 125hz and 350hz, and the dip around 400hz. The hi frequency looks consistent with the stimulus.
You can also use the impulse response calculation to find trouble some reflection points. Look at the time gap from impulse to the first reflection this needs to be significantly lower in volume with a nice time delay.
You can order the calibrated usb mic from miniDSP. Makes a big difference in the quality of the data.
|
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 7, 2015 12:29:37 GMT -6
Thx no but I'll check the room nodes.
I always find my lower freq hard to discern so problems below 400 or so are consistent with my experience .
Solution is more mid band and bass traps or howdo I attempt to remedy the 4 specific frequencies?
|
|
|
|
Post by jazznoise on Sept 7, 2015 15:54:45 GMT -6
It can fix peaks, but room EQ can't do much about nulls..
A 6dB null isn't as bad as you might think, I've seen worse! Definitely a tricorner absorber like a superchunk is a good idea.
|
|
|
|
Post by mulmany on Sept 7, 2015 18:18:15 GMT -6
Thx no but I'll check the room nodes. I always find my lower freq hard to discern so problems below 400 or so are consistent with my experience . Solution is more mid band and bass traps or howdo I attempt to remedy the 4 specific frequencies? Well good thing about them is they all are going to be affected by the super chunk. The mode Calc will give you some insight on front, back, sides, or ceiling placement. Your room is holding onto that 125-150hz content at 20 db more then the surrounding freq. |
|
|
|
Post by jeromemason on Sept 7, 2015 18:56:21 GMT -6
What kind of decay (ringing) time is going on? The waterfall cuts it off at 300.... How big is the room??
|
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 7, 2015 20:54:20 GMT -6
11X20x7.5
|
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 7, 2015 20:58:03 GMT -6
Ps thx Guys I didn't get the super chunks built today: had a session with a cellist.
He plays really well ,Raphael, studied at Glen Gould Conservatory in Toronto.
I'll redo graphs when I get chunks in place!
|
|
|
|
Post by tonycamphd on Sept 7, 2015 21:49:10 GMT -6
my brothers room is almost exactly the same size, he too had a null at 70hz, along with a host of other issues, the easiest thing you can do is put super chunks in all your corners, the more you do, the smoother the bass response, if you have any box bottomed couches, chairs, closets or voids, stuff them with ultratouch cotton insulation, that should trap and smooth out the lows throughout the spectrum, you can do to much to the point of inefficacy, but you cannot cause harm by over trapping bass, better to have too much, than not enough IME, the rest is gravy and can easily be done by ear, run sine waves, or white noise if you can pick out freq's by ear, and walk around the room like swurve suggested, especially if your tracking in the same room, you should listen for flutter echos as well as stepping in and out of modes/ antimodes as you go. For more recommendations, it would help to know what materials are in the room, wood, plaster, carpet, drywall, glass? The length of the room can be your friend for diffusion, the ceiling height warrants a complete ceiling kill, but that usually goes better with at least some hardwood flooring and area rugs, software programs are helpful for lows, but after bass trapping, what i would do is trust your ears, and create a character you dig.
hope that helps a little.
|
|
|
|
Post by jeromemason on Sept 7, 2015 22:32:35 GMT -6
Not a large room, so your decay shouldn't be all that long... Of course my room is similar where I like a bit of live in the back and I treat all 1st and 2nd reflection points on the sides, front and rear. I use diffusion on one side of the wall and absorption on the other and it adds to having a large room but with no reflections at the mix position. Super chunks is the way to do it as they've said, and the cloud overhead, if you can fill it with the same material it'll help as well. My room sounds a lot bigger than it is, sounds like you're on the same path. |
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 8, 2015 7:29:53 GMT -6
All good info guys ; thx. I'll do chunks first: a think a step by step approach is prudent. I was interested as Raphael who has recorded in a number of studios warmed up a bit and said he liked the sound of my room I have a LEDE and a cloud over teh dead end, so I set him up facing into the end, a large piece of wood underneath his cello and mikes ( DIYC12, MK-U47 and an octvamod sdc (aimed higher on neck just to see) LDC about 3 feet in front and about 3 feet up aimed down a little above bridge toward the holes: the capture went very well. I'll post some files so you can hear the room/session, as I discuss tweaking the room. Tony had heard my c12 and my -u47 into my d4's before and recommended the c12 as a little cleaner but I tracked with the 3 mikes on separate channels and pres, as I only could only afford Raphael for 2 hours so I thought grab all the sound I can and just get good solid signals no distortion . After every pass I would dup the 3 tracks ( mute old, arm/mute new) as I don't like playlists so I ended up with 36 tracks quickly ! so, we listened as we went, as he wanted to build up some textured parts by overdubbing; very expressive player and, of course, the cello sounded lovely; very pleased, even though my room needs some help ! I have a 2nd D4 now so I will re track my acoustic part and likely the lead vox and finaly get a percussionist ( Matt Aston) over as I want a madrigal like a hand percussive part not drums.. when ever possible, I want to save up some dough and hire real musicians rather than midi, that way I am helping them with their careers, getting real playing which I prefer for my songs and increasing my recording skills too. And I will keep tweaking the room |
|
|
|
Post by gouge on Sept 9, 2015 2:20:25 GMT -6
I'm going out on a limb here but I think you have a null from the side wall. ie the 11' wall. I could very well be wrong though.
100hz is a half wave length of about 11'
to test is it's a node or bass build up you can open your door and repeat the measurement. that lets the bass out of the room.
it could also be a reflection from the front wall or the floor. 34" is the quarter wave length of 100hz.
I agree you can never heave enough bass absorption but this may be a null from the side wall so that's where I'd put the absorption. just hold up a bass trap and measure again.
not sure of the accuracy of your plot though. 3db variation is pretty much flat. 6db is pretty good. 10-20db not so. |
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 18, 2015 9:55:47 GMT -6
The 11 is the front and back wall.
I will get chunks for front corners done this weekend.
I'll redo measurement then and post but show from 750 or so down so we see more detail.
Thx again !
|
|
|
|
Post by swurveman on Sept 19, 2015 7:47:52 GMT -6
You can also use the impulse response calculation to find trouble some reflection points. Look at the time gap from impulse to the first reflection this needs to be significantly lower in volume with a nice time delay. Good stuff mulmany. I've never understood how you find troublesome reflection points based on an IR. Below is a sample of one. From the example, how would you discover where in the room the reflection point(s) are that are letting this "Discrete Reflection" be so prominent so that you could fix it? Thanks if you can provide some insight!  |
|
|
|
Post by mulmany on Sept 19, 2015 8:46:27 GMT -6
You can also use the impulse response calculation to find trouble some reflection points. Look at the time gap from impulse to the first reflection this needs to be significantly lower in volume with a nice time delay. Good stuff mulmany. I've never understood how you find troublesome reflection points based on an IR. Below is a sample of one. From the example, how would you discover where in the room the reflection point(s) are that are letting this "Discrete Reflection" be so prominent so that you could fix it? Thanks if you can provide some insight! Simple answer is a tape measure and calculation of the reflection path. This is done one speaker at a time. If you look at the delay time you can then find the distance traveled, sound travels @ 1130ft/s at 74 degrees. I usually take the direct path measurement and subtract it from the reflection path measurement then measure for objects that reside at that distance from the mic. The example is pretty nice and clear, in smaller rooms it gets harder as the initial delay gap is smaller. But this delay gap is very important to be larger then the space you record in, or you will never hear the acoustic coloration of the recording space. Hope that helps. |
|
kcatthedog
Temp
Super Helpful Dude
Posts: 16,082
Member is Online
|
Post by kcatthedog on Sept 19, 2015 9:04:00 GMT -6
I cut and placed the super chunks but haven't framed them yet.
Its almost two bags of roxal 2x4. I went from about 18 inches off the floor to ceiling. Cut in half ( 2 24 inch squares and then diagonally in half)
Its very apparent that the room absorption characteristics are immediately positively affected.
If I clap my hand at my monitoring position , the decay is much faster (50%?) and it sounds dryer and flatter as I guess the chunks are already doing their thing absorbing lower freq resonances I was used to hearing before.
|
|
|
|
Post by swurveman on Sept 19, 2015 9:17:45 GMT -6
Good stuff mulmany. I've never understood how you find troublesome reflection points based on an IR. Below is a sample of one. From the example, how would you discover where in the room the reflection point(s) are that are letting this "Discrete Reflection" be so prominent so that you could fix it? Thanks if you can provide some insight! Simple answer is a tape measure and calculation of the reflection path. This is done one speaker at a time. If you look at the delay time you can then find the distance traveled, sound travels @ 1130ft/s at 74 degrees. I usually take the direct path measurement and subtract it from the reflection path measurement then measure for objects that reside at that distance from the mic. The example is pretty nice and clear, in smaller rooms it gets harder as the initial delay gap is smaller. But this delay gap is very important to be larger then the space you record in, or you will never hear the acoustic coloration of the recording space. Hope that helps. Thanks for your reply Patrick. Sorry if I'm being obtuse, but by" delay gap" are you talking about the difference between the "Arrival of Direct Sound" and the "Early Decay" as shown on the IR posted? |
|
|
|
Post by mulmany on Sept 19, 2015 9:47:55 GMT -6
Simple answer is a tape measure and calculation of the reflection path. This is done one speaker at a time. If you look at the delay time you can then find the distance traveled, sound travels @ 1130ft/s at 74 degrees. I usually take the direct path measurement and subtract it from the reflection path measurement then measure for objects that reside at that distance from the mic. The example is pretty nice and clear, in smaller rooms it gets harder as the initial delay gap is smaller. But this delay gap is very important to be larger then the space you record in, or you will never hear the acoustic coloration of the recording space. Hope that helps. Thanks for your reply Patrick. Sorry if I'm being obtuse, but by" delay gap" are you talking about the difference between the "Arrival of Direct Sound" and the "Early Decay" as shown on the IR posted? no problem... its what they call the reverb build up area. The longer and lower that area is the better. you also want the early delay to be very steep or non existent. |
|
|
|
Post by Bob Olhsson on Sept 19, 2015 13:44:10 GMT -6
There's an elephant in the room. We started doing all this stuff in the 1970s and soon most people felt they were getting mixes that translated better from the NS-10s on their meter bridges! For a good laugh, just walk between two THX screens that are showing the same movie.
In a nutshell, room eq. can't work because we hear the sound of the speakers separately from the sound of the room. The only exception is the early reflections. They simply need to reflect a flat frequency response. I've never heard an attempt at attenuating them that wasn't much worse for mix translation than just doing nothing.
|
|
