Help! Is my room okay? (REW results)

[Darkdays]

Hi! I've been doing music for a long time and usually my mixes translate pretty well to other devices, even in other people's homes. As I was thinking about getting studio monitors,
I thought about somehow verifying how good or badly treated my room is. From my subjective feel and experience, there's nothing sticking out in my room like rumbling or clashing frequencies. If I clap my hands, there's only a tiny short echo.

However, I just followed a quick and short youtube tutorial on using REW. That one wasn't very in depth, so I just did the following:

1. Used Rode NT1-A for measuring
2. Logitech consumer speakers for the output (they're around 150€ with good sound and a big subwoofer)
3. Tested the levels with measurement, it said "okay" in green font
4. Hit start and sweep. Sound came only from my left speaker. Microphone was slightly in front of me but on my ear's height.

Here are the results:

I was just interested in seeing if there are any big spikes, cancellations or something going on. The big initial spike is because I think that's when my sub kicked in and before there was just barely anything audible. And the only big other spike I see is around 100Hz. Everything else looks moderate to me.

What do you think? Is there any big problem or is it an okayish room?

PS: I don't know if I messed something up during measurement, because considering the tutorial I watched, it would be weird if all the frequencies in the waterfal diagram have such a long fadeout..

Thanks for your help!

 

[kemmler3D]

The room looks pretty average, but that's not to say it's good enough for mixing... You have pretty normal peaks and dips that are typical of an untreated room, you (and everybody) should definitely consider doing some room correction with EQ.

You also have barely any bass below 50hz so I hope you aren't mixing hip hop or dance music.

You didn't ask, but my advice on studio monitors is to figure out which ones you want, and "buy once, Cry once". I spent too long with mediocre monitors because they were more affordable, and should have just gone for Genelecs to begin with.

 

[Keith_W]

Please read the REW guide.

You can't use a Rode NT1-A for acoustic measurements.

In the future, when you obtain a valid measurement, please post the MDAT because we can see things that you may not know to look for. Right now there is no point looking at your measurements because it wasn't taken properly.

 

[Darkdays]

The room looks pretty average, but that's not to say it's good enough for mixing... You have pretty normal peaks and dips that are typical of an untreated room, you (and everybody) should definitely consider doing some room correction with EQ.

You also have barely any bass below 50hz so I hope you aren't mixing hip hop or dance music.

You didn't ask, but my advice on studio monitors is to figure out which ones you want, and "buy once, Cry once". I spent too long with mediocre monitors because they were more affordable, and should have just gone for Genelecs to begin with.

The barely-any-bass-below-50hz only happend with the test in REW, I don't know why that is. The bass of my system is usually really full and goes all the way down. When I mix my music and work with EQ in frequency ranges way below 50Hz, I can hear all the changes easily.

 

[Darkdays]

You can't use a Rode NT1-A for acoustic measurements.

Why not? I've read that for checking for major issues it's fine. Unfortunately I don't have the budget to get a specific measurement microphone right now.

 

[rdenney]

Why not? I've read that for checking for major issues it's fine. Unfortunately I don't have the budget to get a specific measurement microphone right now.

Because it isn't linear, and doesn't come with calibration data that REW can use to make it linear. It won't provide the information you would need to do any better than what you have now.

And it may explain why the measurements don't match your perception of the bass, though 50 Hz is a lot lower than most people imagine.

I would also want to see the anechoic measurements for your speakers. Usually, that becomes the basis for spectral corrections (using EQ) below about 200 Hz. Below that, you'll be dominated by the room modes and measurements are everything.

There's a difference between reverberation and echo. If you detect any gap in the sound before you hear a reflection, you have more serious echo than you imaging. Damping echo reflections with wall treatments always helps, in my view, even though a realistic amount of reverberation will keep you from being tempted to add too much processed reverb in the mix.

Rick "thinking the mythology of needing special studio monitors to make good mixes needs a thorough laundering" Denney

 

[kemmler3D]

Why not? I've read that for checking for major issues it's fine. Unfortunately I don't have the budget to get a specific measurement microphone right now.

You can, but the frequency response isn't flat:

So any measurements you take will be affected by that. There's also the fact that microphones' frequency response can (and do, more than you'd think) vary from unit to unit. Measurement microphones are calibrated at the factory and come with a calibration file to account for that.

Realistically the NT1-A is probably good enough to knock down big peaks in the bass and at least help things more than you hurt them, but for other uses you should be careful.

For example, there's +3dB around 13Khz in the NT1-A FR graph, but your own measurement shows about 3dB down. So what's the real story there? Is there a problem with the speaker, mic, room, or a combination? Is your treble really lacking or does the mic deviate from the stock response? Without a calibrated mic you just can't know for sure.

The barely-any-bass-below-50hz only happend with the test in REW, I don't know why that is.

In that case, either the sweep settings were incorrect, or there's a problem with your mic...

Definitely read the guide linked by @Keith_W , it will help you troubleshoot these things and get more confident in what the real problems are and what's causing them.

 

[Darkdays]

Because it isn't linear, and doesn't come with calibration data that REW can use to make it linear. It won't provide the information you would need to do any better than what you have now.

You can, but the frequency response isn't flat:
So any measurements you take will be affected by that.

I didn't plan on using it to correct my room, it was just really thought as a quick test to get a rough overview, e.g. if there are terrible cancellations or boosts going on that I might be unaware of due to being used to it.

For example, there's +3dB around 13Khz in the NT1-A FR graph, but your own measurement shows about 3dB down. So what's the real story there? Is there a problem with the speaker, mic, room, or a combination? Is your treble really lacking or does the mic deviate from the stock response? Without a calibrated mic you just can't know for sure.

Yes, but what was more important for me now is not the actual numbers in itself but the relationships. The coloration doesn't matter so much for me now. I've been listening to over I don't know how many ten-thousand songs in this room, living here a bit over 10 years, so when I mix, I already have in mind what good music has to sound like in this room. Knowing the sound of a place and having good references is worth much.

But if I would've spot some major cancellations or other, very powerful artifacts, then I would have to question how much sense it would make to buy studio monitors if the room would've needed some serious treatment first.

The main graph looks solid to me.. the other one with the waterfall, according to the tutorial I watched, frequencies shouldn't look like this beyond 150Hz.. I've looked into the guide and specifically for the waterfall too, but I don't understand from it what it should look like. What does it tell for my room?

 

[kemmler3D]

I already have in mind what good music has to sound like in this room. Knowing the sound of a place and having good references is worth much.

Sure, no doubt about that.

if there are terrible cancellations or boosts going on that I might be unaware of due to being used to it.

Well... for example... 130hz looks to be >20dB above 100hz. That's a big enough disparity for the masking effect to become important and I think that could easily throw off your mixes, and this is not something you can get used to. Content at 100hz will at times become inaudible if something is playing at 130hz at the same time.

You probably aren't EQing mixes that sharply, but it just illustrates why getting used to your room and monitors is not a complete solution. If it were, there would be no need for room treatments or real monitors, mixing engineers could just spend a lot of time with random speakers in a random room and get perfect mixes.

The long and short of it is, your room is a typical room based on that graph, but typical rooms aren't considered good for mixing below 200-300hz for this reason. This is why you will sometimes hear people saying you should monitor bass with headphones.

What does it tell for my room?

Looks like you have some noticeable modes that are ringing all the way up to 350hz. The RT60 also looks a bit long for a small room. IMO not out of the ordinary but definitely indicative of things that could be improved with EQ, or better, room treatment.

 

[Keith_W]

@kemmler3D don't attempt to read RT60, waterfall, or anything else. Those graphs don't mean anything, and attempting to read them will mislead him. He is using an uncalibrated CARDIOID microphone meaning it won't capture sound from all directions equally, let alone be linear across the frequency range. This is the typical polar pattern of a cardioid mic:

It will underestimate the RT60 (and also spectrogram and waterfall), especially if he is pointing the mic at the speaker - i.e. it will reject any reflections coming from behind the mic. And if he is not pointing the mic at the speaker (pointing upwards) the frequency response will be inaccurate - as you can see, cardioid mics only have a flat response at 0 degrees.

Yeah, we can see some room modes and you could even say that the mode at 110Hz is about 20dB down. But the overall measurement does not say anything about the tonality of the system.

I am sorry, but those measurements are uninterpretable. If you can't afford an omni condenser mic right now and you desperately want to know what is going on, borrow one.

 

[kemmler3D]

CARDIOID microphone meaning it won't capture sound from all directions equally,

Oh, right, duh!

@Darkdays - what he said. Omni mic is pretty much a requirement for this, frequency response aside.

 

[Darkdays]

Thank you guys! I've seen there's a cheap beringer measuring mic for only 30 bucks, I could get that one.
Another question in the meanwhile.. I realized that my sub actually only goes down to 45/50Hz. I still hear something at 40Hz, but that's really just subtle (tried a sinewave generator at different Hz).

Now my question: I'm as well mixing on studio headphones and they go down all the way to 10Hz. Unless my room is treated really well, especially with bass traps (which are out of my budget as of now, I've been told they have to be stacked higher and not just sit at the bottom of the corners), wouldn't it be superior to mix bass in headphones? Or to be more specific, deal with the <50Hz frequency especially in headphones. Of course everything above sounds very powerful on my audio system and reflects how most consumers listen to music perhaps.

I know I know, it's not quite the same with the air that you feel in a physical room and such. But if I can't ensure that my room is not adding artifacts / nulling / boosting to the bass, there is no point in having a sub that goes even lower, no?

Also I heard people say you need at least two subs to make up for nulling..

 

[kemmler3D]

wouldn't it be superior to mix bass in headphones?

I don't know how common it is, but people definitely do this to avoid room issues.

But if I can't ensure that my room is not adding artifacts / nulling / boosting to the bass, there is no point in having a sub that goes even lower, no?

You can use room correction EQ to deal with it to a significant extent... And Room modes usually exist up to 200-300 hz, so the fact that they are present is not necessarily a deal breaker when it comes to subs.

 

[Darkdays]

You can use room correction EQ to deal with it to a significant extent... And Room modes usually exist up to 200-300 hz, so the fact that they are present is not necessarily a deal breaker when it comes to subs.

I think I've read earlier that correction EQ does work with mids and highs but not for the very low end. And if I have standing bass waves and stuff in my room, then the subwoofer isn't giving my any transparent and correct information to work with. That's why I was thinking that a subwoofer might be burned money in a not very well treated room.

How do you mean it that it's not a deal breaker?

 

[kemmler3D]

I think I've read earlier that correction EQ does work with mids and highs but not for the very low end. And if I have standing bass waves and stuff in my room, then the subwoofer isn't giving my any transparent and correct information to work with. That's why I was thinking that a subwoofer might be burned money in a not very well treated room.

Depending on the nature of the correction, arguably it's the opposite. The peaks due to standing waves can be knocked down using parametric EQ and if you have the headroom on your sub, you can even try to bring up the nulls a bit. (Usually only lowering peaks is recommended because most subs don't have 10-20dB of output to spare, and true nulls can eat 100% of a boost anyway...)

Broad correction for tonality works on mids and highs, but correcting standing waves at high frequency only works for smaller and smaller areas in the room, because the nulls appear in areas that correspond to the wavelength. Once you are in the Khz range, EQ can only correct a spot a few inches across or even just a fraction of an inch... Useless.

It works better in the bass region in part because the wavelengths are very long, therefore the area that gets corrected is big.

How do you mean it that it's not a deal breaker?

Well, you have peaks and nulls all through the range from 20hz to 200-300 hz in most rooms. So IMO the fact that you have these nulls is not a reason to avoid trying to listen to 20-40hz any more than it's a reason to avoid listening to 100-200hz. my opinion is you should try to reproduce everything and correct what you can in any case.

 

[Darkdays]

Well, you have peaks and nulls all through the range from 20hz to 200-300 hz in most rooms. So IMO the fact that you have these nulls is not a reason to avoid trying to listen to 20-40hz any more than it's a reason to avoid listening to 100-200hz. my opinion is you should try to reproduce everything and correct what you can in any case.

Well I wasn't planning on avoiding to listen to and treat everything below <50Hz, I was just thinking about moving mixing decisions for that range to good studio headphones where I can be sure that I hear 100% of what's real, rather than maybe fighting some room modulations. Sure I wouldn't feel the air pressure the same way or how reflections interact, but that's different for anyone who will listen to my music anways. So I came to the conclusion that the most important thing to have control over is perhaps just the very crisp details, attack transients, rumbling and such in the low end. I mean actual rumbling from the source, not because the room emphasizes frequencies.

I guess there's some disadvantage in using these room correction tools, otherwise people would just throw that in an save room treatement. All the research I've done in the past hours stated that a sub in an untreated room does more harm than good..

.. but I would be of course happy if it's really just that easy, that I (properly) measure the room and use room correction to make a sub worth using in my place.

So, what's the truth? I'm feeling confused by all the different informations.

 

[NotInYourHead]

I (properly) measure the room and use room correction to make a sub worth using in my place.

This should be your priority. Usually, you make the measurements, upload them here so the good people can take a closer look and go from there.
A good subwoofer, correctly placed and tuned, can be very helpful in dealing with room issues.

 

[kemmler3D]

Well I wasn't planning on avoiding to listen to and treat everything below <50Hz, I was just thinking about moving mixing decisions for that range to good studio headphones where I can be sure that I hear 100% of what's real, rather than maybe fighting some room modulations. Sure I wouldn't feel the air pressure the same way or how reflections interact, but that's different for anyone who will listen to my music anways. So I came to the conclusion that the most important thing to have control over is perhaps just the very crisp details, attack transients, rumbling and such in the low end. I mean actual rumbling from the source, not because the room emphasizes frequencies.

I guess there's some disadvantage in using these room correction tools, otherwise people would just throw that in an save room treatement. All the research I've done in the past hours stated that a sub in an untreated room does more harm than good..

.. but I would be of course happy if it's really just that easy, that I (properly) measure the room and use room correction to make a sub worth using in my place.

So, what's the truth? I'm feeling confused by all the different informations.

Headphones can be good for monitoring bass, but AFAIK it's harder to get it to translate than it would be if you just had a properly treated room with subs.

A sub randomly slapped in an untreated room can definitely make things worse. A sub properly placed and integrated with EQ can make things a lot better.

PEQ applied based on a good measurement is a great tool, and if you have a proper mic and follow a tutorial it's not that complicated.

What EQ can fix well is peaks, what it can't fix that well / at all are dips (nulls), and it can't deal with decay times except to the extent that lowering the amplitude decreases the decay time. So this is why multiple subs and room treatment are still used, EQ can't fix everything.

That said, even just fixing peaks is a big deal, I definitely recommend it if you're willing to give it a shot.

 

[janbth]

…
I guess there's some disadvantage in using these room correction tools, otherwise people would just throw that in an save room treatement. All the research I've done in the past hours stated that a sub in an untreated room does more harm than good..

.. but I would be of course happy if it's really just that easy, that I (properly) measure the room and use room correction to make a sub worth using in my place.

So, what's the truth? I'm feeling confused by all the different informations.

So what research have you done? Subs in untreated rooms are routinely used to fix problems with room modes. I myself am using two subs in a multisub setup (plus some DSP) to combat some nasty room modes in my listening room. It works.

 

[Darkdays]

So what research have you done? Subs in untreated rooms are routinely used to fix problems with room modes. I myself am using two subs in a multisub setup (plus some DSP) to combat some nasty room modes in my listening room. It works.

Except the sub introduces these strong modes in the first place.. unfortunately I don't have to budget to get two right away in order to deal with this, this is why I am thinking about how to get the best out of my situation.

I don't think I even have problematic modes going on (judging by ear / "experience"), as my room is mid-sized with a kitchenette at the other end, maybe that helps with carrying off low end energy.. and perhaps a sub would be totally fine in my room and minor issues could be fixed by moving it around. But of course I don't know what I don't know, until I can objectively measure the place, which requires probably uncolored speakers and a measurement microphone first.

I've seen many people saying they get much clearer and better mix if they mix without a sub, and that many professionals would mix that way too. And they just occasionally check back on subs. I've never tried yet to mix entirely without a sub. Getting two bigger studio monitors that get down to somewhat 45Hz + basic room treatment (2x early reflections 2x basstrap to stack in at least the close corner to the sub) would be an option. However, then there's the problem of the monitors not sounding as clear as they could because they need a lot of energy to produce the low end.

It's being said that with covering some early reflection points and some bass traps a major chunk of the room acoustics can be improved. I can only treat one corner with bass traps. But the other two (yes two, the last one is in said kitchenette) are at least 4 and 6 meters away and by just doing the test of going there and listening in these spots with the head down there, I don't hear any standing waves building up at least.

Another option would be my original plan, to go with 1 pair of studios + 1 sub and just work with the room I have. Like I said, it's perhaps an okayish or even good room, nothing terrible.. but I figured out yesterday that I can improve my room by simply hanging a big blanket to my right side over the door (which has a middle glass window), it reduced the reverb coming from the side and gave me a cleaner listening experience. I am actually a bit baffled by how minor yet big the impact feels in the same time.

Well, and my last option is to just roll with an amplifier + headphones and work mainly in headphones, and just check back occasionally on my consumer system or so.

Maybe I'm stressing too much, given that I never had problems with my mixes translating. They sound the way I envision and finish them here, on any device or place. But I think I would benefit from a bit more clearity so that I can make faster mixing decisions, whether that comes from just buying studio monitors that are more neutral or doing the room treatment combo or using better headphones.


Decisions, decisions..