• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

REW, A Graph and What It All Means

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,374
Likes
234,443
Location
Seattle Area
I do wonder at what SPL are target curves from some companies derived at, like is Harman’s target for 80dBC or 90dBC or...?
He generally follows this recommended level based on ITU-R BS1116:

1570819269356.png
 

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,244
Likes
11,470
Location
Land O’ Lakes, FL
What does that have to do with room correction measurements done with a mic? This was in reference to what SPL to use for calculating correction filters.
Meaning, your EQ is dependent on the SPL of the measurement. It’s better to EQ to your target at ~90dB than it is at say 50dB, as our hearing changes with level, so it’s better to use the level you normally listen at to EQ, and ideally have a loudness compensation.
 

audimus

Senior Member
Joined
Jul 4, 2019
Messages
458
Likes
462
I think there is mixing of two related but different concepts here: Measuring and calculating room correction filters to smooth the response curves and EQing for personal tastes.

You can measure for room correction at any level as long as it is sufficiently above the noise floor to make it accurate. The filter corrections being relative (+/- dB) will try to maintain the same target curve shape at any gain setting that determines how loud you listen to. You can offset the measurement curves to any SPL level in REW after measuring but the filter calculations will not change.

Dynamic EQ and Loudness compensation, etc., that have to do with ear’s sensitivity differences across the spectrum with loudness are orthogonal to this. You can have room correction without them or you can have those without room correction.

The Harmon curve is not a correction for loudness. It is based on empirical evidence that people, in general, prefer to have low frequencies at boosted levels from flat and a sloping down of higher frequencies. At ANY level.

Another way to look at it is think of using different target curves at different listening levels (with varying levels of “room gain”) to compensate for varying sensitivity to low frequencies. The deltas between these are relatively small and varies a lot between individuals to come up with a single absolute value curve to use. People also listen at very different levels as normal for them.

The “Harmon curve” that people like is more for the general shape than the absolute values at some SPL. The relative low frequency gain can be typically anywhere from 6db to 10db depending on taste and content. The high frequency slope can vary depending on how bright or warm one prefers their music. The exact curve can vary between individuals.

But it will also have to tempered with what your speakers are capable of. If the speakers have a sharp roll off at higher frequencies then it is not a good idea to have a low slope and try to boost via correction filters. If your woofer-tweeter balance (or sub-small speaker balance) without EQ does not show a Harmon bump in the low frequencies, then you don’t want to use EQ correction filters to boost that gain. With a sub, I experiment with its gain levels 6dB-10dB over the mains before doing measurements and applying eq so that, it naturally provides a low frequency bump and the correction filters just smooth it out.

So you can measure with a mic at any suitable level and then you experiment with the target curve (mostly “room gain”) to suit your personal tastes or hearing abilities at your typical listening level unrelated to reference or measurement levels. A lot of people do seem to boost the SW levels after say a Audyssey XT room correction because most don’t listen at measurement levels and they don’t necessarily like what dynamic eq does with its fixed formula.

If you have a dynamic eq or a loudness feature, it will automatically introduce small deltas in tonal balance as you go up or down from this typical listening as a way of approximating those different target curves but there is no guarantee that this fixed delta calculation will suit every taste or ear or content. It is most effective when the volume is turned down very low where the low frequencies are barely audible for typical human years (like the loudness toggle in old audio equipment). The continuously variable change with tonal balance in response to volume is a very crude approximation and makes sense in theory but not so universally appreciated in practice (unlike room correction). It does provide a lot of marketing talking points.
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
753
Likes
625
Location
Herts., England
I tried REW for the first time following a combination of guides. Ultimately, I took a measurement, had REW EQ it and applied what it said to do in JRiver through the parametric EQ. I did this twice. The first time, I took separate measurements for each speaker, averaged out what was found and then applied the REW eq. The second time, I had both channels measures simultaneously. I was surprised that with both attempts, there were few areas of change. (6 frequencies were identified by the EQ when averaged and 10 when taken simultaneously.) I preferred the sound of the eq using the simultaneous measurements rather than the averaged, for what that's worth. Further, I didn't, ultimately, reduce two frequencies as much as recommended after adjusting the numbers more to my liking. All this leaves me with a few questions (after my before graph which is the simultaneous measurement - I ran it twice with a minimal deviation near the end of the range):

View attachment 35662
1. Is there a preferred way to measure: individually, simultaneous, multiple points in a room (I did them only from my listening position).?

2. With eq, there was less pronounced bass (I reduced the reduction here by 1/2) and a less bright high end (I reduced the reduction here as well, but by 1/4). Is that to be expected or are there different ways to apply the filter (I used what the guide suggested) that change the sound more or less rather than simply aiming for flat?

3. I expected a much larger range of frequencies to adjust. Not because I could hear that it would be necessary, but by virtue of reading what others had done vis-a-vis REW. Does the above graph indicate that there wasn't a whole lot of eq-ing to be done?

4. Where would I go from here assuming I feel further room correction is necessary/desired/warranted?

1. I measure initially for the lower frequencies, say 20-500Hz. I measure each speaker separately, placing the microphone where my left ear would be for the left speaker, pointing at the speaker. I use REWs average of four measurements using the 256Hz signal. For the higher frequencies or full range I measure both speakers together, placing the microphone at the centre of my head position pointing between the speakers.

I suspect others will say that you should measure the bass from both speakers and the higher frequencies with speakers separate!

2. As you found you get less pronounced but more even (smoother) bass. I EQ to flat in this region and am happy. Others use the Harman slope or even a steeper one. There is no right or wrong, just your taste. The Harman slope seems popular.

What I did was EQ up to 200Hz (as I ran out of PEQ filters on my equaliser) and have just listened like that. I should EQ above 200Hz using the equaliser like a tone control, say at 1/6 octave, but I haven't felt the need to do that yet.

When studying the bass performance I use no smoothing on the frequency response graph.

3. I don't know what smoothing you have used on that graph but it doesn't look bad overall. I don't know what speakers you have but let's assume they don't go much below 30Hz then there would be no need to be concerned about the peak at 23Hz. The peak at 47Hz should be brought down though. You've also got dips at 90Hz and 330Hz which you might try and EQ but may not be possible to correct. If you look at a 'no smoothing' graph you might find those dips to be very narrow. If they are they probably will not be noticed.

One way of checking the bass smoothness is to listen to test tones. You can download a free test tone CD here:

https://realtraps.com/test-cd.htm

4. There are a number of ways to get a better sound:
a) Positioning of speakers and listening chair.
b) Room treatment - bass traps etc..
c) DSP/EQ - to complete the job after a and b, or on it's own if you can't do a and b.
b) multiple subwoofers.

Finally, don't get obsessed with the perfect frequency response graph. The proof is always in the listening.
 
Top Bottom