• 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!

EQ'd speakers are still too shouty

Joined
Sep 9, 2024
Messages
9
Likes
6
I was using MR4 for a couple months, after EQ, they sounded very good and natural
I then got these RP-600M II. Been using them for a month and I can't help but feel like they are too shouty and have too much ear gain at 3 kHz even after EQ. Vocals and electric guitars sound horrible
I feel like it's the horn tweeters' fault but it doesnt make sense because I EQ'd them to the same target as the MR4
1.jpg

IMG_1986.jpg
IMG_1987.jpg
 
Klipsch sound shouty? How unusual! (Sarcasm, that's a common reaction.)

Have you tried toeing the speakers out at various angles? I personally like horns but can't take them when they are pointed right at me. I'd try that first.

The 1200ish peak likely plays a role as does 2200, and you are a bit above your target above 2k, so maybe try rolling off the high end by a db or so after 2k.
 
That's a rather wiggly-looking measurement, despite apparently already being averaged. Try MMM and a lower window size.

I kind of have my doubts about RP-600M IIs in a nearfield scenario... I mean, they are crossed over low enough for the driver spacing to work out, but they're definitely not exactly optimized for it.
 
i have done dozens of measurements and EQs, ones way more compliant to the target than that, they still sound the same
 
If that's raw no EQ measurement to listening spot straight 0° cal do it at hi 70's dB, zip it up and attach it. What's the length of the room? For which part is the table responsible for (usually in the 70~100 Hz range).
Target is a bit of and tad too sharp (1~2 dB) to about 78 dB. Do a proper calibration one to 85~88 dB (linear in bass to 1 KHz and -6~7 to 12~20 KHz) and 105 Hz self filter for bass (Q 0.71) and mic (SPL meter in REW) to adjust it too couple preferable SPL. Or use some equal loudness compensation implentation that does it automatically (including highs boost on low SPL) as it's a game/work desktop station after all.
 
@duckhue One thing that I would try is to first apply @pierre's EQ preset based on Erin's anechoic measurements of the RP-600M II ("Data & EQ" tab), then measure the in-room response with it applied, and finally correct room modes <~500Hz only.

This is the advice the OP should be listening to.

In general, bad sound after DSP can be divided into 2 causes:

1. Improper measurement technique.
2. Wrong DSP strategy.

Re: measurements. Please go through the checklist in this post to make sure that you took a proper measurement in the first place. If you try to correct a garbage measurement, you will get a garbage result.

Re: DSP strategy. As @staticV3 said.

As for "shoutiness". It is caused by a broad dip in the upper bass / lower midrange region. Either improper EQ created the shoutiness, or your speakers have nonlinear behaviour that produces the shoutiness at different volumes (i.e. one amplitude response at volume A, different amplitude response at volume B). The way to test this is to do a dynamic compression test. This is how:

- set up your mic 1m/3ft from the speaker.
- using an SPL meter*, play pink noise and adjust the volume of REW until the SPL meter registers 76dB.
- take a sweep at 76dB.
- repeat sweeps at 86dB and 96dB by increasing the volume in REW in 10dB increments without using the SPL meter.

* If you don't have an SPL meter, then repeat the measurement from "normal listening volume" with 10dB increments until it is unbelievably loud.

In REW:
- overlay all the graphs and normalise them to the same volume (All SPL tab, right click, align SPL, average SPL)
- all the graphs will be referenced to 76dB. So Trace Arithmetic A-B, and subtract the 76dB curve from 86dB. Do the same for 96dB. Finally subtract 76dB from 76dB to produce a flat line.
- Rename the curves to something sensible like 76dB, 86dB, and 96dB. Go to "All SPL" again and select these graphs for display. You will see something like this:

1732025076889.png


(random measurement I grabbed from Erin's Audio corner). See that 0dB flat line? With a perfect speaker, you will see 3 flat lines perfectly overlaid on top of each other at 0dB. Most speakers don't do that, they will change the frequency response with volume. I suspect that this is where you will find the cause for the "shoutiness".
 
I drive bi-amplified hybrid ESL speakers using a DSP crossover which has both manual and auto-EQ capability. The auto-EQ does a good job of flattening the response but I still hear a harshness at [just under] 3kHz.

I read somewhere that human hearing is biologically programmed to be overly-sensitive to a baby crying at around 3kHz, so it kinda figures.

After I run the auto-EQ, I final tune my system by ear using various music tracks as reference. Typically, I will manually overlay a parametric EQ to pull down the ear-sensitive 3kHz band. My reference music tracks are "Holiday" by Erin Bodie, or anything by Allison Krause. These ladies have clear high voices that can cut diamonds if the tuning is hot.

After I take the harsh edge off Erin's and Allison's voices, nothing else I'm likely to play will sound harsh. I then follow up with Dave Brubeck's "Take Five", and if Morello's high-hat still has some sizzle, that confirms that I haven't taken too much off. That pretty much makes me happy.

Give it a whirl and see if it works for you.
 
I was using MR4 for a couple months, after EQ, they sounded very good and natural
I then got these RP-600M II. Been using them for a month and I can't help but feel like they are too shouty and have too much ear gain at 3 kHz even after EQ. Vocals and electric guitars sound horrible
I feel like it's the horn tweeters' fault but it doesnt make sense because I EQ'd them to the same target as the MR4
View attachment 407962
View attachment 407960View attachment 407961
Nearfield - you need to look at the direct curve. Check 2.5-4 kHz, preferably a bi lower power, secondly, check 7-8 kHz, should also a bit lower. 1-2 dB.
 
My reference music tracks are "Holiday" by Erin Bodie, or anything by Allison Kraus[ s ]. These ladies have clear high voices that can cut diamonds if the tuning is hot.

Apropos of (almost) nothing, I just wanted to second the Allison Krauss torture test for treble resonances (e.g., but not exclusively, for horn loaded treble drivers). The live version of Ghost in this House from Allison Krauss and Union Station Live (recorded in 2002, per the internet) is my go-to track for such shenanigans. My Cornwalls failed spectacularly. :(
 
Another test track...

 
Try with the grills on, who knows?
 
Horns can have issues that EQ cannot fix.

Diffraction is something EQ cannot fix. EQ can correct for the effects of diffraction on the measured frequency at the microphone location, but the diffraction itself is still happening. There is still a delayed-and-distorted artifact arriving after a very short time delay. It is primarily this short time delay which makes diffraction more audible and objectionable than its effect on the frequency response measurement would imply, and EQ cannot do anything about that time delay.

The ear's sensitivity to diffraction is level-dependent; that is, the perceived harshness caused by diffraction becomes more audible and objectionable as the SPL goes up. This is often incorrectly attributed to non-linear distortion; actually diffraction is a linear distortion of which the ear has a non-linear perception.

@klipschsoundsystem if this describes what you're hearing (it gets worse and worse as the SPL goes up, but goes away at very low SPLs), then my guess would be diffraction in the horn is at least part of the problem. If the edge at the mouth of the horn is fairly sharp, imo that could do it. Or if there is an abrupt discontinuity in the throat, at or near the junction between compression driver and horn, imo that could do it.

With horns, the devils are in the details.
 
Last edited:
I EQ'd them to the same target as the MR4
Others have provided good input. To add my two cents, I would ditch the Harman Curve. I used it for a while, but have found that, on modern recordings, I don't like it as well as just using a target curve with a linear slope. Currently, my preference is a -1.0dB/Octave slope starting at 100 Hz (i.e., down -1.0dB at 200 Hz, down -2.0dB at 400Hz, down -3.0dB at 800Hz, etc.). Try that. If it is too bright or too dull, then try adjusting the steepness of the slope, in small steps, until you find the slope that sounds best to you.

- set up your mic 1m/3ft from the speaker.
That is good when the speakers will be used for listening at larger distances, but the OP is using the speakers on the desk. Chances are that the listening position is closer than 1m. If that indeed is the case, I would place the microphone at the listening position.
 
Last edited:
The way to test this is to do a dynamic compression test. This is how:
Or, just look at Erin's data for those speakers:

Klipsch RP-600M II_Compression.png

The compression isn't bad, especially for use as desktop speakers.

I think the issue is the dip in the upper bass/lower midrange that is in the Harman Curve, which was used by the OP as the EQ target.

EDIT: Also note that the dip in the compression data in the upper bass further accentuates the dip in the Harman Curve in that range. Nonetheless, it doesn't really become signficant until past 96dB. And, that volume level is at 1m. The OP probably will be closer to the speakers when seated at the desk. Even considering dynamics, listening to speakers that loud at that close of a distance risks hearing damage.
 
Last edited:
Maybe far-fetched, but I'll throw it out anyway:
We have the classic Klipsch = shouty sound. Which you have tried to tame without experiencing any improvement. Then you might have; shouty=broken or half broken tweeters?

Do both speakers sound equally shouty?

Can you measure your tweeters with a multimeter? To see what kind of values it gives and if they differ much from each other.
 
For one thing, these speakers aren't designed for nearfield listening. Those are horns for cryin' out loud. Designed to THROW sound and project into the midfield at the very least. Horn loaded drivers are by their very nature 'shouty'. Think megaphone. Forward by design.
 
Back
Top Bottom