Hi, I've been playing around with the DEQX for about a year now and
I've tried many speakers and combinations of measurements/corrections.
While I'm not completely thrilled with the DAC capabilities of the Hd Express II
I think that the room correction capabilities cannot be ignored.
So I wanted to share something I tried that I think sounds really good and also ask a question about it.
The usual method to correct the drivers (in this case Kef LS50's) with open baffle 15" woofers below)
is to measure each driver from up close and then correct and only then do eq corrections on the room.
Well I decided to try something different this time and I was pleasantly surprised.
I wired BOTH speakers (highs/lows separately) to the same amps so that both speakers
were playing in mono. Then I did my measurements from the listening position and then cut and pasted
the one side side to the the other so that both speakers are using the same FIR filter. I then separated
the drivers so they each get their own amp. And then I took the average response on both speakers playing simultaneously
and applied EQ.
And well it does sound really good. Im wondering if tehre's any explanation for this? Only thing I could think of is that using the same filter sort of 'averages' out the response in the room better.
Anyways there it is, hopefully someone more knowledgable can chime in on whether this is a good idea or a bad one.
I've tried many speakers and combinations of measurements/corrections.
While I'm not completely thrilled with the DAC capabilities of the Hd Express II
I think that the room correction capabilities cannot be ignored.
So I wanted to share something I tried that I think sounds really good and also ask a question about it.
The usual method to correct the drivers (in this case Kef LS50's) with open baffle 15" woofers below)
is to measure each driver from up close and then correct and only then do eq corrections on the room.
Well I decided to try something different this time and I was pleasantly surprised.
I wired BOTH speakers (highs/lows separately) to the same amps so that both speakers
were playing in mono. Then I did my measurements from the listening position and then cut and pasted
the one side side to the the other so that both speakers are using the same FIR filter. I then separated
the drivers so they each get their own amp. And then I took the average response on both speakers playing simultaneously
and applied EQ.
And well it does sound really good. Im wondering if tehre's any explanation for this? Only thing I could think of is that using the same filter sort of 'averages' out the response in the room better.
Anyways there it is, hopefully someone more knowledgable can chime in on whether this is a good idea or a bad one.
I think you misunderstood. I am listening in stereo, I only did my room measurements for FIR filters in mono using two speakers playing sweeps instead of one.
What I do - if it can help in any way. Is to measure both in mono combined - but also mono seperate.
What I mean by this. Is that around 300-500hz - your room goes from being dominated by the speakers to being dominated by the room.
So I measure my front speakers - each by themselves - mono. This tells me what my speakers are doing before the room interfere.
Then I measure my woofers and subwoofers - all together in mono - same signal. Since lower frequencies are longer wavelengths - you can se them more or less, as a combined sound source - this is why multisub works. You should note that doing so with woofers and subwooferes. Can ad an extra level of sound that are not present when playing in stereo - which should be copensated for, before playing music.
Point is, to measure how the woofers and subwooferes work as a unity .... cause this is essentially what they do. Also individually to see if they have a similarity - cause this is mostly your room-mode.
mid-woofers, midrange and tweeters, are measured gated or as free from any reflections possible. This to ensure that you only EQ what is possible to EQ. Remember - EQ works only with linear distortion - deviations in amplitude and phase. As soon as the sound bounces around between any surface - you can only correct with EQ - in ONE place - cause then you will also change something else - somewhere else. You are only changing sound from the speaker with EQ - not the room itself. This is why it's important to measure like I try to describe
Please ask further..... cause my explaining can surely miss something
Exactly as I wrote
Summed in the listening position, works for lower frequencies..... not for higher. Low - under ca. 150hz. High above 4-500hz in most rooms. Between 150 - 500hz is the transition area, where it's kinda like a grey area because theoretically it's not possiple to correct it with EQ and practicallly you can't measure it in normal rooms - short story
I tried the DEQX... did not work for me. The problem - IMO - is that it tries to hard to get the phase right... witch also gives you a nice step response. But I have to say, that I dont care about nice curves if it does not sound good.
I have read through tons of material and tried almost all of the hardware out there. If you dont understand the theory behind how sound waves and our hearing works. Then we all can get easily caught up in technology and nice specs, but will never get really good sound.
We have to turn things around a bit - IMO. We dont need a DSP to force a beautiful curve in one tiny space of a listening room. We need to understand how we enjoy music in general. And then go backwards in different directions towards the source and then try to fix the problems at their core.
If you force a speaker to play whatever needed, to get a good step response at one location in a given room. Then you might have the speaker play something that makes everything worse - like the original signal/source material. And/or create new artifacts and any kind of distortion that might not have been there to begin with.
That's why you have to analyze what is going on with different types of measurements, so that you can better understand what can be solved where and how.
Most systems I've ever seen or heard about. Mostly measure a room response and by this they try to "clean" things up. But by this, the software does not know how the curve got distorted to begin with. Was it a bad driver, bad room, bad positioning, bad speaker in general? It mostly just EQ the crap out the speaker to get a nice response - but "cares" less about what is possible in the real world. This is in my opinion why we all get so many different results out of EQ - passive or active.
It is vital, to really know how and what you measure - before you correct.
When you manually measure close, far, off-axis, loud low, stereo, mono and so on.... you look for patterns and things to compare. By this you slowly figure out where and what the problems hides. In a simple room response, you get a huge mess and mix of everything - where you have no real clue what is going on.
There's a lot of evidence that humans can mostly listen "through" most rooms. Meaning that we can mostly figure out who's voice or what instrument is singing/playing - by tune, pitch and volume. This indicates that the source is really important for us, when it comes to realism of sound. And it also gives us an idea that when measuring if this source is good or needs correction (our speakers) - then we need to measure with minimum interference from the room - when searching for the errors/the distortion - that we want to correct.
All of this changes with frequency - mainly split up between low and high - reflected and direct sound. Meaning lower frequencies have longer wavelengths and therefore have more reflections and then again allow for room response measurements, because we mainly listen to reflections at these frequencies - in normal smaller listening rooms - less than 80-100 sqm.
At higher frequencies we first hear a direct sound - but then have all kinds of reflections playing fun and games with us - cause rooms are so very different in both size, shape and reflective construct. This is why we need to make different kinds of measurements to figure out where things are possible to be corrected - and how.
Hopefully this clears up why I'm not a fan of a lot of auto-measuring products like ex. what DEQX are offering.
But none the less..... I'm happy that you got something good out of it. My thoughts should never deprive you from enjoying what you do
I have a DEQX as well but as far as I know the DEQX does not do any room correction at all. It was never meant to function as room correction, it is meant to correct the freq/phase/pulse response of the loudspeaker itself, not the room. I have tried it in room for fun but the results where allways terrible, audible and measurable. The advised "Room correction" on the DEQX website isn't any room correction at all, they refer to the parametric EQ that also has a basic automated function, that's it.
Even with loudspeaker corrections you have to be aware of that you are not correction artefacts like baffle difraction or first reflections, the results will be terrible. The DEQX is not an easy tool.
Even with loudspeaker corrections you have to be aware of that you are not correction artefacts like baffle difraction or first reflections, the results will be terrible. The DEQX is not an easy tool.
Exactly. You can only correct sound from the speaker driver, cause you correct the signal at a line level - before the amplifier drives the speakers. This means that you only change the amplitude, frequency and phase of the signal going to the speaker. The speaker will still have it's sound signatur in the power of baffle design, dirver size and shape. This is something that can never be changed with EQ.
So if you take any speaker of any design - and change anything with EQ on one axis - then it will change too on any other axis, at the same time. Because EQ can only change sound in one place at one time. Any dispersion or ressonanse from the speaker, that is not of a linear type - will always be controlled by the driver type and size - aswell as the enclosure and baffle it is placed in.
I am pleased to see that DEQX actually sum it up rather nicely. First the speakers and then the room - dont mix!
Quote: "Calibrate the speakers first..."
Home - DEQX
And then their video to explain - which I kinda like for it's simplicity. I would still question the audibilty of group delay and use of EQ to correct at bad designed speaker.
The DEQX Factor - Why Timing Matters - YouTube
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.