Ok back on track! Need to get this methodology wrapped
so limiting the EQ to 4 bands and +/-6db for all drivers, Yay or Nay ?
No more than 3dB within the bandpass seems reasonable to me.
that's pretty conservative for someone who wanted an all-out EQing !
That's pretty elusive for someone who own obsolete measurement gear !
is my obsolete gears are acceptable as level-matching tools or do i need to buy a brand new machine from Radio Shack ?
Eq flat to the intended bandwidth and compare at various volume levels. Seems pretty straight forward.
but what happen if ''favorites'' and ''realistics'' gives opposite answers ?
What's wrong with that? Many think that tube amp sounds more realistic. But what if one favors minimum distortion? (distortion is usually the trade-off or compensation for some variables that relate with "realistic sound").
If the response is equalized, whatever the result, I would prefer choosing my preference driver based on some "numbers".
Any "number" on non-linear distortion? (are they all below "threshold"?)
No rules, no limits, come on, the sky is the limits, yes?
+1 No limitations, use as much EQ is as needed to get the job done. Just be sure to let us know what You did once You have the results.
Maybe the $200 driver only needs 2 bands of EQ with 1.5 dB adjustments, and the $50 driver needs 7 bands varying up to 6 dB adjustments. The conclusion might be they sound very similar indeed. But someone could still opt for the $200 driver, because that would work better in their system, if they are not into the hardcore DSP thing. 🙂
What would be interesting, would be to have one listening session with Jon's rig and another with Tomahack's rig...
Eq flat to the intended bandwidth and compare at various volume levels. Seems pretty straight forward.
+1 No limitations, use as much EQ is as needed to get the job done. Just be sure to let us know what You did once You have the results.
Maybe the $200 driver only needs 2 bands of EQ with 1.5 dB adjustments, and the $50 driver needs 7 bands varying up to 6 dB adjustments. The conclusion might be they sound very similar indeed. But someone could still opt for the $200 driver, because that would work better in their system, if they are not into the hardcore DSP thing. 🙂
What would be interesting, would be to have one listening session with Jon's rig and another with Tomahack's rig...

+1 No limitations, use as much EQ is as needed to get the job done. Just be sure to let us know what You did once You have the results.
Maybe the $200 driver only needs 2 bands of EQ with 1.5 dB adjustments, and the $50 driver needs 7 bands varying up to 6 dB adjustments. The conclusion might be they sound very similar indeed. But someone could still opt for the $200 driver, because that would work better in their system, if they are not into the hardcore DSP thing. 🙂
What would be interesting, would be to have one listening session with Jon's rig and another with Tomahack's rig...![]()
I agree, just make sure all the drivers curve the same, flat or not.
I am very interested in the result. How much difference is between the drivers after dsp with the same curve?
Wow, what a day.
We go from ''no EQ please it's too advanced for us folks'' to ''your EQing is not enough''
The EQ you propose is bound to include room artefacts. An REW sweep smoothed 1/12 would too, if you don't use any type of gating.
So I'd call it 'over use' of EQ. Or: "Hammering the response into shape".
Correcting for driver and room as if it were one "feature".
So I'd call it 'over use' of EQ. Or: "Hammering the response into shape".
Correcting for driver and room as if it were one "feature".
Wesayso, i can see you're kind of worried about it, but you shouldnt be. Really.
Again: 400hz-7khz band + nearfield measurements (around 2m distance) + large room + moderate SPL... I'm very confident we won't have to ''hammer the response into shape'' because of the in-room situation.
Correcting for driver and room as if it were one "feature".
an anechoic chamber wouldnt be any more useful, on the contrary.
Our environement will show something closer to real-life, repeatable context. The distance from the center of the driver to the floor will be approximately the same as anyone would do.
And, more importantly, ALL the drivers will be compared in/with the very same environment, context, equipement, measures, etc...
I agree, just make sure all the drivers curve the same, flat or not.
then, might as well target a -2db around 350hz for all drivers. Doesnt seem much, but it's really more listenable than 400-430hz
Which means they will sound very similar 🙂.
Nobody mentioned the need for an anechoic chamber. But you could easily use a REW sweep and gate that with a frequency dependent window (of about 1/6 octave) to leave the room out of the equation as much as possible.
EQ it like that and all measurements taken without gating will look a lot cleaner. This type of EQ will sound better off axis too as it isn't as fixed to that sweet spot.
Nobody mentioned the need for an anechoic chamber. But you could easily use a REW sweep and gate that with a frequency dependent window (of about 1/6 octave) to leave the room out of the equation as much as possible.
EQ it like that and all measurements taken without gating will look a lot cleaner. This type of EQ will sound better off axis too as it isn't as fixed to that sweet spot.
then, might as well target a -2db around 350hz for all drivers. Doesnt seem much, but it's really more listenable than 400-430hz
Actually, mid frequency is more like 250Hz to 2khz.
Which means they will sound very similar 🙂.
Nobody mentioned the need for an anechoic chamber. But you could easily use a REW sweep and gate that with a frequency dependent window (of about 1/6 octave) to leave the room out of the equation as much as possible.
EQ it like that and all measurements taken without gating will look a lot cleaner. This type of EQ will sound better off axis too as it isn't as fixed to that sweet spot.
Agree.
Actually, mid frequency is more like 250Hz to 2khz.
250hz is not possible.
And crossed @ 2khz wouldnt be listenable for the vast majority of people.
... you know all that 🙄
250hz is not possible.
And crossed @ 2khz wouldnt be listenable for the vast majority of people.
... you know all that 🙄
I'm confuse. Are we testing midrange drivers or else?
oh, please..
...really. 3khz and up is not midrange territory. We are talking wideband or fullrange here.
- Home
- Loudspeakers
- Multi-Way
- World's best midrange Blind Testing - Need your help.