The quest for perfection

[Buckle-meister]

Just the opposite. It is a small hole to the long waves, insignificant, hence no change. On the other hand, the shorter waves as you go up in frequency, it becomes a huge hole, out they go and no reflections.

Thanks Mtrycrafts. Your comment had me totally baffled because you say that the opposite of what I wrote is in fact the case . After re-reading your words though, I think we are actually in agreement! . Can you confirm?:

The way I picture things in my mind's eye, low frequency waves would travel right through the door's mass as if it didn't exist in the first place, so that whether the door is or is not there, there is no difference (as may be seen from the applicable plots).

However, since higher frequency waves won't travel through the door's mass, it does make a difference whether the door is there or not because in one case (the door is there), the waves will be reflected, and in the other (the door is not there), they continue on out into the hall.

Isn't this true?

Also, what is your opinion of the response without the doors (the way the room normally is)? Is it terrible would you say? not bad? ok? good?

Regards

 

[sploo]

Hi BM,

Right...

I believe that in your impulse response graph the 1.41ms is 1.41ms after the speaker to listening position impulse. I.e. you have a reflection path that's 1.41ms * 343m/s = 0.00141s * 343m/s = 0.484m = 48cm longer than your direct speaker to ear path. This may well be the first reflection point on the side walls.

Absorption at first reflection points will greatly help your impulse response, and will result in less comb filtering.

Your 3D plot looks pretty clean, but with sharp peaks. One of the things that bass trapping will do is to smooth those peaks (thereby making changes in sound level less severe with changes in the frequency).

In general, you seem to have a better starting point than me (my room is thin, so having the speakers near the side walls made for some nasty reflections) but you'll be able to greatly improve it with room treatment.

Don't be surprised to get big effects with small changes. After my first round of treatment, I added a nasty null in the bass range. Ethan pointed out that it was probably always there, but masked by a peak that I'd removed with the treatment. Overall, you will get an improvement, but accept that you may see some odd results now and then.

 

[Buckle-meister]

I believe that in your impulse response graph the 1.41ms is 1.41ms after the speaker to listening position impulse. I.e. you have a reflection path that's 1.41ms * 343m/s = 0.00141s * 343m/s = 0.484m = 48cm longer than your direct speaker to ear path. This may well be the first reflection point on the side walls.

Oops! I ball$ed up there huh?!

In general, you seem to have a better starting point than me...

This certainly ties in with the fact that I always said that my own room didn't sound too bad to my ears, whereas you always said your own did. One thing I just can't fathom though; if my square room is the 2nd worst shape that there is, why aren't my results worse than your own?

Cheers Sploo.

Anybody else?

Regards

 

[sploo]

I believe that the square room thing affects the modal response - i.e. low frequency peaks/nulls.

Because I have a long thin room, and ended up with speakers near the (hard, shiny) side walls you get strong reflections from the walls, ceiling, and (hard wooden) floor.

If your speakers are 3' from your side walls, and you've got a carpeted floor, and maybe even a ceiling higher than normal, you'll get far less problems with the midrange.

Absorbing these reflections made the single biggest difference for me. The bass traps reduced the 'boom' quite a bit, but the reduction in reflections turned the room from foul to pretty good.

 

[mtrycrafts]

Thanks Mtrycrafts. Your comment had me totally baffled because you say that the opposite of what I wrote is in fact the case . After re-reading your words though, I think we are actually in agreement! . Can you confirm?:

The way I picture things in my mind's eye, low frequency waves would travel right through the door's mass as if it didn't exist in the first place, so that whether the door is or is not there, there is no difference (as may be seen from the applicable plots).

However, since higher frequency waves won't travel through the door's mass, it does make a difference whether the door is there or not because in one case (the door is there), the waves will be reflected, and in the other (the door is not there), they continue on out into the hall.

Isn't this true?

What I was saying is that the wavelengths get shorter as you increase frequency and short wavelengths will go out that door opening and not reflect from that portion of the wall. So, depending on where that door opening is in relationship to the measuring point, it will affect the high frequency response more than the low frequency that is longer than the door opening and will not as readily exit and gets measured as though it was a solid wall.

Also, what is your opinion of the response without the doors (the way the room normally is)? Is it terrible would you say? not bad? ok? good?

Regards

I am not sure as the response curves I have seen do some sort of averaging protocol so it looks smoother. What you posted apperas not to be averaged or smoothed and very ragged. Not very good looking. I don't think your plot is realistic, but no sure?

 

[Buckle-meister]

I don't think your plot is realistic, but no sure?

How do you mean 'not realistic'?

Regards

 

[mtrycrafts]

How do you mean 'not realistic'?

Regards

I haven't seen such spikes, so many in other speaker modeling. Perhaps the averaging used in the plots?

 

[Buckle-meister]

This quote was taken from the thread this thread continued from (refer to post #1).

Air has a higher co-efficient for higher frequencies. Higher frequencies will be absorbed/damped by air at a greater rate per distance as compared to lower frequencies.

Doesn't the above statement imply that the further one sits from the speakers, the worse music will sound due to a loss of upper frequencies? It can't all be relative, i.e. even though the upper frequencies are decaying, so are all the others (though they are), because the upper frequencies decay "at a greater rate" than those lower.

What's the solution? Turn up the volume?! I ask because when I look at my full frequency response in post #14, I see what appears to my eyes to be upper frequencies dropping-off quite early, yet with my room being a relatively modest 4.2x4.2m, I don't sit far from the speakers. Why then do my upper frequencies drop off so quickly?

Regards

 

[mtrycrafts]

This quote was taken from the thread this thread continued from (refer to post #1).



Doesn't the above statement imply that the further one sits from the speakers, the worse music will sound due to a loss of upper frequencies? It can't all be relative, i.e. even though the upper frequencies are decaying, so are all the others (though they are), because the upper frequencies decay "at a greater rate" than those lower.

What's the solution? Turn up the volume?! I ask because when I look at my full frequency response in post #14, I see what appears to my eyes to be upper frequencies dropping-off quite early, yet with my room being a relatively modest 4.2x4.2m, I don't sit far from the speakers. Why then do my upper frequencies drop off so quickly?

Regards

EQ is a solution to correct room interaction? Door opening reduces higher frequency reflections from that sector where the door opening is vs closed. So, I wonder how much of the high frequency is absorbed if it travels further, yet arrives still to affect.

I wonder how your speaker measurement would be at 1 meter, with respect to the high fr fall of, vs the one at your listening position?
Consult with an acoustician much more informed than me?

 

[WmAx]

What's the solution? Turn up the volume?! I ask because when I look at my full frequency response in post #14, I see what appears to my eyes to be upper frequencies dropping-off quite early, yet with my room being a relatively modest 4.2x4.2m, I don't sit far from the speakers. Why then do my upper frequencies drop off so quickly?

Regards

The solution is to use an exactly inverse equalization curve[though in reality, I doubt you would want to compensate for this by 100%, as this would make many recordings probably sound to dominant in high frequencies]. Equalization also a solution to bass problems[traps can only solve some of the problems, and equalizer is needed in combination with traps for optimal response].

-Chris

 

[sploo]

I wonder how your speaker measurement would be at 1 meter, with respect to the high fr fall of, vs the one at your listening position?
Consult with an acoustician much more informed than me?

I did try this when I was messing around with ETF.

Putting the SPL meter close to the speaker, and setting the gate time to reject any paths longer than speaker->SPL (i.e. reflections), results in a much smoother and flatter response - presumably the speaker response (within the accuracy limits of the SPL meter).

 

[Buckle-meister]

The solution is to use an exactly inverse equalization curve[though in reality, I doubt you would want to compensate for this by 100%, as this would make many recordings probably sound to dominant in high frequencies]. Equalization also a solution to bass problems[traps can only solve some of the problems, and equalizer is needed in combination with traps for optimal response].

Thanks WmAx, but (my fault, sorry!) what I should have written in place of "What's the solution" is "What's the explanation". In that context:

1. Was my first paragraph correct?

2. Again, if the distance is so short, why would upper frequencies fall-off so quickly? Or isn't it so surprising, going by the plot you submitted in the previous thread?

Regards

 

[Buckle-meister]

Putting the SPL meter close to the speaker, and setting the gate time to reject any paths longer than speaker->SPL (i.e. reflections), results in a much smoother and flatter response - presumably the speaker response (within the accuracy limits of the SPL meter).

I haven't tried this yet, but I see the logic of your words.

Regards

 

[sploo]

I haven't tried this yet, but I see the logic of your words.

That's the nicest thing anyone's ever said to me