Beginner trying to figure stuff out.

Okay so, I just took a bunch of screenshots of some of my graphs. I have no idea if this is what you were talking about.

This shot is of my left and right channel together.... Just the simulated results for the frequency response.

This next one is zoomed up simulated impulse response on both speakers. Not sure if it's better if I zoomed up even more?

This one is "step response simulation" of one of the channels. And making sure the check at the bottom is on "simulation front right" and "step response Eva nr front right" (whatever that means)

Again, not sure if you want me to zoom up more on one of those....?

This is a zoom up on the simulation step response I think.... Of the left speaker.

That's all I got for now. Please tell me if I'm even taking screenshots of the correct things.

Thanks for your help

Awesome, thanks for the input. Just a few more pictures for reference. One screenshot of my filtered frequency response for both speakers, just showing how the bass on my speakers drops, and where are my target curve is.

Closer shot on the impulse response for both speakers.

And just one shot showing the correction procedure designer. I don't have anything boosted, and I only use 1dB for Max correction boost. Most of these decisions were the results of just listening.... And preferring this sound.

juicehifi said:

Looking

Click to expand...

So in the correction procedure designer..... It's set to 10hz.

My microphones frequency response is 20hz - 20khz.

Should that be set to 20hz then in the correction procedure designer?

Thanks

So...I was experimenting a bit with the target curve, and I ended up trying to follow more closely the tweeters response at the very end of its extension. Basically from 12kHz and up.

Wow, what a difference in sound. I know I've read this suggested by others and even in the help files.... But I never actually tried it. It sounds so much smoother this way.

My question is, does it matter how many reference points you use on your target curve. Clearly as you can see, I used quite a bit for the tweeter..... Is it better to use less reference points? Or does it not matter?

Like, I could probably delete 50% of those and still maintain basically the same shape.

So, I did try partial correction but.... I didn't like it as much. I tried a bunch of different points, but it didn't sound as clear. Imaging was a little bit more diffused. Oh well... I tried.

Just out of curiosity, is there a way to change the channel balance of the outputted convolution file?

Like say, make one channel a couple DB louder(or quieter) then the other. I know that the playback programs generally have this ability.... I'm just curious.

Thanks

My measurements look good I suppose, but.... I'm just a guy who doesn't know as much as yourself.

Not sure what you mean by consistent delays?

I had another question. I've been using these convolution files with HQplayer.... And in HQplayer you can plot your files.

The picture of included is the plot of one of my channels.

The designer of HQplayer says about this plot "Here you have some phase modification too, not angle just proceeding spinning at continuous rate, but instead you have a discontinuity in the spinning at least at one point where you have phase adjustments.
It is possible to negate this delay related phase spinning, but it also kind of makes the response "fake", since it is "modified reality".

Could you please give me your take on this? Thanks.

Yep.... I made sure my speakers were time aligned the best I could.

That HQplayer guy suggested I go no more then 18dB per octave. Not sure with Audiolense, if it's really necessary to be that conservative.

Oh sorry....the amount of bass drop-off. He says your target falls too rapidly and should not exceed 18dB per octave.

So I made two targets, and just compared them by listening. One following his advice, and one where the target follows the natural bass fall off of the measurement, but continues downward till around -70dB.

I much preferred the steeper bass drop off.(The pic with the red arrow). Plus, I have a two-way speaker, the vocal sound much cleaner the more bass I remove.

LOL...ya, I already made an adjustment to that to make it smoother, just didn't bother updating the picture.

Should this be ok? I am zoomed out quite a bit.

Awesome, very helpful....thanks.

So...when it comes to high frequency extremes, the help file says the target should be "straight" at the extremes. I've also read that following your tweeters natural FR is recommended.

But my highs fall off fairly rapidly passed 21kHz.

Should my target trend down following the tweeters drop at 21kHz then quickly straighten just before 24kHz?

Or should it just straighten out at 20khz and stay flat till the end?

Thanks again for your help.

I've included 2 examples of the targets.