View Full Version : what do you make of these graphs?


bulls
10-23-07, 10:39 AM
played with onkyo sr800's mic plugged in my computer (pretty good freq sensitivity compared to my cheap headset mic)

dark blue TOP LINE IS MEASURMENT
the bottom lighter blue is background noise/baseline/"mic silence" :P

recorded in the spot where i sit, i wanted to see any visible differences (i hear audio differences)...but i dont see too much visibly in these...maybe my technique/equipment is wrong (probably :D) (i played with individual frequencies, 20, 35, 45, 50 etc, and the sub/mic hit them spot on) (20 was really low/almost inaudible from more than 2-3 feet), 35 was good in my spot - 7 feet)

this is playback of a wave (48Khz samplerate), mono, generated sinewave sweep 35-200Hz, 10sec.

onkyo sr800 receiver volume at 50, crossover is at 80hz, sub level is at 0db, and i have it on a manual equalizer.

http://i22.tinypic.com/21diz4l.jpghttp://i22.tinypic.com/2qn9ypz.jpg
^heres a bar graph with the Equalizer ... OFF (left)...................... eq ON (right)(i have 40hz +3, 80-1, 160-2)

http://i21.tinypic.com/27xidq9.jpghttp://i23.tinypic.com/wrgkd4.jpg
the Equalizer is OFF .............................. and back .........................ON(i have 40hz +3, 80-1, 160-2)

http://i23.tinypic.com/e7zapx.jpghttp://i23.tinypic.com/11hxw2f.jpg
changed crossover to 60hz... and doublebass on (too boomy for me) ..... and crossover 60, doublebass off


layd out like this hard to tell differences , my ears can tell the eq 40hz +3 or 0, and doublebass, and crossover (when at 60 or lower it becomes too flat for jmusic)

does this look like anything to anyone?

Ethan Winer
10-23-07, 01:15 PM
i dont see too much visibly in these

When you consider that each vertical division is 10 dB, that means you have a 20 dB error between ~110 and 150 Hz.

does this look like anything to anyone?

It looks like a typical room that has no bass traps. :D

I'll also mention that a display of raw LF response tells only half the story. Just as important with room acoustics is the ringing (decay) times.

--Ethan

bulls
10-23-07, 01:26 PM
thanks, for response,

heh my couch is a bass trap ...no?

i dont know what that means a 20db error but i guess somethings wrong up the measuring wires!
i dont know the technicalities... decay, mmm... (does that mean how long until a sound fades (from the reflecting waves)) ?

does this look like this sub's best response is around 60Hz ? (or is this messed up by the room-which i cant change)

OvalNut
10-23-07, 02:34 PM
But Ethan's valid point is that you CAN change the room .... with bass traps. They can help tame that big peak at 125, and good ones well implemented can help significantly with that peak at 60.


Tim

SpectralD
10-23-07, 03:04 PM
The light blue lines suggest that you've got more than 20dB worth of 60-cycle hum (and its first harmonic at 120) in your no-signal measurements. Whether this is from sources internal to the computer and mic setup or from outside sources like power cords and lights I can't say.

OvalNut
10-23-07, 04:18 PM
Awesome observation SpectraID ! Touche', excellent.

Ethan Winer
10-24-07, 02:38 PM
heh my couch is a bass trap ...no?

Maybe a tiny bit. But you need real bass traps to tame that response to be acceptably flat.

i dont know what that means a 20db error

It means your response deviates from flat - what it should be - by 20 dB. In truth, I bet your real response is actually worse due to using too low a resolution when measuring. Now that SpectralD pointed out the hum, you can see that what should be displayed as a single narrow blip at 60 Hz actually looks more like a wide peak. And this means the measurement is fairly low resolution. I'm not sure how deeply you want to get into this stuff, but this article on my company's site explains how I measure rooms and talks about the tools I use:

http://www.realtraps.com/art_etf.htm

--Ethan

bulls
10-24-07, 07:24 PM
1. thx for the 60hz notice :) i went all crazy and unscrewed the mic out of its enclosure, but it had no other components, just a simple mic on a wire)... and when i'd hold the mic with my hand the 60hz would disappear - uh-oh... grounding issues i thought
and then ... duh .. 60hz... but it was a laptop i thought,
so i unplugged the power, just on battery
and the 60-120-180 spikes almost disappeared!

2.
what is this receiver doing ???

ponder this: ... i change the F/L/C speaker sizes to SMALL,
and the subwoofer frequencies and level output changed!!! that is not normal to me...

the sub now is louder at freq's below 40 and even can hear it at 20! (kind of lame at this point but anyway, before it didnt work here!)

anywho: heres more pictures (sorry didnt notice the software made a new db scale today)(also not using my nice sine sweep like yesterday)
eq on http://i21.tinypic.com/2upu5aq.jpg eq off http://i20.tinypic.com/no6atz.jpg

LTD02
10-24-07, 07:56 PM
so much for the bass traps theory

Jonomega
10-24-07, 08:08 PM
so much for the bass traps theory

What do you mean? An even FR is only half the battle. You still need appropriate RT60 decay times...

LTD02
10-24-07, 08:09 PM
What do you mean? An even FR is only half the battle. You still need appropriate RT60 decay times...

all i meant was that to vendors with a hammer, everything looks like a nail.

clubfoot
10-25-07, 07:19 AM
2.
what is this receiver doing ???

ponder this: ... i change the F/L/C speaker sizes to SMALL,
and the subwoofer frequencies and level output changed!!! that is not normal to me...

the sub now is louder at freq's below 40 and even can hear it at 20! (kind of lame at this point but anyway, before it didnt work here!)


This is quite normal depending on receiver. Set all spaekers to small, input their distances to listening position and match levels.