AVS Forum banner
Status
Not open for further replies.
1 - 20 of 20 Posts

·
Registered
Joined
·
900 Posts
Discussion Starter · #1 ·
I'm dialing in my system. Need a good RTA.


Have a very old GoldLine that's 1/12th low and 1/3 above 144hz.


Looking for something that's got a LOT more resolution above 144hz ... like 1/12th or 1/24 up through 10k or 16k.


(I could use my Dell laptop as the platform, but it's only got USB and firewire: no slots for soundcards, etc. It was a "standard" tiny travellers XP based laptop.)


What would you recommend?
 

·
Registered
Joined
·
17,407 Posts
.... and don't assume that your brand new microphone is accurate. Mine was more off than I would have imagined and calibration only cost in the $40 range.
 

·
Registered
Joined
·
900 Posts
Discussion Starter · #5 ·
(1) How would ease of use and customer support differ, if any, between the TrueRTA and ETF products?


(2) Any appreciable difference in "speed"/"ease" of measurement? Thing I like about the old GoldLine is that it's easy to just play wideband pink noise and observe the changes in the RTA as I play with the EQ


Thanks.
 

·
Banned
Joined
·
2,352 Posts
Quote:
Originally Posted by Brucemck2
(1) How would ease of use and customer support differ, if any, between the TrueRTA and ETF products?
Don't know about the customer support, but the ease of use goes to TrueRTA. For both of them you can get more help from the forums. Actually I recently wrote a small quide regarding the TrueRTA.

http://www.avsforum.com/avs-vb/showthread.php?t=572477
Quote:
(2) Any appreciable difference in "speed"/"ease" of measurement? Thing I like about the old GoldLine is that it's easy to just play wideband pink noise and observe the changes in the RTA as I play with the EQ


Thanks.
ETF can't do pink noise at all. TrueRTA can perform just as you discribed.


ETF is more precise, but it can't work in real time.
 

·
Registered
Joined
·
17,407 Posts
Ilkka,


Have you had any experience with the Allen and Heath real time analyser software? If so, what did you think about it?
 

·
Banned
Joined
·
2,352 Posts
Quote:
Originally Posted by JimP
Ilkka,


Have you had any experience with the Allen and Heath real time analyser software? If so, what did you think about it?
No I haven't, but I just installed it and it doens't look very good. It is has only 1/3 octave resolution, cheesy GUI, no proper calibration for mic and system anomalies.


1/3 octave resolution is the worst thing, even 1/6 isn't enough. You need atleast 1/12 or preferably 1/24 or more.


I would take TrueRTA or ETF over this one hands down.
 

·
Registered
Joined
·
17,407 Posts
Ilkka,


Thanks for the response.


Regarding the 1/3rd octave resolution, if you're using something like a Rane equalizer in 1/3rd octave correction, does having 1/24 octive resolution RTA benefit you or is that something more appropriate for a BFD where you can have smaller increments of correction.
 

·
Banned
Joined
·
2,352 Posts
Quote:
Originally Posted by JimP
Ilkka,


Thanks for the response.


Regarding the 1/3rd octave resolution, if you're using something like a Rane equalizer in 1/3rd octave correction, does having 1/24 octive resolution RTA benefit you or is that something more appropriate for a BFD where you can have smaller increments of correction.
You are on the right track. You must have a parametric EQ to really make the most of those high resolution results.


For 1/3 octave EQ, I'd suggest 1/6 octave RTA.
 

·
Registered
Joined
·
3,258 Posts
Hi Ethan,


You're right about ETF showing time-domain measurements, which an RTA cannot do. But ETF "MLS" is not intrinsically better than pink noise. MLS refers to two things:


1. A Maximum Length Sequence (pseudo-random) of square waveform lengths, which is "white" acoustically.


2. The traditional MLS analysis method, which is a very fast frequency transform (Hadamard transform) based on the signal's simple two-valuedness.


I believe that while ETF uses 1, it no longer used 2, favoring the "dual FFT" method instead. That latter allows a reference signal to be subtracted from the test signal. You've been in more recent contact with Doug than me.


Pink noise weights the low frequencies more than the highs, compared to white noise. So for the same FFT length, it could actually be more accurate for subwoofer frequencies.


Hey, don't forget who turned you on to ETF in the first place! :)


Regards,

Terry
 

·
Registered
Joined
·
900 Posts
Discussion Starter · #15 ·
Terry and Ethan,


Perhaps you could weigh in the "suggested path" for tuning an existing room?


The basics have already been attended to. My room has pretty good dimensions (checked via several mode calculators; approx 18'L x 15'w x 12'h). Have good bass traps across four wall junction boundaries. (MegaLenrds across front wall/ceiling juncture, up the front/side wall junction on both sides, and across the rear/ceiling juncture.) Have also placed 2" of 703 on the front wall.


Speakers and chairs are all placed where the sound is "good". Two SVS Ultra subs (run as mono) are placed in front in symetrical locations. Used the Goldline to confirm the locations of seating, chairs, and speakers. Side benefit is that three of the four chairs have highly similar frequency response patterns (fourth abuts a side wall, and I expected it to have much poorer sound.)


My next step was going to be to EQ the sub(s) and the three front channels.


I used the Meridian's room EQ to measure the fronts and the subs, and it gave a pattern similar to the Goldline in 1/12th octave setting. (The Meridian would automatically produce only 1 filter unless I manually intervened. Even then, it only gets up to about a dozen filters across all three fronts and the sub, out of 45 available, so it at least believes the room measures well.)


Currently the sub is plus 8 minus 10 db, ignoring two narrow but deep nulls. Currently the three fronts are closer to plus/minus five db, again ignoring two narrow but deep nulls. Using 8 parametric filters on the subs I got to plus/minus 2db for three of my four chairs, and plus 3 minus 8 for the fourth chair., from 25hz to 100hz. That's with no more than an 8db cut in any filter, and no more than a 4db boost, with only two boost filters in the overall mix.


Was intending to do the same for each of the three fronts, but needed an RTA with resolution finer than my GoldLine's 1/3rd above 144hz (hence my earlier post) to get the proper filters.


Figured once I'd done that, then I'd add some room treatment at first reflection points and on the rear wall behind my head (likely some combination of RPG BAD panels and/or RPG Skylines's.)


Is there another sequence that would make more sense? Does it not make sense to EQ the three fronts too? How would "modal ringing" etc. that Ethan mentioned be taken into the mix?


Thanks!
 

·
Registered
Joined
·
3,258 Posts
Quote:
Originally Posted by Brucemck2
...

I used the Meridian's room EQ to measure the fronts and the subs, and it gave a pattern similar to the Goldline in 1/12th octave setting. (The Meridian would automatically produce only 1 filter unless I manually intervened. Even then, it only gets up to about a dozen filters across all three fronts and the sub, out of 45 available, so it at least believes the room measures well.)


Currently the sub is plus 8 minus 10 db, ignoring two narrow but deep nulls. Currently the three fronts are closer to plus/minus five db, again ignoring two narrow but deep nulls. Using 8 parametric filters on the subs I got to plus/minus 2db for three of my four chairs, and plus 3 minus 8 for the fourth chair., from 25hz to 100hz. That's with no more than an 8db cut in any filter, and no more than a 4db boost, with only two boost filters in the overall mix.
I'd say +/- 2 dB was just great for the combination of your two subs over 3 chairs. One can't optimally EQ subs for a large area.

Quote:
Was intending to do the same for each of the three fronts, but needed an RTA with resolution finer than my GoldLine's 1/3rd above 144hz (hence my earlier post) to get the proper filters.
For higher frequencies, you don't need the same kind of RTA frequency resolution as for subwoofer frequencies. EQ correction at mid to high frequencies should be broader, to correct the overall shape of the frequency response as necessary. Narrow surgical correction does not work well at these frequencies.

Quote:
Figured once I'd done that, then I'd add some room treatment at first reflection points and on the rear wall behind my head (likely some combination of RPG BAD panels and/or RPG Skylines's.)


Is there another sequence that would make more sense? Does it not make sense to EQ the three fronts too? How would "modal ringing" etc. that Ethan mentioned be taken into the mix?


Thanks!
I would install those treatments before EQ. They will influence your frequency response -- hopefully, for the better!


As for modal ringing, that comes out in the wash with EQ. Fixing a low frequency room mode also fixes its time-domain behavior.


Regards,

Terry
 

·
Registered
Joined
·
3,854 Posts
Bruce,


Judging from your thorough post, it looks to me like you have a good handle on things:
Quote:
Using 8 parametric filters on the subs I got to plus/minus 2db for three of my four chairs, and plus 3 minus 8 for the fourth chair., from 25hz to 100hz. That's with no more than an 8db cut in any filter, and no more than a 4db boost, with only two boost filters in the overall mix.
It’s hard to ask for better than that.

Quote:
Does it not make sense to EQ the three fronts too?
It does in some cases, but it might not be worth the expense. You’ll either have to have pre-out and main-jacks for all three front channels or use outboard amplifiers. Otherwise you won’t be able to connect the equalizers. Needless to say, of you have to spring for both the amps and the EQ’s, it can get pricey.


Regards,

Wayne A. Pflughaupt
 
1 - 20 of 20 Posts
Status
Not open for further replies.
Top