My continuing EQ adventure - EFT users should read (1 Viewer)

[Rob Formica]

Ummm... but if i recall correctly, you can get a "NON-logorithmic" graph of the frequency response between 20 and 200hz using the low frequency add on in a resolution greater than 1Hz. It'll be more than enough to set up your EQ. It is covered in the demo room as well as on TAG (?) room eq web site. Perhaps someone could post you the link for the latter on as I don't have it handy...

No doubt that the learning curve is steep... perhaps even more than a hardware RTA. You must first understand the science of audio measurement, but you can look at it as an educational tool.
Rob

 

[Rob Formica]

Ummm... but if i recall correctly, you can get a "NON-logorithmic" graph of the frequency response between 20 and 200hz using the low frequency add on in a resolution greater than 1Hz. It'll be more than enough to set up your EQ. It is covered in the demo room as well as on TAG (?) room eq web site. Perhaps someone could post you the link for the latter on as I don't have it handy...

No doubt that the learning curve is steep... perhaps even more than a hardware RTA. You must first understand the science of audio measurement, but you can look at it as an educational tool.
Rob

 

[Ethan Winer]

Kincade,

> I can't seem to get decent resolution on the "low response" graph <

I use ETF's LF response display all the time and the results are repeatable and seem accurate. I agree that Doug at Acoustisoft should extend the LF display from 200 Hz up to at least 300 Hz, and 400 Hz would be better still. I've already suggested this to him.

So why not just use the LF display? I'm not a big fan of EQ generally, but it can help at the lowest modal frequencies. In a test I did a while ago I found that below about 160 Hz the modal resonant frequencies remain constant around the room. So the LF display is definitely usable there.

--Ethan

 

[Ethan Winer]

Kincade,

> I can't seem to get decent resolution on the "low response" graph <

I use ETF's LF response display all the time and the results are repeatable and seem accurate. I agree that Doug at Acoustisoft should extend the LF display from 200 Hz up to at least 300 Hz, and 400 Hz would be better still. I've already suggested this to him.

So why not just use the LF display? I'm not a big fan of EQ generally, but it can help at the lowest modal frequencies. In a test I did a while ago I found that below about 160 Hz the modal resonant frequencies remain constant around the room. So the LF display is definitely usable there.

--Ethan

 

[Kincade]

Ethan,

Well - I guess I thought that the LF response wouldn't work because doug told me it wouldn't. He said I need the add-on for any resolution under 50-60 hz.

Additionally, the LF display is radically different than the 1/6 octave plots I have done before, with repeatability; this suggests to me that one of the two is wrong. With doug telling me that it won't work as it stands, I guess I just assumed that the 1/6 plots are more accurate.

You DO think that the LF display is accurate then, without the PSD/Sweep component?

 

[Kincade]

Ethan,

Well - I guess I thought that the LF response wouldn't work because doug told me it wouldn't. He said I need the add-on for any resolution under 50-60 hz.

Additionally, the LF display is radically different than the 1/6 octave plots I have done before, with repeatability; this suggests to me that one of the two is wrong. With doug telling me that it won't work as it stands, I guess I just assumed that the 1/6 plots are more accurate.

You DO think that the LF display is accurate then, without the PSD/Sweep component?

 

[Rob Formica]

TAG seem to have been bought out by International Audio Group (IAG) part of Quad, Wharfedale and Wharfedale Pro. The room demo pages featuring ETF5 no longer seem to be on their new website.

http://www.tagmclarenaudio.com

I got the link from these forums. There used to be a downloadable PDF file of the tutorial... maybe someone here kept a copy. I only have a printed version. Their tutorial was more complexe than the ETF demo room.

good luck...
Rob

 

[Rob Formica]

TAG seem to have been bought out by International Audio Group (IAG) part of Quad, Wharfedale and Wharfedale Pro. The room demo pages featuring ETF5 no longer seem to be on their new website.

http://www.tagmclarenaudio.com

I got the link from these forums. There used to be a downloadable PDF file of the tutorial... maybe someone here kept a copy. I only have a printed version. Their tutorial was more complexe than the ETF demo room.

good luck...
Rob

 

[Ethan Winer]

Kincade,

> I guess I thought that the LF response wouldn't work because doug told me it wouldn't. He said I need the add-on for any resolution under 50-60 hz. <

I don't have that add-on so I can't comment. But I think Rob may be correct, that you misunderstood Doug Plumb.

> the LF display is radically different than the 1/6 octave plots I have done before <

Understand that a 1/6 octave display averages all of the data within that 1/6 octave band. So getting different results doesn't surprise me. Unless I'm missing something. How did you measure 1/6 octave originally? With sine waves or pink noise?

The screen cap below shows the exact same data expressed at 1/3 octave and again at 1/12 octave.

--Ethan

 

[Ethan Winer]

Kincade,

> I guess I thought that the LF response wouldn't work because doug told me it wouldn't. He said I need the add-on for any resolution under 50-60 hz. <

I don't have that add-on so I can't comment. But I think Rob may be correct, that you misunderstood Doug Plumb.

> the LF display is radically different than the 1/6 octave plots I have done before <

Understand that a 1/6 octave display averages all of the data within that 1/6 octave band. So getting different results doesn't surprise me. Unless I'm missing something. How did you measure 1/6 octave originally? With sine waves or pink noise?

The screen cap below shows the exact same data expressed at 1/3 octave and again at 1/12 octave.

--Ethan

 

[Kincade]

Ethan,

Thanks for the tip. That may be what i'm seeing. It looked to me like there were peaks and dips in different places when comparing the logarithmic and low freqency graphs. I'll run a few other sims though and see what I can do.

So... If I am going to try to EQ using ETF, you just recommend using the low frequency graph? I really wish that I could see a 1/6 or 1/12 octave logarithmic scale of the low frequencies in this program. It seems much easier to correlate the filters in the BFD to that, rather than a linear scale.

 

[Kincade]

Ethan,

Thanks for the tip. That may be what i'm seeing. It looked to me like there were peaks and dips in different places when comparing the logarithmic and low freqency graphs. I'll run a few other sims though and see what I can do.

So... If I am going to try to EQ using ETF, you just recommend using the low frequency graph? I really wish that I could see a 1/6 or 1/12 octave logarithmic scale of the low frequencies in this program. It seems much easier to correlate the filters in the BFD to that, rather than a linear scale.

 

[Ethan Winer]

Kincade,

> I swear he told me that the program is not accurate below 50 hz without the PSD sweep add on. I'll email him again and see what he says. <

Please do and let us know. When I bought ETF some months ago I explained to Doug Plumb what I need to do. He said I'll get a resolution to better than 1 Hz, which is all I care about. But if I missed something I'd sure like to know!

Better still, invite him to chime in here in person.

> you just recommend using the low frequency graph? ... It seems much easier to correlate the filters in the BFD to that, rather than a linear scale. <

Yes, just use the LF Response display. But no matter whether the scale is log or linear, numbers is numbers. Since you can pinpoint the exact frequency in ETF with the cursor, it shouldn't matter if the screen is linear.

--Ethan

 

[Ethan Winer]

Kincade,

> I swear he told me that the program is not accurate below 50 hz without the PSD sweep add on. I'll email him again and see what he says. <

Please do and let us know. When I bought ETF some months ago I explained to Doug Plumb what I need to do. He said I'll get a resolution to better than 1 Hz, which is all I care about. But if I missed something I'd sure like to know!

Better still, invite him to chime in here in person.

> you just recommend using the low frequency graph? ... It seems much easier to correlate the filters in the BFD to that, rather than a linear scale. <

Yes, just use the LF Response display. But no matter whether the scale is log or linear, numbers is numbers. Since you can pinpoint the exact frequency in ETF with the cursor, it shouldn't matter if the screen is linear.

--Ethan

 

[Felix_F]

I bought a BFD today. Brought it home and took about an hour to get accustomed to it operation, then measured my system with 10-120Hz sine waves at 1Hz increments. It took me a few hours, using 5 filters, to get my sub to +/- 2dB from 30-100Hz measuring after every change. I also set a house curve and an "excitement" curve, for the lack of a better term, while I was at it. At the risk of sounding insensitive, the manual method sure seems easier and less frustrating than using software to approximate the interaction at hand.

Felix

 

[Felix_F]

I bought a BFD today. Brought it home and took about an hour to get accustomed to it operation, then measured my system with 10-120Hz sine waves at 1Hz increments. It took me a few hours, using 5 filters, to get my sub to +/- 2dB from 30-100Hz measuring after every change. I also set a house curve and an "excitement" curve, for the lack of a better term, while I was at it. At the risk of sounding insensitive, the manual method sure seems easier and less frustrating than using software to approximate the interaction at hand.

Felix

 

[RichardHOS]

Actually, I found using ETF to set up the BFD to be a snap. Even allowing for the time it took to become accustomed to the BFD's user interface, it took less than an hour to dial in my EQ settings. That included at least a dozen iterations through ETF before I was satisfied with the filters. With a new room response generated in seconds, I just can't imagine going through the same process manually taking readings, writing them down, typing them into a spreadsheet, and then calculating filters. It would take hours... if not days!

 

[RichardHOS]

Actually, I found using ETF to set up the BFD to be a snap. Even allowing for the time it took to become accustomed to the BFD's user interface, it took less than an hour to dial in my EQ settings. That included at least a dozen iterations through ETF before I was satisfied with the filters. With a new room response generated in seconds, I just can't imagine going through the same process manually taking readings, writing them down, typing them into a spreadsheet, and then calculating filters. It would take hours... if not days!

 

[Felix_F]

Richard, that does sound very easy. Does EFT work off 1/6 octave delineations?

Felix

 

[Felix_F]

Richard, that does sound very easy. Does EFT work off 1/6 octave delineations?

Felix