In article , MD
wrote:

MINe 109 wrote:

In article ,
MD wrote:


I own a Behringer Unit that allows me make parametric DSP changes to
each of my room modes (all were measured by plotting individual tones -
in 1hz increments - not warble tones etc)

I love what the unit does and it's negatives are very small (a small
amount of noise added and the input has a hard time with high line level
inputs. I fixed this by changing the final gain stage of my DAC)

I have heard the Rives in a demo and liked it. Was never able to compare
it to the Behringer

I have never heard the Tact T or any other digital correction system

Here's my question - Why would I spend more than the $150 I spent on the
Behringer? I can digitally set freq (within 1hz) set bandwidth (within
a few hz) and set gain - all with a DSP that runs at 24 bit - 46khz
64/128 oversampling.

If the answer is that the other units do this at a higher frequency -
would i be able to hear the difference (red book CDs) and I I could
would I pay thousands more? Now that Behringer has a new unit out that
has 96khz sampling and is only $400 - wouldn't I buy that?

And please - save me the answers where you assume that cheaper -
pro-audio gear is crap - unless you have heard it.


For the extra cost, the TacT uses test tones and room measurements to
generate correction curves more accurately than the method you employed.
If you're satisfied with your results, bravo!

snip

Thank you. Actually the newer/more expensive ($400) Behringer has
tones/mic. Why would my hand plotting be bad - in 1hz increments? I
use the radioshack meter. Unless the mic in the TacT is more linear?

It's the test tone itself. TacT uses 'click' tones to measure transient
response, arguably more important to reproduction than steady tones. At
least, Ralph Glasgal argues this:

http://www.ambiophonics.org/Tact.htm

This is rather old (1999) but there are lots of reviews out there.

Transients aside, the TacT measurements will be more accurate than the
Radioshack can provide. Of course, the TacT costs twenty times more that
your Behringer.

Thank you, Mr. Google. Here's a better explanation:

http://www.regonaudio.com/Digital%20...o%20Part%20I.h
tml

RCS Measurement and Correction Model

The best way to understand what the RCS unit does and why it works so
well is to imagine first, for contrast, an idealized version of the old
"slider" band-by-band analog EQ devices and figure out why they did not
work right. By "idealized" I mean I am going to suppose that the device
simply does what it is supposed to do operationally, with no distortion.
The old idea was this: Run a broadband, steady-state test signal through
each channel (separately) of your system. Measure the steady-state
response at the listening position in frequency bands corresponding to
the sliders' frequencies. (The highest resolution, to my knowledge, had
30 bands, each one-third octave wide.) Move the various "sliders" up or
down as needed to get the measured steady-state response essentially
flat. Do this for each channel. Now your system is "flat" and the
channels match.

In reality, this process often produced worse results than what you
started with... There isn't much wrong with using steady-state response
as the measure in the bass. The problem is that you didn't have narrow
enough bands.

The second problem is a little harder to understand because it involves
a surprising property of how we hear. In the bass, we really have no way
to tell the difference between the "first arrival" and later, reflected
sound. You cannot really get a handle, even mathematically, on the
energy at, say, 100 Hz, in some sound until that sound has been going on
long enough to produce a cycle or two at that frequency. You need
somewhere between 10 and 20 milliseconds. And we are unable to treat
reflections that arrive during that rather long window separately from
the direct, first-arriving sound. This is true for microphones and
computers, too. That is why you cannot readily separate the effects of
the room from the response of the speaker when you do measurements in
your listening room: The room gets in the picture before you have time
to latch onto the energy content of the bass in the direct sound.

In the higher frequencies, this changes: If you are interested in how
much energy there is at, say, 5 kHz--for that, you need only 0.2 to 0.4
milliseconds --you have plenty of time before any reflections arrive,
typically. You can measure the high-frequency response of a speaker in a
room without "hearing" the room at all. You can get the "anechoic"
reflection-free response by just chopping out everything after the first
little bit of the direct arrival of, say, an impulse signal.

snip

Now you can see what is wrong with the old-style steady-state EQ: It did
not "hear" right. The bass was heard correctly, but the microphone
picking up the steady-state noise signal was lumping the whole sound
together in the higher frequencies, treating reflections and direct
arrival as a unified whole. By contrast, the ear-brain was taking the
direct arrival more seriously than the reflections, and ignoring (at
least to some extent) the peaks and dips that arise from reflections.
(Much experimental work has been done on the thresholds for this
phenomenon.)...

End quote.

This refers to old analog EQs, so your Behringer results will be better.
The article is worth reading, especially to see the remarks in context
without my snips.

Stephen