At 7:27 PM +1000 8/19/10, Paul Jacobson wrote:
>
>On 19/08/2010, at 1:30 PM, Rob Danielson wrote:
>
>> So we don't confuse anyone, recorders have input noise, mics have
>>self-noise.
>
>Hi Rob,
>
>Sorry to cause confusion. In reality we are looking at combined
>mic/preamp noise output rather than mic self-noise and preamp
>equivalent input noise in isolation.
Correct. There are two different sources of the noise in the test.
Mic self-noise is contributing much more which I feel it is crucial
for people to grasp. A phrase like, "combined self-noise and pre
input noise" is much clearer.
I'm not hearing anything close to a "significant" difference in noise
performance. For me, this means not having to use one combination of
mics/recorder over the other for about 98% of the recording I do. I
do like extra low end response and I'm very familiar with it from
using a 744T regularly and making .1 (LFE) tracks. I should carry a
back-up 4+ channel recorder and the 680 looks very tempting if the
money comes along.
> > Many level references and measurements can
>> be incorporated but it doesn't solve the predicament that the pres
>> have different frequency response under 80 Hz.
>
>It's not a predicament unless you are setting out to prove all
>preamps sound the same. ;)
? That's cruel. :-) .
> It's simple enough to use an alignment tone that is above the
>problematic frequencies.
We are in a discussion about two mic pres that sound nearly identical
and require sonograms to reveal more content under 50Hz. I'm okay
with adding another reference, but I'm not okay with using only one
reference. Pink noise and higher tones as you suggest introduce
prejudices to be leveled out too.
Any testing we do will highlight some aspects of performance over
others. I encourage you to propose another way to do such comparisons
that are a better fit for your criteria. For folks work in Reaper, in
the future, I can send the Reaper session I used so they can play
with the levels and further the experience.
>
>> If the perceptual tests we've been making over the last 6+ years
>> proved to conflict with results in the field or with each other, I
>> would have stopped making them.
>
>I agree the tests have generally been useful. In cases like this
>comparison of Emil's test recordings those who relied purely on the
>level matching in the posting got the wrong answer, simply because
>the two preamp/mic systems are so close that a 1-2dB is enough to
>change the ordering of noise levels for a casual listener. Of the
>four who got it "right", Raimund, Mike and myself all used analysis
>tools in addition to listening tests while Emil conducted the
>original tests. John and Tom who just listened to the file got the
>order wrong. I have to admit if I hadn't spent some time
>investigating after noticing that clip 4 was quieter than the
>previous clips, I would have picked Recorder A as having lower self
>noise.
Again, its "input noise," if its generated by the recorder's pre. I
assume you mean combined mic self-noise and pre input noise. This
test is very challenged to tell us how much recorder input noise we
are hearing.
If you determined Clip #4 to be quieter, the explanation is
complicated. Its mixed .9 dB _higher_ than Clip #1. There seemed to
be a slight difference in the sensitivity of the two AT4022's but its
a challenge to discern it amidst the robust, rapid clock ticks. To
deal with the difference I was hearing, I elected to mix clip #4 with
.9dB more gain than Clip #1 (the other mic on the same recorder) in
the overall flow. Any level difference between the two mics was
harder to discern on the A recorder so I elected in the end to give
Clips #2 and #3 get the same gain. Reaper will make .1dB
distinctions, but not in real time. The channel faders have minimum
steps of .9 to 1.0 dB. I'll try using the finer resolution settings
next time to see if it helps. I worked at the matches about 3 hours.
Its taxing on the ears splitting the dB's when there are several
clips.
Another factor the test revealed is the sound file created by the
DR680 required about 14dB gain to match that made by the SD702. (Both
at max gain). This suggests to me that the maximum available gain in
the 680's pre is around 52-55 dB. This could create modest
limitations in the field. I'd hesitate, for example, when recording
ambience in a very "quiet" location with moderately sensitive mics to
use 16 bit mode opposed to 24 bit mode. Rob D.
>
>cheers
>Paul
>
--
|