What Audio Codec IC is used by the T2i

[Robert Rozak]

Greetings,

Can anybody tell me what audio codec IC is used by the T2i? I'd like
to be able to look at the datasheet to see the register table. Is the
the AK4646 again (same as the 5D)?


Best Regards,

Robert

[Trammell Hudson]

On Sun, Aug 08, 2010 at 07:39:11PM -0700, Robert Rozak wrote:
> Can anybody tell me what audio codec IC is used by the T2i? I'd like
> to be able to look at the datasheet to see the register table. Is the
> the AK4646 again (same as the 5D)?

I believe it is the AK4646 (or a compatible chip) still. All of the
audio gain functions use the same register values and SPI commands
to write to it, so I didn't modify our audio functions at all. Even
the functions to read the audio levels is unchanged.

--
Trammell

[Ba...@aol.com]

Robert, Trammel:

 

Hey guys.

 

I have tried the headphone listening yet.  

 

I am kind of shooting in the dark here,    but I am guessing the zero gain for the camera is a noisier higher level than the 5D has.   I remember that when we set the 5D at 10 analog,  zero on the dgain, we had a pretty quiet noise floor.    I messed with a lot of variables in these initial tests,  but the one showing the floor is not moving,   tells me I am not affecting gain levels by entering different numbers in the gain settings like we did with the 5D.  Auto gain seems to be off,  and I got the feeling I was affecting things  at first, but I was also messing with different level settings on the Juiced link gain.   It seems like I can cut the Juiced link levels back a lot, into the middle level switches,  but none of it gives a lower noise level floor.    I could pump up the Juiced link,  recording at the higher levels,  and then bring the floor down by reducing gain in post,  I guess.

 

chris

 

--
http://magiclantern.wikia.com/

To post to this group, send email to ml-d...@googlegroups.com
To unsubscribe from this group, send email to ml-devel+u...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/ml-devel?hl=en

[Robert Rozak]

Chris,

"It seems like I can cut the Juiced link levels back a lot, into the
middle level switches"

That seems to indicate to me that you're heading in the wrong
direction ... ?

Noise in the camera comes from 2 sources:
1) The SNR of the mic pre. These types of amps will typically yield
better SNR at lower gains.
2) High digital gain after the A/D (AGC on or off), which forces you
to inject a low signal into the A/D, which is not signifigantly higher
than the A/D 1-bit quantization noise. I beleive that this is the
bulk of the N setting system SNR.

As you throttle back digital gain after the A/D, you should be pumping
up the gain/trim in the juicedLink. So, you will be injecting a much
higher signal into the A/D relative to its 1-bit quantization
noise ...


Robert

[Ba...@aol.com]

Robert:

 

Yes that is what I was thinking,  too.   But I am not so sure that with what we got so far, that we are really cutting back the digital gain by changing/editing the config file we have been working with for last few days in this new version of ML for the T2i.   I think so far we have disabled auto gain,  but I am not so sure that by using same settings as we used in the config file in ML for the 5D that we are actually affecting gain level.

 

 

Chris

[Robert Rozak]

Sorry for being so software ignorant.

But, is there a way to work this problem backwards? Is there a way to
use something like the AGC disable signal into the T2i, and then read
from some registers in the T2i to find out what the AGC algorithm is
setting the digital gain to, and then use that to program those values
into ML?

-R

On Aug 8, 7:56 pm, Ba...@aol.com wrote: