Deity PR-2 Pocket Recorder

[tapeheadtoo]

Setting it to mic in automatically provides the 5V PIP, correct?  I didn't see any separate way to set it.
If you set the gain at 0, and no levels are seen while recording a rock show, then there's no way to visually confirm you're recording anything, and there would be a TON of noise in post. 

I have only used the recorder once and kept the gain at zero. After processing the gain in Adobe Audition I noticed some noise.

Were you able to see levels on the screen?  What mics were you using?

Didn't check the levels in the screen and used DPA 4061 mics with the 5 volts PIP

[one8ung]

Manual says: Mic Plug-In Power   5V / Line In Switchable

[grawk]

The reason to run gain 0 is the gain is all digital so there's no benefit to running higher than that.

Is this also true for 24bit or only for 32bfp?

There is no analog gain at either bit depth.

[tapeheadtoo]

The reason to run gain 0 is the gain is all digital so there's no benefit to running higher than that.

But if I run gain at 0 and get an extremely low volume recording I need to apply digital gain in post anyway...?

Please, someone explain it to me like I'm 5:

I'm plugging my DPA 4061s terminating in 3.5mm straight into "in" jack on PR-2.  Input = mic (not line).  Auto Gain off.  Gain L=0 / R=0.  Fresh batteries (tried with both Deity-supplied lithiums and regular alkalines).  Turned up my speakers as loud as they go and hit record holding the mics close to them.  While recording I see levels on PR-2 screen no higher than -40.  Resulting file is very low volume which needs to be cranked up in post to be usable.  Can someone point out what I'm doing incorrectly?  Or is my unit defective and not supplying the 5V plug in power?

[grawk]

There's no difference between normalizing in post or running louder on the recorder, except that you can't go over if you do it in post. If you need louder going in, you'd want to use a preamp.

[tapeheadtoo]

The only reason I got the PR-2 was to eliminate the need for a battery box or preamp.  (I was the one who started the thread "CA-UGLY bit the dust, need another preamp solution for DPA 4061s" over in the battery boxes, preamps, etc. subforum.)

There's no difference between normalizing in post or running louder on the recorder, except that you can't go over if you do it in post. If you need louder going in, you'd want to use a preamp.

[grawk]

Maybe get to a louder spot for your recording?

You absolutely can turn the gain up on the pr2, it's just not going to lower the noise floor vs doing it in post.

[aaronji]

Or try more sensitive mics...

[one8ung]

The reason to run gain 0 is the gain is all digital so there's no benefit to running higher than that.

Is this also true for 24bit or only for 32bfp?

[Ronmac]

The above chart seems to confuse bit depth and sampling frequency characteristics.

Bit depth dictates dynamic range.

Sampling frequency dictates resolution.

Perhaps I am taking this out of context? Can you provide link to your chart, please?

[tapeheadtoo]

Update:
Tested above referenced configuration from front row of medium-loud rock show.  No gain setting issues with much higher SPL than at home.  I ended up setting the PR-2 gain to +3 and was happy to see levels bouncing around. 

What threw me off before:
Previously for these mics (DPA 4061) I always used a 9V battery box whereas PR-2 supplies only 5V plugin power.  It didn't occur to me there must be a linear relationship between voltage and response.  So the response was markedly different from what I've seen before testing other setups at home (speakers, claps, etc).

[TheJez]

The above chart seems to confuse bit depth and sampling frequency characteristics.

Bit depth dictates dynamic range.

Sampling frequency dictates resolution.

Perhaps I am taking this out of context? Can you provide link to your chart, please?

The charts show the difference between linear samples and floating point. With floating point, very low volume samples can be stored much more accurately than with linear samples.
I don't know where these charts came from, but here you can find a nice explanation of 32bfp vs linear:
https://www.sounddevices.com/32-bit-float-files-explained/

[Ronmac]

The above chart seems to confuse bit depth and sampling frequency characteristics.

Bit depth dictates dynamic range.

Sampling frequency dictates resolution.

Perhaps I am taking this out of context? Can you provide link to your chart, please?

The charts show the difference between linear samples and floating point. With floating point, very low volume samples can be stored much more accurately than with linear samples.
I don't know where these charts came from, but here you can find a nice explanation of 32bfp vs linear:
https://www.sounddevices.com/32-bit-float-files-explained/

I am well versed in the differences between 24bit and 32bit fp and how to use. My question is in the claim that 32bfp provides "higher resolution". It does not. The advantage is in dynamic range only.

It would be useful to see if I am missing some context by not being able to see the full explanation within that snip.

[aaronji]

^ I think you are correct. Nyquist-Shannon and whatnot...

[TheJez]

The above chart seems to confuse bit depth and sampling frequency characteristics.

Bit depth dictates dynamic range.

Sampling frequency dictates resolution.

Perhaps I am taking this out of context? Can you provide link to your chart, please?

The charts show the difference between linear samples and floating point. With floating point, very low volume samples can be stored much more accurately than with linear samples.
I don't know where these charts came from, but here you can find a nice explanation of 32bfp vs linear:
https://www.sounddevices.com/32-bit-float-files-explained/

. My question is in the claim that 32bfp provides "higher resolution". It does not. The advantage is in dynamic range only.

I tend to disagree... Consider very very quiet parts in your recording. When using 16 or 24 bit linear storage, you'd be using just a handful of these 16 or 24 bits for these samples, so you'll have relatively big quantization error compared to loud samples. Then, if you'd amplify this quiet audio, you will amplify the quantization error.
In case of 32bfp, even very quiet samples will use plenty of the available bits, so much less quantization errors compared to linear storage. So quiet samples can be amplified without amplifying the quantization error! That's what's being shown in the graphs...