Topic: Tascam FR-AV2 new compact 32bit float recorder.

[Niels]

Roland Harris has a very interesting test of the FR-AV2 compared with the SD MixPre 3 (v.1) and SD 788T.

https://drbadphil.com/dual-adc-and-32-bit-float-tests-with-the-tascam-fr-av2-with-comparisons-to-the-sound-devices-24-bit-mixpre-3-and-788t

The lack of noise on the quiet MixPre 3 (v1) recording when lifted 70db in post is very impressive.

[datbrad]

Great Article. Aside from the topic of dual switching ADCs with 32 bit float, I was reminded how proper gain staging between microphones and mic preamps is still one of the most important considerations when matching gear for a signal chain, regardless of the final capture method. The math has been around since the days of analog tape, and there are x/y type charts available online to make it easy. You just need to know the sensitivity, self noise, and max SPL of the microphone being considered. Then you need to know the preamp's max input capacity and it EIN rating.

He mentions a Sennheiser shotgun with a sensitivity of -24dbu, and I found it's max sound pressure is 129db.

With a mic that sensitive and a preamp with max input of +4 dbV, the preamp will overload before the max sound pressure of the mic is reached, but not until your source is 120db.

Based on this, I see almost no way the overload point of the mic inputs could be realistically reached with the mics we use for recording music with average sensitivity specs in the -36 to -40 range, not out putting +4dbV to the input of the preamp until they are subjected to 134db to 138db of sound pressure.

I assume we all know the dynamic range of a microphone is the max sound pressure, minus the self noise. Reading the article, I was reminded how the effective audible noise fron the mic to the preamp is impacted by mic sensitivity. Going back to the mk4/cmc6, even though the self noise is 14db, with -36 db sensitivity the effective noise level (what you can actually measure) is down to -116 db at 1 pascal. Pretty darn quiet I think.

[aaronji]

The lack of noise on the quiet MixPre 3 (v1) recording when lifted 70db in post is very impressive.

Probably because the first generation MixPre's do the whole dual ADC trick, but don't save it in a 32-bit floating point format.

[TheJez]

The lack of noise on the quiet MixPre 3 (v1) recording when lifted 70db in post is very impressive.

Probably because the first generation MixPre's do the whole dual ADC trick, but don't save it in a 32-bit floating point format.

Saving the output as 32-bit float (instead of 24bit) won't affect the SNR at all, so this has nothing to do with it.

[grawk]

The lack of noise on the quiet MixPre 3 (v1) recording when lifted 70db in post is very impressive.

Probably because the first generation MixPre's do the whole dual ADC trick, but don't save it in a 32-bit floating point format.

Saving the output as 32-bit float (instead of 24bit) won't affect the SNR at all, so this has nothing to do with it.

The dual adc is what enables the lower SNR, regardless of format used for the container.  So the 24 bit mixpre gets all the benefits, without the buzzwords. 

[Gutbucket]

^ 

[Kyle K]

I just noticed this and it appears to be a major defect with these devices.

In the image attached, the particles you see are *trapped under the screen*. I only just removed the original plastic covering the device. I like to keep it on as long as it will generally remain affixed. I have not ran these devices particularly hard. Less than 20 runs. It is a bit unfathomable how this is occurring. Never had any similar issue on any device with a screen, those of which I've ran for years longer.

[Gutbucket]

Dead pixels?  Possible if they failed to on/white, rather than off/black.

If you have a particularly clear photo of it, zoom in to more closely examine the edges of each spot.

[goodcooker]

Those are dead pixels for sure. Screen is beginning its decline. Return for warranty replacement if you can.

[Kyle K]

No no, not dead pixels, somehow physical dust or skin flakes or some shit are getting UNDER THE SCREEN. I'm not some gross person living in squalor, I have no clue. All I do is take my gear to show, use it, store it. I have no idea where the air gap is that is creating this issue but I would have to imagine this will begin occurring with all devices. And it is the sort of issue they may refuse to acknowledge or resolve, we shall see. It's occurring with both my units.

I imagine I can find the culprit with some compressed air testing each potential opening, and blasting it to the edges, but wtf. It is SO MUCH and it is so gross. It is a defect related to air gap and the screen overlay not settling flush. I can press down on it lightly and sethr articles moving, far more 'give' than there should be.

This will stop be from purchasing another. I was considering a third.

[gmm6797]

...This will stop be from purchasing another. I was considering a third.

What does Tascam say about it?

[Kyle K]

Awaiting their reply, will report back. Curious to what degree others see or don't see particles under screen with device off after ~10 runs.

[gmm6797]

Awaiting their reply, will report back. Curious to what degree others see or don't see particles under screen with device off after ~10 runs.

I have run mine some 50 times this calendar year alone, no issues. I ordered mine as soon as it came out... about 100 bands (not shows) so far.

[datbrad]

I'm going look at mine as soon as I get home from work. The particles seem to extend beyond the actual OLED screen into the matt frame under the outer plastic cover. Have you had any performance problems or just the pixie dust?

Btw, if I showed a pic like that to my taping buddies 15 years ago, they would say I need to stop using the bullet directly over my deck, lol.

[grawk]

The dust wouldn't bother me, so I can give you $125 for your defective recorder.