RODE NT1 5th Gen 32-Bit Float?

[KenTheriot]

Does AA support 32-bit float such that clipped audio is repairable? I'm testing the RODE NT1 5th Gen and even though I've installed the RODE ASIO driver AND can select that in AA, and AA can see the 32-bit Float mode, it actually can't record (?). At least on my Windows 11 computer. I keep either getting an error, or sometimes it DOES record, but cannot deal with the actual benefit of 32-bit float such that I can just normalize clipped audio back under 0dB and have it be completely fine (no clipping).

I am able to do this just fine in Reaper. But if it does work in AA, I'm not seeing how. Thoughts? Thanks!

[Rag and Bone · Accepted solution]

This page on the Rode site says that 32 bit float is supported on Audition for Mac, but not for Windows. (Scroll down to "Recording With the NT1 5th Generation in 32-bit Float").

Maybe take this up with Rode? Or, record in Reaper, and import the file into Audition?

[SteveG_AudioMasters_ · Accepted solution]

The trouble here is that there are several different issues to consider. No, Audition on Windows doesn't appear to support 32-bit Float recording, and I think that this is something for Adobe to look at, rather than Rode - if Reaper can do it, then it's clearly not inherently a Windows issue, is it? The second thing to note is that this system cannot repair clipping caused by actual mic overload - all it does is to enable you to record the entire dynamic range of the mic without the need for gain alterations. In other words, if you overload the mic from an acoustic point of view and the diaphragm hits the end stop, then nothing is going to correct this; 32-bit Float recording is not a universal panacea for poor mic practice. So no, it won't correct for plosives, because they're inherently physical. Do we really need 32-bit Float recording? It's amazing that we've managed pretty well without it for all these years, isn't it? I think it's just a way of encouraging bad practice, personally.

Don't start me on Windows 11... I've now seen first hand SSDs that only run at half the speed they do with Windows 10, and the whole system seems to be somewhat unresponsive. And we've had other reports here about sync and graphics issues. I will not be installing it on anything until I'm absolutely forced to.

[SteveG_AudioMasters_]

I should perhaps explain why I think that the 32-bit Float is an Adobe issue. Audition is perfectly capable of creating a 32-bit container and letting you record into it - it does that now. But what happens is that although the Float values are entered (the mantissa and sign values), the exponent is left full of zeros. And these are the eight bits that the Rode provides values for. Personally I think that this should be a relatively straightforward fix, and is all that's required. And I think it's a bit off that it works on a Mac and not a PC!

[KenTheriot]

Thanks guys! Just wanted to to get the facts and this explains it. I did end up diong my recording in Reaper.

BTW, in my experience, clipping is not from the physical mic. But rather from having input gain on either an interface or software being overloaded. Not that it doesn't happen. But I've sung some pretty belting notes (like literally almost yelling) into LDCs and not had clipping. I think it's much more common for it to be something in the chain after the mic.

[Christopher5FD1]

Very confused with this because of lack of information out there. Adobe Audition on windows certainly has the record in 32 bit float at 192kHz, and I have generated loads of files. Looking at the wave forms on the tracks, there does seem to be some slight clipping when  moving mic around. I hardly used Audition so not sure if I just need to readjust the track hight or what ever. 
Once saved I have the Wav files and the Raw file, but unsure with workflow from here. Has there been an update, or is this still not 100 percent 32 bit float.
I actually bought it for survelliance, as a neighbour was firing shots at my wall during a party last year, and I didnt want any clipping. Lol, yeah I know, pardon the pun or double entendre rather.

[SteveG_AudioMasters_]

The look of the waveform has no impact on the recorded material whatsoever - anything that looks like clipping is purely visual, and reducing the screen amplitude fixes it.

I haven't checked recently on whether anything has changed regarding 32-bit float recording on PCs, simply because for me, it's not an issue, but hey, I'll look into it.

Why is it not an issue? I bought a Tascam Portacapture X8, and that records 32-bit float directly onto SD cards. I can still set record levels perfectly sensibly on my own, primarily because I tend to record in controlled environments.