Cannot play at 24 bit depth 192 khz

[tam2000k2]

Hey everyone,

My old Mac Mini has ended and I have purchased a new Sierra 10.12/MacPro (2.7 GHz 12-Core Intel Xeon E5, 16 GB 1866 MHz DDR3, AMD FireFro D700 6144 MB) to take it's place.

On my old Mac Mini, all my projects were set at 16 bit/44.1 khz. With the new Sierra 10.12/MacPro, there is no support for firewire, and this means my ProFire 2626 is not supported, which lead me to purchase the Focusrite 8 Prex Thunderbolt. As a result, I have the ability to change my project settings to 24 bit/192 khz, but it will NOT play the VSTs properly - jagged sounding, like a car that will not start sound.

My question is, do I have to create new projects with the 24 bit/192 khz first, followed by copying and pasting the tracks(midi) for this to work?

I find it hard to believe that this computer does not have the processing speed or power to play/record at 24 bit/192 khz or is it that the VST(s) (Native Instruments Kontakt 5, Toontrack EZ Drummer, etc...) that cannot handle the settings?

Thanks in advance

[PeppaPig]

I just don't see the point of such a high sample rates - many others' would argue against me, but (IMHO) you're just shifting much much more data around to attain an improvement that is quite frankly undetectable. What is your ultimate medium? CD? DVD? personal FLACs or WAVs? Lossy MP3? I can see the point of recording at 88.2 and then dithering down for CD. Surely all you're doing is providing perfect reproduction of incidental ultra high frequencies to keep dogs and other animals amused/(distressed!)

Just my 2c of course - other opinions are available...

[djw]

I can see the point of recording at 88.2 and then dithering down for CD.

For the record, dithering is used when lowering the bit depth, not frequency.

Frequencies higher than double human hearing is unneeded if your plugins can oversample by itself, otherwise it can prevent aliasing (artifacts that fold back into the audible range). In the end you basically take a low pass filter to get it down to 44.1/48kHz.

[PeppaPig]

Yes, I know, but note the "24 bit" bit? Here's a post from me a couple of days ago explaining dithering is for bit depth reduction: viewtopic.php?f=226&t=106327

[djw]

Yes, I know, but note the "24 bit" bit? Here's a post from me a couple of days ago explaining dithering is for bit depth reduction: viewtopic.php?f=226&t=106327

I mentioned it to avoid having threads like this derail into disinformation as I've seen happen many times. I wasn't trying to say that you don't know.

[Raphie]

You can't use many plugins realtime at 192/24 you will need al hell of a system and even then you will be limited to a few instances. So there is nothing wrong. 44.1 is fine btw, some prefer 48 from nyquist perspective, just to be sure, others 88.1 but higher isn't really needed.

[PeppaPig]

You can't use many plugins realtime at 192/24 you will need al hell of a system and even then you will be limited to a few instances. So there is nothing wrong. 44.1 is fine btw, some prefer 48 from nyquist perspective, just to be sure, others 88.1 but higher isn't really needed.

Good point, each doubling of sample rate pretty much doubles the processing power required, many high quality plugins upsample anyway to avoid aliasing so you're not necessarily going to get an improvement at higher sample rates. Personally i think 48 gives you a slightly better sound stage (given quality DACs and soundstage sensitive material) but beyond that I cannot tell the difference. Certainly for most material 44.1 is fine, that said I would produce in 88.2 to be "safe" given no hardware limitations.

[neilwilkes]

I am a fan of high resolution audio but I honestly do not see the sense in using a 192kHz sample rate as it is actually more likely to degrade your audio than it is to improve it, depending on the source. For example of you are coming off tape you are actually going to cause problems with things like biasing frequencies and unless you use a properly configured LPF you could very well alias these down from a spike to something larger & a lot nastier in the audible range.
96kHz is fine, and so is 48kHz too (the old argument about 96k not being resampled correctly to 44.1kHz is long since a dead issue with a decent SRC such as R8Brain Pro or Saracen or WaveLab's Crystal Resampler) and they do sound different to me although I would not say one ios going to be always better than the other - it depends on the project.
But in all honesty, why bother with 192?