TheNavigator wrote:And yes, I'm all for 192 kHz processing in some plugins, processors, also synthesizers. But neither for recording, nor playback.
That begs the question:
If it is OK to have a number of high resolution intermediate states, doesn't it make sense to start with and maintain that resolution for as much as possible?
FFTs help transform time-based measurements into frequency-based measurements.
The significance of this is that:
1. A possible infinite number of frequencies are represented by a finite number of frequencies which are sub-multiples
of the sampling
frequency and hence are NOT necessarily harmonically related to ANY frequency in the waveform. The significance is that ALL those representation frequencies MUST be included in calculations, even if they are above (or below) the hearing in-band.
2. The resulting finite values are scalar, which means calculations are reduced from a possible infinite number of trigonometric ones to finite matrix arithmetic with linear values, making processing much more predictable, time-wise!
What the matrix operations are to emulate an LA-2A are, I don't know. The above is what I gleaned from the web after someone on the Soundsonline forums also offered FFTs as some sort of proof of their sample rate assertion. I still don't know how such knowledge of FFTs really offer any explanation in relation to choice of sample rate frequency as the calculations are in a whole other domain to the way I tend to think about waveforms.
Also, the process sounds simple conceptually, but there are other considerations, such as introducing guard bands at each end of each block of samples to take the sample from 0 and back to 0, all to prevent introduction of spurious 'noise' from the abrupt ends into the calculations.
@TheNavigator, please explain how FFT theory backs up your assertions about sample rates. Without that, or a reference to such an explanation, I cannot ascertain the veracity of your assertion. Basically, you have made an appeal to 'false authority' by referring to an unknown person without a traceable line of proof of your assertion.
I do NOT really know here, but I am not prepared to blindly accept unsubstantiated assertions without an explanation or a way of reasonable finding one. I did not originally want to get into FFTs with the other discussion on Soundsonline, but I was willing to take a leap of faith and follow the provided info.
However, it was another red herring in that instance because neither the person or the info they cited actually provided the causal link between FFT theory and a practical choice for a recording and processing sample rate, at least not without a lot more explanation. It was clear that while they worked with DSPs, they did not seem to have the sound theoretical knowledge to provide that explanation.
It did help me understand their benefits, as I explain above.
In my other line as a Technical Writer, I do not accept vague assertions and partial explanations as true. It often takes iterative interviewing of multiple people to get at the full explanation and ALL the steps (no 1 in 3 origami steps) involved in processes. One cannot build knowledge on a bed of sand, and assertions are logical sand, and ridicule is just trying to blind one with that sand!
Half of the folk music duo, DevaKnighT. http://devaknight.com/ http://youtube.com/devaknight2
Music available from CD Baby (MP3/FLAC at http://cdbaby.com/Artist/DevaknightT
) and the usual culprits. All recorded and processed on Cubase 7.x at 192k.Comp:
i7-4930X, Asus P9X79-E WS, 32GB, 3xPalit GeForce 9400, SSDx5, UAD-2Q, Dell S2340Tx3, Dell 3008x2, TV. HW:
Nmn U87 Ai, JA251x2, YRG, Korg padKontrol+microKEY2, RME FF400+800, M-Patch-2, Tannoy Reveal Active+TS-12. SW:
Win 8.1 Pro 64, Cubase 7.5.10 64, RX3Adv, Goliath, SO Plat Pro, Pianos, Gypsy, Sup.Drum 2.0, AT7, Ozone5Adv.