Recent content by Yuri Korzunov
-
Why do high up-sampling/ over-sampling rates (DSD,) kill PRAT and aliveness of music? Any ideas?
Al, If there are features in numbers even, "good" term can't be defined. "Better" term is easier. It can be defined in numbers. But again: what is "threshold of audibility"? -
main difference between real sound and repro sound
Sound is wave Not once I noticed that live sound have more "volume"/"space", than recorded one. Sound is wave. And there works wave theory. Sound waves like light (optical/electromagnetic waves). Acoustic sound is sum or acoustics rays from source (musical instrument, singer, etc.)...- Yuri Korzunov
- Post #80
- Forum: General Audio Forum
-
The 24-Bit Delusion
For achieving maximal resolution used FFT window is very important. For spectrum analyzis tool recommended -200 ... -300 dB lower border for opening all adwantages of FFT 2^16. It need for analyzis of 32/64-bit float point formats. -
The 24-Bit Delusion
Ok. I agree. Multitones are cool ;) -
The 24-Bit Delusion
1. Multitone maybe have sense for a hardware devices that work in real-time. As example, for simultaneous estimation at several frequencies, measurement time economy, etc. It is a clear reasons. Same way may be used white noise for estimation of amplitude-frequency response. But using... -
The 24-Bit Delusion
Ok. You used multitone for testing. What conclusions about this test for perception you may officially claim as vendor? -
The 24-Bit Delusion
We talks about measurements. There all in numbers. Let describe "more revealing" term in numbers, please. -
The 24-Bit Delusion
Real music is stochastical process. Independent number of test sines (upper 2) we can't cover all cases. I suppose, we can't predict what percent of the cases covered by N sines even. So it lead to complicated analyzis only. Without practical sense. -
The 24-Bit Delusion
Yes. But without signal there will not noise. For checking overload ability enough take 2 sines and check full spectrum step-by-step by frequency or in sweep mode. This way slower than multitone. But it allow check overload ability for full frequency range in details. It may be automated... -
The 24-Bit Delusion
For quantization noise inside pure digital system need signal. For measurement of noise of apparatus - no. Modulation of noise is non-linear distortions. I noted above «distortions + noise floor». What is mean careful? Need exactly know that you want to get by measurement. Referring to any... -
The 24-Bit Delusion
Noise floor measured in silence mode. For single sine measured distortions + noise floor. I'd not recommend using realistic signals, including musical, because there to complex response that unseparatable with test signal. Enough use two sines and sweep sine. Anaysis should be in... -
The 24-Bit Delusion
When considered bit depth, reviewed noise floor and maximal 0 dB. But need consider 0 dB and minimal signal for given signal/noise ratio. Example: Dynamic range = 0 dB - V(dB); V(dB) is level of signal with SNR=40 dB. Because SNR for high signal is 110 dB, but for -70 db SNR about 40 dB (it... -
The 24-Bit Delusion
Except double blind hearing test. It is kind of measurement too. -
The 24-Bit Delusion
In the picture we really see measured noise level about -140 dB. If there are some doubts need exactly know settings the measurement tool. For same cases used protocols of measurements where exactly described all knobs/virtual knobs positions, modes of measured apparatus and measurement... -
The 24-Bit Delusion
In post above I corrected numbers for more right example.
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |