alfie wrote:One of my Indian friends is unhappy about Deep's dismissal as well ! Sometimes the technology is a bit vague but I don't think any injustices were done today
The Deep dismissal is a good example of the limitations of Snicko, and why we should not be using it in cricket. This explanation will probably go right over everyone's head, but a simple look at the envelope of the soundwave produced shows that the sound is not natural, it is not possible for instance for a ball striking a bat to create a sound with 4-5 different attack phases leading to zero decay/release or sustain periods. This would indicate multiple events causing new sounds, not one continuous sound. No percussive strikes produce such a sound, and as they are non-linear, its not resonant frequencies either.
The envelope shown indicates strongly that Snicko uses sidechain compression, which makes perfect sense. Sidechaining is an audio mixing technique that uses some form of trigger (either timed to a beat in music, or triggered by a frequency in a sound) that momentarily adds gain reduction to a set frequency range, while boosting whatever sound is filtered into the sidechain at higher gain - the goal is to add loudness and clarity to something you want to momentarily stand out in the overall sound, at times when that sound would be masked by others. I guess the best way to explain it is whenever anyone on TV or the radio talks over a soundtrack in the background, you setup a sidechain on the microphone so when you turn it on, you trigger a reduction in sound in the soundtrack so you can hear the voice clearer.
In this case, you setup the sidechain to pick up the fundamental frequency of the edge of the bat, at which point that sound will be gain amplified and be more pronounced on your oscillator showing the edge, while at the same time placing a filter on the ambient noise to remove interference. The problem with sidechaining is, using frequency triggers are notoriously unreliable because anything can trigger them, at which point the gain boosting occurs. This "double pop" or phantom strike is a common problem in audio production - even the air displacement of a drummer pressing a kick pedal down can trigger the sidechain, giving you the end result of what sounds like him striking the drum. You look at a phantom strike soundwave, and the give away is multiple attack phases with no decay/sustain etc. You can quite literally make it sound like something has been hit when it has not... with the drum example, its sounds very clearly like the drum has been hit twice.
Taking this dismissal, the third spike on Snicko looks like a classic edge - maybe it is, but we cannot be sure. What is clear is, as the ball gets closer to Deep the oscillator is picking up more ambient noise as the ball gets closer to him. There are 3-4 quite large murmurs on snicko which do not appear as the ball is far away in flight, so this is quite conclusive proof to me that the movement or displacement of the air or something external is registering - and if the oscillator is picking up other interference, and increasing as the ball gets closer, then you cannot say with any certainty that the "edge" registered is not that ambient sound setting off the sidechain, which when it goes boom with all that gain addition, makes it look like an edge when it could just be anything.
The key is the first two sounds before that big spike - what are they? They are in no way consistent with strike contact, but are increasingly audible/higher amplitude. And the 4th attack phase after the spike for the edge? The ball hasn't hit anything, so what is producing that rise in amplitude? It cannot be anything, unless its something artificial in the setup ... its clear indication that Snicko is de facto creating these sounds.
Bottom line - Snicko needs binning. It's an unsophisticated and hugely inaccurate way to judge what its there for.