Should the DRS be mandatory?

[sussexpob]

I think there is only one problem with Snicko, and that's the way it's presented - as an animated waveform. If they did what I suggested earlier, I think it could be pretty reliable: we'd have a video lasting just 20ms in which you may or may not hear a click. It's pretty unlikely that some extraneous source would produce a sound that sounded like a ball nicking a bat in that time. Things like crowd noise and wind would sound completely different and can be filtered out anyway.

In the Computers thread you helped me set up audio recording software, I have experience in recording bands as a hobby and my old man and his friend use to run a music studio where I used to sound engineer, so I am not just talking generally here but with some knowledge(although amateur, so I dont know everything and all the science behind it).

To put it frankly, recording a live sound from a microphone so restricted is really tough. Each sound has a "sweet spot" where the direction and the distance of the microphone to the sound source create a vast difference in the sound you hear.... a sound source very close to a microphone will cause a huge bass boost for instance and increase amplification naturally, the further away you get more reverb/delay and sustain which will produce a far less defined and clean sound... so in essence, a edge sound can be a hugely different just in the position of where in the crease the batsman hits it.

I would guess that they record ambient sound, so things like the wind/crowd will all produce background interference in the mix. Its not a case of "filtering" these sounds out, you only have once source of sound produced from a single microphone, so filtering away different frequencies also effects the main sound that you are trying to anayse.

Everything is amplified, so the sound of a bat coming towards the ball in the air is also amplified. In order to distinguish the sound of an edge in the mix you would have to boost higher frequencies in the mix, while lowering your mid and low range bass frequencies.... this means that the "thud" of the pad will also sound higher, it means the "whoosh" of the bat will sound higher, a scratch of a wicketkeepers feet behind the stumps will feel higher.... and as its ambient sound, they will probably filter a lot of the reverb and sustain in each sound away for a crisper sound...

In essence mixing a sound can make anything sound like anything... and without it frequencies of different sounds clash and produce a "muddied" effect, or with ambient microphones, sound will delay, meaning that a clicking edge sound slowed down to correspond with a picture could just be the delayed sound of something that has already happened and still occuring.

Also with such a live recording of sound where the source is moving, a standard sound setup will also give you high natural differences in the same sound produced because it happens to occur in different places for the microphone to pick up. An edge from 18 meters away with a filtering of low bass sounds will produce a more defined knick then from a few inches away, because the knick will sound much much much more bassy anyway.

[DeltaAlpha]

Yeah, DRS is like Mary Poppins, D/L, "practically perfect in every way". How stupid of some of us even to discuss it...

[DeltaAlpha]

I've read your post, sussex, and, while there is a lot of truth in what you say, I'm not sure that it applies to a sound clip that lasts just a fraction of a second.

[D/L]

Yeah, DRS is like Mary Poppins, D/L, "practically perfect in every way". How stupid of some of us even to discuss it...

Adding little to the debate there, DA.

[sussexpob]

I've read your post, sussex, and, while there is a lot of truth in what you say, I'm not sure that it applies to a sound clip that lasts just a fraction of a second.

The same applies to sound regardless of length of the sound clip, surely DA? Honestly I can master a sound by mixing the frequencies to sound like anything... in fact a bass drum slowed down with its base removed sounds like the whistling of a train going by....

[DeltaAlpha]

The same applies to sound regardless of length of the sound clip, surely DA? Honestly I can master a sound by mixing the frequencies to sound like anything... in fact a bass drum slowed down with its base removed sounds like the whistling of a train going by....

Yes, I know that can be done, sussex, but there are two differences. One is that the sounds we're talking about aren't well-defined frequencies but basically noise, which contains a wide range of unrelated frequencies, and the other is that they wouldn't be tweaking the frequencies anyway. Maybe amplifying the sound or filtering some out, but not altering the actual frequency.

The thing about the the length is that, in such a short time, background noise is unlikely to change significantly, and it is very unlikely that a similar-sounding extraneous noise would occur in that time.

I've made a video of a marble being dropped on to a table, and clipped it to just two frames. I've yet to figure out how to post it (I don't intend to join YouTube), but two things are clear: it sounds like a marble being dropped on to a table and neither of the frames shows the marble in contact with the table.

[DeltaAlpha]

Well, I've tried to upload it somewhere! Photobucket plays it, but doesn't return to CMS when it's finished; I did try YouTube, but it says it can't process it and CMS won't accept it as an attachment.

Anyway, I find it convincing...

[KipperJohn]

I'm more interested in this concept of 'a second chance'.

The whole basis of the sport is that the batsman has no 'second chance - being given out should be the equivalent of being shot dead!

The sound of being bowled was called 'the death rattle' for a reason - there's no coming back. The raising of the finger by the umpire was the final death knell of an innings - it was a finality which set cricket apart from other sports.

By the same token, the words 'not out' signalled a reprieve from the hangman's noose, whilst the bowler and fielders trudged back like lions robbed of the kill.

With DRS much of that has changed and cricket is much the poorer for it. These days I can't wait for the 'reviews' to have been used up - then one can watch, and share, the disappointments and celebrations knowing that they are not a false dawn or a prelude to finding out that Hamlet hasn't died yet- in fact he's going to be on stage a bit longer - indeed he might never die if they declare....

[sussexpob]

Yes, I know that can be done, sussex, but there are two differences. One is that the sounds we're talking about aren't well-defined frequencies but basically noise, which contains a wide range of unrelated frequencies, and the other is that they wouldn't be tweaking the frequencies anyway. Maybe amplifying the sound or filtering some out, but not altering the actual frequency.

When I refer to "mixing" I refer to it in musical or sound engineering terms, not saying you "tweak" or change Frequencies DA! As frequency governs pitch and key, that would be a form of transposing (I would guess!). You are quite right though, the sound is made up of a range of different frequencies and when you "mix down" the sound you are increasing or decreasing the amplitude of the wave at certain levels of frequency, which in turn boosts the presence of the sound at certain levels and helps distinguish between sounds. You can filter the sound too, its common to use a low/high pass filter to disregard extreme frequencies at a certain cutoff threshold.

Your marble example doesnt really apply to be honest. The sound setup of a stump microphone will pick up ambient sound all around, and because the microphone is so far away from what it is measuring, and is being used in such a poor acoustic environment, the microphone sound will have to be amplified quite significantly. The problem is that it is one sound source, so any filtering or amplification of that sound source through mixing also has the effect of filtering, amplifying or limiting elements of the sound that you actually want to record and hear(and all those you dont). Remember also that a very slight snick of a bat is probably not as loud as a lot of things that are going on around it, and humans hear midrange frequencies much better than anything else, so in essence a high pitched snick and a mid range clip of the pad at the same time would mean the second would mask the former, even if they are the same decibel level.

What ends up happening is the filtering of all frequencies from the mid to low range, and possibly even amplification of high frequencies so what you are left with in terms of noise is things that produce a high pitched sound, deemed most noticeably to be the snick of an edge of the bat. When you hear a sound it is assumed to be the bat.

The problem stems from close calls, say for instance when the ball goes through the gate with a batsman playing outside the line of the ball and an audible sound is heard after flicking something... the sound itself could stem from the inside edge or have flicking the top of the pad, both sounds in terms of timbre would be very similar because they both have low attack, low sustain and slightly low decay levels(envelope, which actually makes sounds of the same pitch and volume sound different).... and with the "dumbing down" of the actual sound the difference in terms of what you hear is diminished, the sound has been adapted to make it sound high pitched in order to make sounds off the bat more detectable! You could go further by commenting on the fact that ambient microphones placement also completely interferes with the attack and sustain making similar sounds sound identical in terms of Timbre!!

Hotspot also hasnt helped snicko. As I said before snicko came first, and the focus was not on the production of sound but on the presence of certain sound. What you find now is that, if an umpire has given a batsman out and hotspot doesnt detect anything, the presence of any sound on snicko or the microphone will mean he is out... just check KP or Khawaja, the clear indication even on these boards was "a sound was heard". It didnt matter that these sounds didnt really sound or look like edges!

Another thing that puzzles me is, if snicko registers a snick surely it is impossible that Hot spot does not? When a sound is generated usually the sound is absorbed in the form of heat energy in the surface that produced the sound, so this would surely show on hot spot( I would stand corrected on this point as my understanding of thermoacoustics is not great)

[mikesiva]

The Rogers incident is a sore point with me. I believe the system should be used to arrive at the "right" decision. If he's not out caught behind but he's out lbw then IMHO he should be out. This technicality of what you're appealing for is a nonsense....

[DeltaAlpha]

It's clear that you have a great deal of practical experience in audio technology, sussex, whereas I don't, but really what you're saying comes before what I'm arguing. It's clear from looking at the Snicko waveforms on TV that somehow they've managed to overcome most of the difficulties you outline and produce an audio signal that, from my experience, would easily be audible as a click. Below is a waveform from my 'marble' video, and I'm sure you'll agree that it looks very similar to a Snicko waveform - and the click's very clearly audible.



The reason I'm arguing this is because of the apparent issue with the time it takes for the Snicko result to be shown. I attribute most of this time to the time it will take to convert an audio waveform into a video and synchronise that with the actual video, neither of which is necessary if one simply plays the sound. And that's quite apart from whether one can attribute the sound to its source by listening to it.

Part of the reason I made the 'marble' video was to demonstrate that there may not be a video frame corresponding to the time at which the ball passes the bat, and that seems to trouble some members; this was clearly evident in the video because, although the marble clearly bounces, it's never shown in contact with the table. Unfortunately, I haven't yet found a way to post it.

[braveneutral]

I don't know much about the Rogers incident as I haven't seen it but fielders do not need to specify what they are appealing for (unless there is an ICC regulation) as far as I am aware.

[Arthur Crabtree]

I think everyone accepted that. But DRS is a decision review system. It doesn't review the appeal, it reviews the decision.

I think most are happy with how that went. My only slight quibble is that the rule says that the lbw, or secondary review, is automatically opposing the onfield decision. But it seems fair enough if the third umpire asks the onfield umpire to make a new decision. It may be that the umpire would have given LBW had he known that there was no edge. But the rule assumes he wouldn't have.

[hopeforthebest]

I think DRS should be forced on everyone, why should we have all the fun.

[braveneutral]

I think everyone accepted that. But DRS is a decision review system. It doesn't review the appeal, it reviews the decision.

I think most are happy with how that went. My only slight quibble is that the rule says that the lbw, or secondary review, is automatically opposing the onfield decision. But it seems fair enough if the third umpire asks the onfield umpire to make a new decision. It may be that the umpire would have given LBW had he known that there was no edge. But the rule assumes he wouldn't have.

Perhaps they should be reviewing the basic message of the decision - out or not out - without taking into consideration the mode of dismissal that the standing umpire judged for. Of course I am talking blind here.