Hi Mike,
Having just finished the production period for the Australian Tour of "Titanic - the musical", a couple of things have come to light which would suggest a possible modification.
The first one is Fader Scaling - in this show, as one could imagine, there is a greater than usual dynamic range in the SFX playout - between subtle water effects and the ship hitting the iceberg, for instance. I've found that in this case the on-screen faders are too "jumpy" ( dB change per minimum movement) at the lower end, whereas they could have less detail at the upper end. Any chance of an alternative scaling for such shows? I had to re-process some sound files with a built-in attenuation to get the fader up to a useable point in it's travel, and most of the quieter cues required level changes via the text box as the minimum move of the fader via the mouse caused too big a jump. I'm talking around the -30 range.
The next thing is that it would be great to have some way, in the editor, of playing a full cue (with all it's subcues) to audition the effect of a change on one of the subcues while still in the editor, rather than having to exit the editor to do that. My way of programming complex effects uses a single cue per effect, containing subcues concurrently running different "stems" (to use a film term) of the elements of the effect. If one wants to increase one of the elements, it is useful to be able to make the change then easily hear it in the context of the whole cue.
Cheers,
nick
Fader Scaling and other things
Fader Scaling and other things
Nick Reich
System Sound Pty Ltd
System Sound Pty Ltd
fader scaling/scales
I tend to agree with Nick on the resolution too low at the bottom of the fade matter. As a compromise in signal/noise on the analogue side, and using the most bits of resolution on the digital, I'll pick a level (typically around -18dBFS) to equate to 0dBu on an analogue desk (that is to say, that the average level of a voice-over will be -18dBFS leaving 18dB of headroom for peaks, "yells", and other transients; when that sound hits the input of the board it translates to average around 0dB on the meter, still leaving 18dB of headroom which falls well within the >22dB of headroom above 0dB on most descent boards.)
What this means though, is that some subtle cues wind up very low on SCS's fader where small movements of the fader result in very large changes in the apparent volume of the cue.
What I would really love to see, is a fader more like those on your typical analogue (or digital) board, with a nominal starting point of 0dB digital gain or reduction, the ability to add +6 to +10dB digital gain, and a fairly constant fader resolution over the course of travel between say +10 and -40dB. Better boards with 100mm faders (not those damned annoying 60mm faders on the smaller mackies) will typically give you a control "sweet spot" that wide, and I've only rarely come across cases where that was not adequate.
Thanks again for a great product.
What this means though, is that some subtle cues wind up very low on SCS's fader where small movements of the fader result in very large changes in the apparent volume of the cue.
What I would really love to see, is a fader more like those on your typical analogue (or digital) board, with a nominal starting point of 0dB digital gain or reduction, the ability to add +6 to +10dB digital gain, and a fairly constant fader resolution over the course of travel between say +10 and -40dB. Better boards with 100mm faders (not those damned annoying 60mm faders on the smaller mackies) will typically give you a control "sweet spot" that wide, and I've only rarely come across cases where that was not adequate.
Thanks again for a great product.
-
Mike Daniell
- Site Admin
- Posts: 3629
- Joined: Sun Jul 24, 2005 8:58 am
- Location: Brisbane, Queensland, Australia. TZ:GMT+10
- Contact:
I'd like to see that as well, but need assistance with the maths! If anyone can translate a linear scale to a dB scale similar to a typical sound board (eg max level of +12dB) then please contact me.What I would really love to see, is a fader more like those on your typical analogue (or digital) board, ...
-
Couch Potatoe
- Posts: 9
- Joined: Fri Mar 17, 2006 4:30 am
- Location: Kent, UK
Hi Mike,
Besides, this is a feature I'd LOVE to see in SCS
I just need to know the Logarithmic range you want to cover (-120db TO +15dB?) AND the linear range you want to cover it (0 to 65536?) and I'll sort something out!
It might also be useful to know how you handle/process gain changes in SCS.
Cheers for now!
Couch Potatoe
BTW Many, many thanks for the "run-cue-at-specified-time-of-day" feature added recently, it worked like a charm at Hever Lakeside Theatre getting the punters in from their picnics around the lake this season
What you're asking for doesn't sound hard to figure out...I'd like to see that as well, but need assistance with the maths!
Besides, this is a feature I'd LOVE to see in SCS
I just need to know the Logarithmic range you want to cover (-120db TO +15dB?) AND the linear range you want to cover it (0 to 65536?) and I'll sort something out!
It might also be useful to know how you handle/process gain changes in SCS.
Cheers for now!
Couch Potatoe
BTW Many, many thanks for the "run-cue-at-specified-time-of-day" feature added recently, it worked like a charm at Hever Lakeside Theatre getting the punters in from their picnics around the lake this season
-
Mike Daniell
- Site Admin
- Posts: 3629
- Joined: Sun Jul 24, 2005 8:58 am
- Location: Brisbane, Queensland, Australia. TZ:GMT+10
- Contact:
Thanks, Couch Potatoe - have just replied to your Private Message - I previously tried to reply via email but it was returned as 'not delivered after 4 hours'.
The logarithmic range you suggest looks good, but probably only needs to go to +12dB. The linear range is 0 (minimum level) to 1000 (maximum level).
Not sure what you mean by handling gain changes unless it's related to the following: DirectSound (the Microsoft sound engine) does not have the concept of gain, only attenuation. So the unity level of DirectSound is full volume. All you can do is attenuate the level. So to implement a conventional mixer fader we need to set our own 'unity' level lower than full volume so we can provide headroom for 'gain'.
The logarithmic range you suggest looks good, but probably only needs to go to +12dB. The linear range is 0 (minimum level) to 1000 (maximum level).
Not sure what you mean by handling gain changes unless it's related to the following: DirectSound (the Microsoft sound engine) does not have the concept of gain, only attenuation. So the unity level of DirectSound is full volume. All you can do is attenuate the level. So to implement a conventional mixer fader we need to set our own 'unity' level lower than full volume so we can provide headroom for 'gain'.
I would also really like to see this feature implemented, and I am here to provide you with some math!
The formula for the dB scale is as follows:
dB = 10 * log(X), with X being the linear scaling factor. In your case, want the inverse of this formula:
X = 10^(dB / 10)
A good dB range to provide would be something like -50dB to +15dB (you could get by with a little less if you wanted). As you mentioned before, you don't want to actually increase the level of digital audio, since you are assuming the songs are mastered at a near-0db level, and raising the gain would cause them to clip.
So, what I propose is a slider for the user that is labeled (and allows us to type the in the dB values) between -50dB and +15dB. However, you translate those numbers inside your program to -65dB to 0dB so you avoid the clipping issue. Set the default volume level of a track to 0dB so we have some headroom built in for quiet tracks.
The math for this is as follows.
-50dB on the user scale (-65dB internal) is:
10^(-65/10) = 3.162 x 10^-7, or really really quiet :)
0dB on the user scale (-15dB internal) is:
10^(-15/10) = 0.0316
15dB on the user scale (0dB internal) is:
10^(0/10) = 1
Hopefully that all makes sense, and you can implement this new scaling for us. The current linear volume controls are not very responsive at all, and don't leave nearly enough headroom for some of my quiet tracks, so I would very much appreciate this feature.
The formula for the dB scale is as follows:
dB = 10 * log(X), with X being the linear scaling factor. In your case, want the inverse of this formula:
X = 10^(dB / 10)
A good dB range to provide would be something like -50dB to +15dB (you could get by with a little less if you wanted). As you mentioned before, you don't want to actually increase the level of digital audio, since you are assuming the songs are mastered at a near-0db level, and raising the gain would cause them to clip.
So, what I propose is a slider for the user that is labeled (and allows us to type the in the dB values) between -50dB and +15dB. However, you translate those numbers inside your program to -65dB to 0dB so you avoid the clipping issue. Set the default volume level of a track to 0dB so we have some headroom built in for quiet tracks.
The math for this is as follows.
-50dB on the user scale (-65dB internal) is:
10^(-65/10) = 3.162 x 10^-7, or really really quiet :)
0dB on the user scale (-15dB internal) is:
10^(-15/10) = 0.0316
15dB on the user scale (0dB internal) is:
10^(0/10) = 1
Hopefully that all makes sense, and you can implement this new scaling for us. The current linear volume controls are not very responsive at all, and don't leave nearly enough headroom for some of my quiet tracks, so I would very much appreciate this feature.
-
Mike Daniell
- Site Admin
- Posts: 3629
- Joined: Sun Jul 24, 2005 8:58 am
- Location: Brisbane, Queensland, Australia. TZ:GMT+10
- Contact:
levels, faders, scales etc
Hi Mike,
I would suggest *real* labeling of the fader travel. By this I mean the top at "0" and the bottom at "-∞"
The "taper" should of course approximate a nice hardware fader as discussed above. Better yet, this taper should be adjustable or have a couple of different presets available to the user.
We are dealing with digital audio. I think it is confusing and misleading to have your fader labeled in an offset manner. We must recognize that digital audio is different from analog, and we must learn to work with it properly. Labeling the faders in real values lets me know exactly where I am levels-wise. Every digital console I know and every serious audio software I know, labels all controls with real values. If you place a fader at "+10" you must be aware that you are boosting 10dB. Unity gain is zero.
The only way I see having "plus" values in the fader travel is if you're actually applying gain. This could be useful, but is not absolutley necessary IMO.
eduardo
I would suggest *real* labeling of the fader travel. By this I mean the top at "0" and the bottom at "-∞"
The "taper" should of course approximate a nice hardware fader as discussed above. Better yet, this taper should be adjustable or have a couple of different presets available to the user.
We are dealing with digital audio. I think it is confusing and misleading to have your fader labeled in an offset manner. We must recognize that digital audio is different from analog, and we must learn to work with it properly. Labeling the faders in real values lets me know exactly where I am levels-wise. Every digital console I know and every serious audio software I know, labels all controls with real values. If you place a fader at "+10" you must be aware that you are boosting 10dB. Unity gain is zero.
The only way I see having "plus" values in the fader travel is if you're actually applying gain. This could be useful, but is not absolutley necessary IMO.
eduardo
Re: levels, faders, scales etc
[quote="eduardo"]Hi Mike,
I would suggest *real* labeling of the fader travel. By this I mean the top at "0" and the bottom at "-∞"
The "taper" should of course approximate a nice hardware fader as discussed above. Better yet, this taper should be adjustable or have a couple of different presets available to the user.
We are dealing with digital audio. I think it is confusing and misleading to have your fader labeled in an offset manner. We must recognize that digital audio is different from analog, and we must learn to work with it properly. Labeling the faders in real values lets me know exactly where I am levels-wise. Every digital console I know and every serious audio software I know, labels all controls with real values. If you place a fader at "+10" you must be aware that you are boosting 10dB. Unity gain is zero.
The only way I see having "plus" values in the fader travel is if you're actually applying gain. This could be useful, but is not absolutley necessary IMO.
eduardo[/quote]
Thats fine by me, I don't really care what they are labeled since the output levels of most sound cards are arbitrary anyways. I just want the default level for a cue that I just added to be at least 10 or 15dB below the maximum. I had problems on a recent show with cues that were significantly quieter than the rest of the show, even with the cue's volume maxed, so I had to go through and lower the volume on every other cue.
I would suggest *real* labeling of the fader travel. By this I mean the top at "0" and the bottom at "-∞"
The "taper" should of course approximate a nice hardware fader as discussed above. Better yet, this taper should be adjustable or have a couple of different presets available to the user.
We are dealing with digital audio. I think it is confusing and misleading to have your fader labeled in an offset manner. We must recognize that digital audio is different from analog, and we must learn to work with it properly. Labeling the faders in real values lets me know exactly where I am levels-wise. Every digital console I know and every serious audio software I know, labels all controls with real values. If you place a fader at "+10" you must be aware that you are boosting 10dB. Unity gain is zero.
The only way I see having "plus" values in the fader travel is if you're actually applying gain. This could be useful, but is not absolutley necessary IMO.
eduardo[/quote]
Thats fine by me, I don't really care what they are labeled since the output levels of most sound cards are arbitrary anyways. I just want the default level for a cue that I just added to be at least 10 or 15dB below the maximum. I had problems on a recent show with cues that were significantly quieter than the rest of the show, even with the cue's volume maxed, so I had to go through and lower the volume on every other cue.