Rational Acoustics



Dr. J
December 2nd, 2011, 03:44 PM
Here is my situation:

I have a digital console (Presonus SL) where I measured the High Pass to be a 6dB Slope. This is pretty Lamo to say the least. I also have Low Shelf & High shelf filters that seem to resemble close to a BW18 type of slope.

There is even a Hi-Q option on the console BUT it measures out to be a 2/3 octave Q. Not very specific but good for tonal shaping or whatever.

I have to use a combination of the Low Shelf and the High pass just to get the low frequency information out of the mix. It is crazy how high -- the high pass has to be set & when I measure it using Smaart -- the information never really leaves the screen. Still hovering around trying to make a comeback.

For fun -- I hooked up a Driverack thru an Aux and routed all vocals mics thru this. Measuring again with Smaart -- I was able to set the High Pass at 150 Hz and not have ANY remnant of any low frequency information visible or audible. Plus there is a set of Parametrics & access to a 28 band graphic. There pretty much isn't anything I can't do to the vocals with this arrangement. I really like this precise set of tools.

I have the ability to feed in Plug Ins into my console as well but I know there can be latency involved.

So the question is: Do plug ins have just as much Latency as digital outboard gear or is it about the same..... in your experience?

The DR measures 1.50ms from the input to the output BUT since I have it in an Aux routed to a vocal subgroup I think it doubles in ms. I cannot hear any latency with the Driverack hooked to the system this way.

I haven't used any of the plug in EQ's yet but do they cause latency as well?

PrestonSoper
December 2nd, 2011, 08:24 PM
I have to use a combination of the Low Shelf and the High pass just to get the low frequency information out of the mix. It is crazy how high -- the high pass has to be set & when I measure it using Smaart -- the information never really leaves the screen. Still hovering around trying to make a comeback.

Tell us a little more about your set up to compare the input channel's hpf & lpf to your reference input.
At which point in the chain/routing are you making the comparison?



So the question is: Do plug ins have just as much Latency as digital outboard gear or is it about the same..... in your experience?

Plug ins can range from very minimal (less than a ms) to several ms.
The more plug ins you stack on any given input or output will increase the latency value.
Most outboard DSP will average 1 to 3 ms of latency with a full range signal.
Conventional filters & fir filters can also play a part of creating latency in the dut.



The DR measures 1.50ms from the input to the output BUT since I have it in an Aux routed to a vocal subgroup I think it doubles in ms. I cannot hear any latency with the Driverack hooked to the system this way.

A comparison of the L/R bus and a discrete aux vocal subgroup (not the one routed to L/R), should give you an answer to any arrival discrepancies.



I haven't used any of the plug in EQ's yet but do they cause latency as well?

No more or less than previously stated.
It depends on the plug in & quantity in series on any given input or output.

Dr. J
December 3rd, 2011, 01:21 AM
Hey Preston -- I will give it a try. I am splitting my pink noise signal right off the cd deck so it goes to my interface input 1 & into the mixer channel strip input....... I then take the signal coming out of the main on the mixer and run it into the interface Input 2. Run delay locator and I get 1.81ms travel time from input to the output of the mixer.

I then do it again the same way BUT this time I go into the input on the Driverack and from the "out" back to the interface. NO mixer hooked up on this one. I get 1.50 ms travel time.

When I combine the two -- I get 3.31ms

When a subgroup is assigned -- another 1.81ms is added onto the trip. Something internal I assume. Thus, my concern about adding latency on a vocal in comparison to the band.

I also measured the travel time from the input on a channel strip to the aux and it is exactly the same as coming out of the main.

NOW -- When I run the pink noise thru the mixer alone -- I can see my EQ changes on the channel strip using Smaart. I have high pass option on every input so I set it at say 150Hz and I see what it is actually doing. I do it again for a Low shelf filter and see how it works. Just getting a visual of the changes I make so I get used to hearing and seeing what I am doing.

So for fun I hooked the Driverack up to an Aux into the DR and out of the DR into an Auxiliary Input on my console PLUS routed to a vocal sub group. I know this seems weird BUT I have no extra outboard EQ's with high pass or anything narrower than a 2/3 octave Q. The reason why I tried this is because a high pass slope of 6dB is just lame and on a vocal -- it is hard to remove what you want without removing too much. So for fun - the DR has a tool set for me to try.

I set up a BW18 High pass. Right off the bat an improvement. A series of parametrics to deal with feedback frequencies just in the vocal mics only. Plus access to a 28 band graphic. This helps me to tame down hot vocal mics which make everything else sound harsh because it finds it way into the vocal mics & I am able to take out feedback frequencies with great specificity, plus a more than sufficient high pass.....etc.

I could use a plugin fed into the mixer via firewire and probably achieve the same thing but was concerned about travel times between a Digital DSP device versus a plugin thru firewire.....

My mixer has a 4 band Semi-parametric plus a high pass BUT there is Nothing on it that makes it anymore specific than a graphic EXCEPT I do get to select the frequency. The "Hi-Q" option is a 2/3rd octave Q of 2.14 which I do not get to control.

Don't know if that explains everything exactly.

Dr. J
December 3rd, 2011, 09:54 PM
Hi Fellas -- I am not on smaart 7 yet but what are the advantages of the different access points that are available? Let me give an example so I am clear:

You can split the signal straight off the CD deck with Pink Noise and then run it into the channel strip input with the other split going into the interface input. This feeds the interface with unaltered pink noise. Anything useful here?

Next I can split the Main out on the console before a system controller and run it into an input. This is where I would normally do a measurement IF I was measuring speakers. The other split off the main goes to an input on the interface. Measurement mic directly into the an input on the interface.

What about the Outs on a crossover. I measured these before and it all I could tell from it was for Level setting. Once I eq'd a driver -- these were all over the place and had to be readjusted.

I am still in learning mode (always) & I know with smaart 7 there is a ton of flexibility here.

PrestonSoper
December 4th, 2011, 07:53 PM
Here is a link to the waves latency chart.

http://www.wavesupport.net/content.aspx?id=2213

Rasmus Rosenberg
April 28th, 2012, 09:22 AM
Hey all,
Bumping this thread as Im curios if you have some inputs. From the Waves list, as I read it, the higher sample rate the more latency the plug in creates, right? But the higher sample rate the faster the conversion (I/O) in the sound card is. Aka if you only measure the conversion lets say 48 khz has 1.44ms (In/out) so 2.88ms in total, and at 96 khz that will roughly be 0,72ms (I/O) so roughly 1.4ms total. But the latency the plug in generate 48 khz vs 96 kHz will not in all cases scale epual . Also the processor use will go way up running 96 kHz vs 48 kzh so depending on what plug in, how many you want to run and processor cap. You can optimize the latency vs processor cap. So I guess we would like to run the I/O at 192 kzh but the plug in at 44 kHz, as that would give the least latency, but I'm not sure you can or how? any tips?
mvh
Rasmus