FWIW - A good general rule for finding the right amount of DeEssing:

1.) Turn on the "Listen Mode"
2.) Turn the depth to full (clockwise)
3.) Turn the frequency to 12kHz (full clockwise)
4.) Preview a clip
5.) Start turning the frequency knob left (lower freq) until you hear some "S"s squeak thru.
6.) Keep turning left until you hear some voiced signal just start to squeeze thru.
7.) Back off to the right a little.
8.) Turn off listen mode.
9.) Adjust the depth to the desired supression.

Hope this helps.

~Bob

Fair enough...

I latched on to your comments...

"and still nothing when actually set on high freq's. It's like backwards." 
 
and… 

“For example I'll put it like 3khz but I'm hearing sibilant freq's there. And if I actually put it on the higher freq's that are usually the culprits I hear nothing.” 

...which is expected behavior.  IOW - lowering the frequency means you are "gathering up" more of the "S" to suppress, raising it is the opposite.  I want to make sure you understand this principle. 

Anyhow...Yeah, Vista and x64 both have their share of oddball behavior, that's for sure.  You might be on to something.  If there is some difference between x64/Vista and win32 versions, we need to hunt it down.  Both flavors should operate exactly the same, and in both cases, your listen mode should be able to be adjusted as to hear only “…..s……sss…..SS…..s…..shh…..” etc.  Again lowering (not raising) the frequency will make the extraction more pronounced. 

Do you have some examples? 

~Bob

Bob, there seems to be a real bug here on 64-bit systems that I came across. That was then published here by somsto is confirmed here on the analyzer. What doesn't really add up is what you stated here:
"...which is expected behavior.  IOW - lowering the frequency means you are "gathering up" more of the "S" to suppress, raising it is the opposite.  I want to make sure you understand this principle. "

DeEssers usually work on a band pass which appears to be the case for the VX-64 plugin as well. The "S" quality is something found in the higher bands such as ones described several times here throughout the thread like 7 or 8k, for example. Lowering the frequency on the plugin should indicate moving the bandpass to other target frequencies. I'm not sure what you are getting at by saying "gathering more of the "S". The band pass should normally be a narrow Q affecting only a small range or target frequency. The frequency targeted should then change and only affect what frequency it is then set at.

The bug here is, as our analyzers have shown, that the frequency setting does not match what the bandpass of the VX-64 is passing; they are as mentioned here, "reversed". Our analyzer would show a bandpass of 10k when the setting was set to 3k. That's clearly a bug. It was brought up to my attention that this is affecting 64-bit systems on Sonar 8.5.1 64-bit.  

Thanks

Tim you must be using Sonar 8.5.1 64 then, correct?

If not then this may be even stranger as it would seem like a random issue.

Hi Bob, it's not necessarily the filter implementation but that the end result is an isolating of a band or frequency. Technically you a passing a band to process. The lowpass and hipass filter pair at their respective crossovers are creating a bandpass. If its anything else that's going on there I'd be really surprised because of the ultimate goal of creating transparent De-essing would definitely be not to affect a group or not too large of a group of frequencies. Sure the algorithms can vary in how it handles the peak relating to it's attack, ratio, etc - I can understand that.

If you take a look at a signal passing through the VX-64 in listen mode, you will notice an isolated band. The direct result of a matched lowpass and hipass filter at a gentle crossover. Essentially creating a bandpass. The bug here is that when our crossover freq is for example set at 3 or 4 K, our spectrum analyzer shows us a bandpass of like 10 K being passed and processed. This just can't be right and has to be a bug.

My latest question is to find the common thread of a system setup that will replicate this issue. What it's pointing to at the moment is that running the latest Sonar on Vista 64 harbors this bug.

Tim, thanks for the info...

So now it seems random. We have some some 32bit guys with no problem but you for example with the issue. And my system for example using Sonar 64 and Vista 64 with the issue.