another marque tool bug this time lasso

ok if its not adobe then please explain why this effect is not apparent on the standard marque tool and only on the lasso?

i would guess that the blending/smoothing is not working as effectively(if at all)on the lasso tool which needs to be fixed.

kb

ok if its not adobe then please explain why this effect is not apparent on the standard marque tool and only on the lasso?

The requirements for a filter that can take out a fixed frequency range over a period of time are rather more relaxed than the lasso filter, which has to have a much steeper response. The steeper the response of a filter, the more ripple you get around the turnover frequency. If you implement a filter that has no ripple at the top or bottom of its passband, then repeatedly applying it will not increase the level of any of the skirt aretefacts, and that's what you can do with the 'square' marquee tool.

OTOH, the lasso filter has to have steeper sides, which will give rise to cumulative ripple effects - you are asking it to take out one frequency one moment, and in very short order, potentially altering this, because the lasso can go wherever you want it to. In order to achieve this transformation, you need a filter that can achieve a fast transition time - which also increases the chance of skirt ringing significantly. To do this, you need to implement a second order filter, which is, in this case,  an IIR specified one.

i would guess that the blending/smoothing is not working as effectively(if at all)on the lasso tool which needs to be fixed.

Sorry, wrong again. As I said, it's the Laws of Physics at work here - not Adobe.

I will give you the kudos for spotting the difference in performance, though, if it will make you feel a little better...

i dont need your kudos to feel better, i expected something that performs better.

kb

i dont need your kudos to feel better, i expected something that performs better.

Then your expectations were completely unreasonable. It obeys the Laws of Physics, but not neccessarily the ones you understand. If you don't want a crumb of comfort, that's up to you - but it won't alter a thing.

Can you describe in a little more detail what you actually were trying to do please. I understand what Steve is saying. I don't understand, but would like to, why you would want to apply the same amplitude cut of 3dB 40 times to one poor bit of audio. Why not just reduce the level by 120dB?

i never suggested you should apply an effect 40 times i gave 20 times as an example to exagerate the effect as sometimes the effect im talking of is not always noticable after one time, but it does happen.

im just trying to reduce unwanted signal in a cetain section, where in the past i have used the std marque tool from 1.5 to great effect. but using the lasso tool which i thought would be quicker and more precise is often a disaster and i do not intend to use it in future unless it can be made to function better.

kb

But that is the issue.  I've had lovely results with lasso-marquee, but I set the amplitude reduction sufficiently (or, just hit "delete") so that the transform occurs only once.  I've never noticed the ripple/ringing, so it seems like applying the effect more than once or twice is a sort of abuse.

as i have already mentioned i get this problem even when just using the tool only once under certain conditions. so it is not a result of abuse at all.

kb[/i]

OK, I've done an invert-paste test with a wierd-shaped lasso area of significant size and complexity at 44.1/32bit.  I used Amplitude set to -3dB and applied it 10 times, then changed it to +3dB and appied it 10 times.  After 20 transforms, I could not tell where the spectral mask was in spectral view, nor could I hear anything odd in the music.  Then I invert-pasted the original over the result.  My findings: an outline of the spectral mask was indeed faintly visible, with no artifacting whatsoever in the middle of the area, and of course nothing outside of it.

Moral of the story: don't abuse your audio.

as i have already mentioned i get this problem even when just using the tool only once under certain conditions. so it is not a result of abuse at all.

kb

Try reinstalling the software.  We can't reproduce this.

Try reinstalling the software. We can't reproduce this.

Or if it's possible to upload the file and let someone work at what you're doing?

Mark

it happens on all files i have tried using the method described.

kb.

Well, is everything else OK. Do you like the program in general. I know you've never been to "general" but is it all right were you are

Mark

i loved audition 1.5 and i still do, the marque tool for me is essential as a professional audio restorer, so its hard to general. i dont like the new layout and gui on the new audition 2 as i find it is an unessary change in favour of looks rather than ease of use. so in general i feel audition 2 is a step back from 1.5 though i so much wanted to enjoy it and embace it as i have done all other previous versions.

in general i would say if you are not into multitrack work then i would recomend people stick with 1.5 until adobe work out the bugs i have found...i know the bugs i am talking of are not excepted by some others but believe me they are real as i use the std marque tool everyday for work.

the situation is so critical to my work that i am back to 1.5 for now and will not return to audition 2 or further until i am convinced it is bug free.

the other part of the marque tool the marque lasso while new is next to useless as a tool for me creating more problems than it fixes.

if anyone wishes me to send them before and after sound files and images pm me with your email and ill be glad to do so.

k.b.