psychoacoustic effects??

hey everybody

I was wondering if there are any psychoacoustic effects around as LADSPA or LV2 plugins?? I recently heard about legendary effect processors like the Aphex Enhancer. I really dont know how these work but i was told if i would use them i would never want to make music without them.

Are there any plugins that can do what de “Enhancer” does??

Can anybody explaim me what they do?

In terms of what they do, this article is an oldie but a goodie:

http://www.soundonsound.com/sos/1997_articles/may97/exciters.html

I’m not aware of any standalone exciter/enhancer plugins for Linux, but this thread tells you how to roll your own in Ardour:

http://ardour.org/node/405

Oh, actually, there is an exciter plugin (DSSI) listed here:

http://holap.berlios.de/list.html

hey thx

now i got something more to fiddle around with.

Hi

The 2nd link you gave me tells me to first put a high pass filter on the signal and then a harmonic generator, wich sounds quit reasonable for achiving this effect.

Concerning this method i have a question about the GENERAL BEHAVIOR of plugins in Ardour

Does it matter the way i insert plugins in a row??
For example:

If i insert a EQ then a Compressor and then a Limiter would it be the same if i first insert the Limiter and then the Compressor?

I know how each of them works, but is the output I get a mix out of three different signals like

input + Eq
input + Compressor
input + Limiter

blended together or are they set one after another the way i insert the plugins??

When i work with my Rocktron Intellifex I have diagrams in the manual that let me see how a effect works.

Is there a way in Ardour I can change the way the effects a processed??
Like parallel or in a row??
Would it be better to duplicate the signal, just put one effect on every line and then mix them together?

Thinking about this makes me some kinda crazy.
Hope you can understand my way of thinking…

All processing in Ardour’s channel strips happens from the top down.

So Compressor then EQ is NOT the same as EQ then Compressor.

  Seablade

if you want to use parallel processing, use FX busses, each bus with one effect, and create sends from the track to these busses (one send per bus). You can mute the track’s main outs if you want, and only mix the busses.

Working with a DAW is completely new to me so ive got to figure out what works best for me. Knowing now that the order you insert plugins does matter i gonna keep an eye on that the next time i record something.

Maybe you can give me some tips for how to insert fx the right way.

I want to record a distroted guitar. Afterwards i want to put following effects to my guitar.

  1. Noise Gate or Hush reduction (got that on a FX processor in a rack, so i would have to get the signal out and in using Jack)
  2. EQ
  3. Reverb
  4. Compressor
  5. (maybe Limiter)

Would you change the order of effects?
Is a compressed signal better for reverb?

Don´t compress your reverb. Reverb is meant to be on a bus, so you can send different signals to the same reverb bus. So it would be:

  1. Noise Gate or Hush reduction
  2. EQ
  3. Compressor
  4. (I would´n use a limiter)
  5. a send to the reverb bus

the best way is the way it sounds best to you :slight_smile:
You’ve got to experiment.

In general, have in mind that a compressor is not frequency neutral, so if you apply some EQ, and then compress heavily, you will definitely affect the EQing.

A way that may be attractive (maybe not to all), is to run certain things in parallel and others in sequence. You could try to comp a track via a bus and still have the uncompressed track going to the master mix. So you can adjust the level of compressed and uncompressed signals. This is to keep a bit of the original dynamic in the image. Of course, if you use compression very heavily for “strange” effects, you may choose another way to set up your effect sequence.

Reverb is typically used in parallel via a bus. If you want coherence, you can send more than one track to a single reverb bus. It would make sense for a group of back-vocals. By “send”, I mean it in the ardour way: you fork the track signal at some point and send one of the forked audio path to somewhere while the other one continues through the fader or other destination.

For your hardware effect, you will need an insert point. Check your latency at that point as hardware latency is not possibly guessed by jack. See other discussion thread about hardware latency compensation.

If I were you, I would first focus on having the right raw sound, forgetting about FXs for a while. Before it reaches the DAW, your guitar sound is produced and captured somewhere. How is this going at this level ? (mic, preamp, AD converters) are you satisfied with what you get before applying effects ? it sounds like you will apply effects only once you have recorded the “raw signal”. So you must have that part (pre-effect) under control first.

Some good advice here. :slight_smile:

I’ve always suspected that good compression needs a decent analysis window size if you’re compressing based on RMS levels, which means latency (similar to the oft-mentioned look-ahead for peak level measurement). I’ve noticed analog compressors that seem to introduce a small delay, which has strengthened my conviction on this. I’ve had pretty good results with a DIY compressor patch implemented in Pure Data, with a tunable window/latency - no good for live stuff, but good for applying compression to a track once it’s recorded.

Noise gating distorted guitar is another area of interest. My favourite kind is…I guess you’d describe it as side-chain based, with the gate after the gain/distortion stages, but triggered by the signal from before the gain stages, where the difference between loud and quiet is still significant. I’m pretty sure the noise gate on my ADA MP-2 works this way too - it works extremely well, even at very high gain. I’ve also put together a gate patch in Pure Data based on the same principle (using a logistic function for ITB distortion, or sending the audio back out to a guitar amp). It’s a bit more complex to set up, and you have latency compensation to figure out, but it’s effective, and very flexible. (For example, you can get a nice volume swell effect if you turn the gate attack/turn-on time up to about a second.)

I would think that for distorted guitar, plain ol’ EQ would have much the same effect as an exciter/enhancer, as there’s plenty of distortion and extra harmonics already in the signal.

Thx everybody for the good and needful advice