Pitch - Need Help

I'm talking about single note solos, with each voice of the shifter at 100% wet mix going through its own amp. No intermodulation to consider. The higher the gain, the better the sound when the shifter is placed first, at least in every setup I've ever had. Javajunkie's mp3 above indicates the opposite is true in that particular instance with the Axe. Not sure why that is. That particular amp or drive doesn not seem to like the higher harmonics in the up-shifted signal. I might try a lpf between the shifter and the amp. I like the Axe's flexibility.
 
Ok, so I just got an Ultra. As I suspected -- tracking is about equal to Eventide and better than anything else. Quality of the voices is not as rich as Eventide, but sounds more like a guitar and less like a synth (sounds a little bit like a better tracking Boss harmonizer to me).

I see what you mean about no intermodulation -- I usually do the original voice and two harmony voices on each channel (L/R) so it works best for me if it's after the amp.
 
bmancini42 said:
Ok, so I just got an Ultra. As I suspected -- tracking is about equal to Eventide and better than anything else. Quality of the voices is not as rich as Eventide, but sounds more like a guitar and less like a synth (sounds a little bit like a better tracking Boss harmonizer to me).

I see what you mean about no intermodulation -- I usually do the original voice and two harmony voices on each channel (L/R) so it works best for me if it's after the amp.
Click to expand...

What firmware are you on?


Tracking is actually better than the eclipse. I was just comparing the two a few minutes ago. The Eclipse handles polyphony much better than the eclipse. When comparing the 2 with the new algorithm, I liked the new algorithm on the Axe-fx a bit better than the eclipse. I liked the eclipse a bit better with the bell tree pitch stuff (pitch w/ feedback). The eclipse just has a certain sweetness to it.
 
javajunkie said:
here is an A major a 3rd up 100% first before the amp, then after.

http://javajunkiemusic.com/Audio/pitch3.mp3
Click to expand...

I downloaded the editor and found that the "Lonely Heart Solo" patch puts the pitch shifter before an amp and runs the unshifted signal through its own parallel amp. Don't know how it sounds since I don't own an Axe, but at least I know this is a valid approach on this unit.
 
steadystate said:
javajunkie said:
here is an A major a 3rd up 100% first before the amp, then after.

http://javajunkiemusic.com/Audio/pitch3.mp3
Click to expand...

I downloaded the editor and found that the "Lonely Heart Solo" patch puts the pitch shifter before an amp and runs the unshifted signal through its own parallel amp. Don't know how it sounds since I don't own an Axe, but at least I know this is a valid approach on this unit.
Click to expand...

Of course it is a valid approach. Running through its own amp has lots of advantages (before or after the amp). At times it is more appropriate to put it in front, other times after, sometimes it just does really matter that much. It comes down to using your ears and experimenting. The loney heart preset sounds good.

You can hear it here:

http://www.youtube.com/watch?v=zXCGfJqc ... re=related

I think he demos that preset.
edit: it is at 4:31
 
steadystate said:
I'm talking about single note solos, with each voice of the shifter at 100% wet mix going through its own amp. No intermodulation to consider. The higher the gain, the better the sound when the shifter is placed first, at least in every setup I've ever had. Javajunkie's mp3 above indicates the opposite is true in that particular instance with the Axe. Not sure why that is. That particular amp or drive doesn not seem to like the higher harmonics in the up-shifted signal. I might try a lpf between the shifter and the amp. I like the Axe's flexibility.
Click to expand...

After extensive experimentation creating hi-gain harmonies for leads (solo notes with a lot of sustain and distortion) I've concluded that using the pitch shifter to create the harmonies first and then running each harmony (and the unshifted note) through their own dedicated distortion units creates a far superior effect than placing the shifter after the distortion. It requires 3 paths (assuming 2 harmonies), one for each note, with each containing either an amp or drive, and a cab or LPF to simulate a cab if the CPU runs short, but to me the sound is far superior. A compromise would be to distort them separately and share the same cab. But placing the distortion before the shifter makes the harmonies lose their "independent" feel to me. It sounds more like a pitch shifter than 3 people each playing different parts with different amps and cabinets. If your method works for you more power (and less CPU power) to you!

I'm frankly at a loss to explain the example that was posted by Javajunkie of the difference in quality between pre and post distortion shifting (it sounds like the dry signal might not have been completely killed - I've NEVER heard artifacts like that when distorting a monophonic pitch shifted signal on the AXE and the results have always been better with the shifter before the distortion - FOR SINGLE NOTES). I agree with steadystate about how to best accomplish creating harmonies for soloing. But again, whatever works for you. The quality of the shifter is good enough such that the difference between the two methods isn't as big as it is when using cheaper units (IPS-33B for example).
 
Yes, it's always better to have individual paths for the harmonies. If you put the harmonizer before distortion you'll get intermodulation products. After the distortion and you'll harmonize the overtones. After sounds better than before but separate processing paths sound best.

There's a couple (or more, can't remember) factory presets that illustrate this. "Lonely Heart Solo" is one.
 
sampleaccurate said:
steadystate said:
javajunkie said:
For the most part, the Axe-fx shifter worker best between the amp and cab (trust me I've extensively tested just about every combination). This is because the axe-fx pitch detection (unless set to local mono or local poly) gets its signal directly from the input for any effects processing. It is slightly different than your setup because of this.

I agree voice level and interval CC control would be really nice. I doubt you'll ever see 3 drive blocks though. you can control the interval through sysex commands if your foot controller is able to send longer custom strings. I'll check on the local custom scales. I've stuck to the global scales when using custom.
Click to expand...


I've never achieved the best results placing a pitch shifter after a preamp. When up-shifted, the signal sounds thin and mosquito-like. When shifted down, all the highs generated by the preamp are gone and I have to put the signal through an exciter (basically another distortion unit). All non-formant-preserving pitch shifters change the timbre of the signal when you shift. The greater the shift and the more harmonically complicated the signal, the worse the timbre changes. I'm trying to wrap my brain around why the Axe shifter would be different from any other.
Click to expand...

Ditto! Well put. I've had exactly the same issues.

Javajunkie: As far as the pitch detector being on the input, what does that have to do with the location of the pitch shifter module if the global input detector is used? I'm sure steadystate is using the global pitch detector at the input. Why would the location of the pitch shifter module matter as far as being able to track pitch??? And how is it any different than what steadystate says he's doing now? It sounds identical to me. The pitch module is receiving the pitch information it needs from the pitch detector at the input regardless of where the module is placed. No?

And what's this about doubting if we'll ever see three drive blocks? Steadystate says he wanted to use one amp block to distort the main lead, and two drive blocks to distort the two harmonies (separately). That's perfectly doable. CPU might be tight, but there's room for a shifter, an amp, two drives, a cab, and a couple of LPFs and EQs as a "low CPU substitute" for cabs on the harmonies - or sum all three signals post distortion and use the same cab.

I don't think you guys are communicating clearly (for all I know neither am I) but I understand everything steadystate said and I agree 100%.
Click to expand...

I don't think steadystate owns the axe-fx he was talking about past experiences w/ the IPS-33B (at least at the time he posted). He later explained that he had experimented with sidechaining the input agreed with me about the track. He then elaborated he was referring to timbre not tracking. We were not talking about pitch tracking but tone. Cliffs post above yours explains why. However, if it is set to local it makes a huge difference if you are using a large amounts of distortion. Location of the pitch before and after the amp makes a huge difference in tone (again nothing to do with tracking).

My comment about the thee drive blocks was not refering to the point you brought up but his comment later in that post,
CC control of the pitch shifter voice levels would be great, as would CC control of the voice intervals (or more CC selectable global scales). The ability to use three or more drive blocks would save a lot of CPU in this setup as well, since the amp block could be replaced with a drive. The possibility of replacing my large, heavy, expensive, and marginally reliable (but great-sounding) rack with a single two-space unit has me very excited. Before I purchase an AXE-FX, I need to verify it can do what I need it to do.
Click to expand...

I have absolutely no disagreement about using individual paths for harmonies. That is where we got off from each other. He was referring to each harmony having a line. My references were to the orginal posters issue (which now that I look back, I misunderstood). Regardless, the pitchshift has been vastly improved in 8.06. I imagine the person originally having the issue is much happier now.
 
Just to give you a rough idea steadystate, the patch I'm running now uses the following modules with a CPU load of less than 92%:

(1) AMP
(2) DRV
(2) PIT (intelligent harmony mode, but I'm only using one, the other I threw in there just because I could to load the CPU)
(1) CAB
(2) FIL (used in LPF mode along with the EQs to simulate cabs for the harmonies)
(1) DLY (stereo)
(2) GEQ
(1) PEQ
(1) REV
(2) CHO
(1) CPR

Take out your credit card, part with the $2K and sell your rack already steadystate! You'll have a smaller rack, no compromises in tone, and a lot of money in the bank from the eBay sale!

Stephen Cole
 
No sooner do I back up a well held belief and make an argument for it backed by an extensive technical justification when, abracadabra, I found a way to make the post distortion shifter work almost as well as the pre-distortion method I so staunchly defended and advocated in my previous post.

This box never ceases to amaze me. It's just amazing. What other word is there? I'm so used to the artifacts of pitch shifters and now they're almost gone! It's like I need to re-think EVERYTHING and throw away all preconceptions about what a processor can do.

I'm just thankful somebody finally figured out not only a way to make great digital guitar distortion, but threw in truly world class effects to top it off.

I feel like a kid discovering stomp boxes for the first time.

Stephen Cole
 
Cliff (or whoever posted from Fractal) was succinct and accurate in the comment about the relative disadvantages of each method and what type of signal degradation you can choose from.
Click to expand...


Right, that was Cliff

The answer he got was “It takes some of the edge off the upper harmonics which tend to be not as stable”.
Click to expand...
That answer was not from me.

I don't think Cliff was talking about polyphonic playing but just using the pitch block when you don't have individual paths (which winds up as necessity for me a lot). You are actually hitting the amp w/ a polyphonic signal this way. Ypu have the dry signal and the one or 2 voices. I use a lot of dual amp configurations and a lot of my patches are already on the edge because of 2 amps and delay couple w/ a lot of controllers. Like I said I have no argument that as Cliff stated individual path harmonies is better. You can try this for yourself just place a pitch block in front of an amp in a preset set your harmonies and mix. Play it, then move it after the cab block. I prefer after. That is what I was speaking of for the most part.

But like you are saying split the signal to 2 amps, pitch shift one w/ mix 100%; before may sound better. Although the pitch shift clip at 100% I posted earlier appears to sound better to me after the amp (it was pretty clean though).http://javajunkiemusic.com/Audio/pitch3.mp3 . It all depends on what you need to do.

I think we did a test of all this stuff a few months back.

A note about the Poly local mode. Try it. Even if you don't do anything polyphonicly, I found it can sound more natural but has a bit more latency (it uses a different algorithm).


EDIT:
I was experimenting with the pitch again this morning briefly. The thing I found was with pitch mix at 100% it seemed to sound better only clean tones after. On heavier stuff, it was a bit better before (which confirms your experiences). Regardless, the pitch shifting is much better after the 8.06 update.
 
A quick tip about your comment on using the PEQ as a cab subsitiute. It may not be necessary for you. You can have up to 4 cabs in a patch. If you place both cab blocks in stereo mode and use proper panning you can get each voice with its own cab with not a lot of extra cpu. Also you could (if you were keeping things mono) just mix the signals into a signal cab.
 
javajunkie said:
Cliff (or whoever posted from Fractal) was succinct and accurate in the comment about the relative disadvantages of each method and what type of signal degradation you can choose from.
Click to expand...


Right, that was Cliff

[quote:3ru1you2]The answer he got was “It takes some of the edge off the upper harmonics which tend to be not as stable”.
Click to expand...
That answer was not from me.

I don't think Cliff was talking about polyphonic playing but just using the pitch block when you don't have individual paths (which winds up as necessity for me a lot). You are actually hitting the amp w/ a polyphonic signal this way. Ypu have the dry signal and the one or 2 voices. I use a lot of dual amp configurations and a lot of my patches are already on the edge because of 2 amps and delay couple w/ a lot of controllers. Like I said I have no argument that as Cliff stated individual path harmonies is better. You can try this for yourself just place a pitch block in front of an amp in a preset set your harmonies and mix. Play it, then move it after the cab block. I prefer after. That is what I was speaking of for the most part.

But like you are saying split the signal to 2 amps, pitch shift one w/ mix 100%; before may sound better. Although the pitch shift clip at 100% I posted earlier appears to sound better to me after the amp (it was pretty clean though).http://javajunkiemusic.com/Audio/pitch3.mp3 . It all depends on what you need to do.

I think we did a test of all this stuff a few months back.

A note about the Poly local mode. Try it. Even if you don't do anything polyphonicly, I found it can sound more natural but has a bit more latency (it uses a different algorithm).[/quote:3ru1you2]

I will try it. I'm sorry I didn't believe you when you said "believe me". It looks like I've got some catching up to do. I spent hundreds of hours experimenting with pitch shifters on other devices. Now I need to throw a lot of that out the window and consider new possibilities.

I have a theory on why the pitch after distortion sounds so good, but I'm not going to go there. If it does, it does. I'll just accept the good news and move on.

Stephen Cole
 
I edited that post a bit, you may want to reread it. I'm going to do some more experimenting this evening.
 
You must log in or register to reply here.
Back
Top Bottom