How do you fix Polar latency issue guys?

[Gulale]

after I put Polar on my guitar, I'm having some amount of latency issue going on. How did you get over this latency issue guys? Thanks in advance for your valuable idea. 

[Benedict]

HI

Pitch Shifters will always have latency as they need time to think about what they are doing. So you either blend the effect with the dry signal and live with it or record the dry audio and then add it back with the Dry Mix section and use the included Dry Delay to line up with the shifted signal. Then you use ReGroove to pull that track backwards till it lines up with the mix. Quite easy once you have done it once.

[Gulale]

Thanks Benedict useful Info.

[hydlide]

i often compensate the effect using DDL (digital delay) in milliseconds with the settings feedback 0, dry / wet to wet. It depends on the algorithm on what the delay time will be.

Its not ideal when doing a "live" setup. But its a descent work around in order to have all the tracks sync up with the rest using one device that takes care of the other tracks.
The delay time is covered in the manual of Polar. I thought it was 48ms till 96ms

[stratatonic]

Its not ideal when doing a "live" setup.

A real shame about that. You cannot lay down live random guitar/keyboard shifting unless it is really slow stuff. :frown:   I suppose Propellerhead is more concerned with the cpu hit, but there should be an option somewhere where you sacrifice some computer power for a faster response time. I can't even do simple proper octave shifting in real time...it's not the same FX-ing everything after tracks have been laid down.

[hydlide]

Its not ideal when doing a "live" setup.

A real shame about that. You cannot lay down live random guitar/keyboard shifting unless it is

really slow

stuff. :frown:   I suppose Propellerhead is more concerned with the cpu hit, but there should be an option somewhere where you sacrifice some computer power for a faster response time. I can't even do simple proper octave shifting in real time...it's not the same FX-ing everything after tracks have been laid down.

Actually, its not really the CPU Hit causing the issue. Its more related to the thing Benedict mentioned:

Pitch Shifters will always have latency as they need time to think about what they are doing

since a pitch shifter in this context (eg: audio), the need to read-ahead before they can translate the pitch shift process. They buffer the audio first, then it gets shifted and so on.

I hope that makes sense.
However, I may add, if you are used to laying down a live performance with lets say 100ms (takes a lot of practice), you'll get used to "performing upfront". Eg, I did some stuff in the past where the audio was already processing a delay with 150ms before it got back to me... However at some point you can adapt to this delay even.

And the cool thing with live performances... you may make mistakes, that is what makes it ... a live performance (and not some quantized note driven burned CD thing... )

[stratatonic]

Actually, its not really the CPU Hit causing the issue. Its more related to the thing Benedict mentioned:

Pitch Shifters will always have latency as they need time to think about what they are doing

since a pitch shifter in this context (eg: audio), the need to read-ahead before they can translate the pitch shift process. They buffer the audio first, then it gets shifted and so on.

Well, I understand about the latency and the audio buffer look ahead, but isn't that also dependent on the amount of computer resources put to that task? (More of that = lower latency?)

[joeyluck]

I've had the same question before.  

It's not that I care to compensate for latency after recording and applying the effect... It's that I want to use Polar live.

I have hardware from 1999 that shifts pitch without any noticeable latency.
Surely it's in the way the pitch shifting is handled?

I also don't notice as much latency when shifting pitch via Neptune.
And I do have it set to 'fast.' Does it achieve the pitch shift in a different way?
Is it perhaps the nature of the RE format?

[Namahs Amrak]

The latency is a major disappointment for me also. I asked the question in the PUF, and the replies were similar to Benedict's post in this thread. However I am sure that I have heard pitch shifting pedals played live that do not suffer such a latency. Can't there be a method in which the input is processed by a filter that simply alters the output signal? I would love to hear from anyone who has used a guitar pedal that does not exhibit Polar's behaviour (such as the Boss PS-3 or PS-5).

[normen]

They all have at least a bit of latency, even the "octaver" type oldschool foot pedals. You get less latency (but also lower quality) if you set polar to "classic" mode. It can't compensate your audio systems inherent latency however so thats always added on top if you use it live.

Btw, see my signature for a possible solution

[Namahs Amrak]

Btw, see my signature for a possible solution

Not for 'live' guitar, though, because playing ahead of a beat is a cumbsersome and emotionless task.

[normen]

Not for 'live' guitar, though, because playing ahead of a beat is a cumbsersome and emotionless task.

Many great guitarists rather see it as an interesting challenge though For example Steve Morse still likes his 1989 Shadow guitar to MIDI system cause he got so used to the amount of latency it does (and how it forces him to pick every note accurately).

[Namahs Amrak]

Many great guitarists rather see it as an interesting challenge though For example Steve Morse still likes his 1989 Shadow guitar to MIDI system cause he got so used to the amount of latency it does (and how it forces him to pick every note accurately).

Well, sure... I seem to recall reading an interview with John McLaughlin (a player I hold in very high regard) in that he would also be playing 'ahead' when using the early Roland MIDI equipment, but I'm not in the same league as the mighty Mahavishnu. 

Nevermind, Polar is still an OK RE, as a production FX unit. Dynamic use of the two pitch-shift knobs makes for some unique sounds - although to raise another small criticism, I would have preferred the steps were fluid (such as the pitch bend wheel for MIDI instruments in the rack) rather than locked in to semitone increments.

[normen]

Well, sure... I seem to recall reading an interview with John McLaughlin (a player I hold in very high regard) in that he would also be playing 'ahead' when using the early Roland MIDI equipment, but I'm not in the same league as the mighty Mahavishnu.

Well it still is an interesting thing though. Actually listening to what you do instead of mixing up what you do and what you appear to hear. Not only because you actually start to form your sound instead of just accidentally having one but also for some technical reasons. For example some guitarists I know seem to "just do", when I see them recording into their computers with latencies they end up with tracks that are out of groove because they don't listen..

All DAWs (including Reason) work the following way: If you have software monitoring enabled they put the recorded sound on the track *where it was heard* relative to the backing tracks during the recording (that is with the computers latency), if you disable software monitoring they put it *where it was played*. So if you always play "on beat" in "real time" you end up with tracks that are shifted when you have software monitoring on.. So DAWs expect you to listen and many people just don't

[Namahs Amrak]

Well, sure... I seem to recall reading an interview with John McLaughlin (a player I hold in very high regard) in that he would also be playing 'ahead' when using the early Roland MIDI equipment, but I'm not in the same league as the mighty Mahavishnu.

Well it still is an interesting thing though. Actually listening to what you do instead of mixing up what you do and what you appear to hear. Not only because you actually start to form your sound instead of just accidentally having one but also for some technical reasons. For example some guitarists I know seem to "just do", when I see them recording into their computers with latencies they end up with tracks that are out of groove because they don't listen..

I completely understand where you are speaking from on this point. Many guitarists (speaking about 'home' guys & girls) end up owning DAWS because, let's face it, it can be very cheap for a hobby player to lay down a couple of guitar tracks and a bass against some drum loops. It can all be done for $50 or so (that is, with Reaper, plus a wide range of VST amp sims that are free).

And you're right, many just don't seem to hear that they are out of sync with the overall 'groove' of the song. It's something I hear in a lot of pure guitar hobbyists, but it's quite rare for electronic musicians to be so out of sync. Of course in many ways that's because most virtual instrument tracks can be quantized at the push of a button. For the instruments with latency (and of which your well regarded RE can compensate for), there are fixes, if the 'producer' is even aware of the problem.

Here lies the problem. Electronic musicians seem more attuned to the latency issues, because they are spending countless hours poring over a track. 'Pure' guitar players seem interested only in laying down the tune, with little regard to post production outcomes. To refer back to your earlier comment about Steve Morse, and in my case of Mahavishnu J McL, these guys are very, very proficient players. That's why they are career musicians held in high regard. They can pull off playing 'early' against a beat, because they are pro's !

I remember when I first stuck my toe in to the icy waters of digital recording, my latency issues were TERRIBLE... speaking purely of electric guitar into a reputable interface (a lexicon) but running a Pentium P4 computer. No matter what I tried, the latency was very off-putting, so my cumbersome work-around was to avoid monitoring my output, then nudging the finished guitar takes to the beat in production. It sucked so much ! Although the positive outcome was that I discovered MIDI-based music, and Reason 4.0 (and to this day, I actually feel that some of my best compositions were in the R4 environment).

The big issue, in the way I see it, is that whether in the case of your Steve Morse example, or that of mine with not listening to my playing while recording due to latency, is that the 'feel' can often be lost, because the player is never playing WITH 'the band' (the band being the virtual backing).

Some quite proficient guitarists even extend upon the importance of 'feel', and claim that all VST amp sims lack that 'real' feeling of a cabinet and tube head. Many go on to say that hardware sims such as the AVID Eleven rack device almost perfectly emulate the 'feel' of a real rig set up, and they are highly favoured. Why is that ? Perhaps Eleven has dialled in the specific latency between the player's ears and a real rig, whereas the VST sims are too pure. In this instance, it might me that in a (human perception) latency-free digital format isn't imperfect enough.

All DAWs (including Reason) work the following way: If you have software monitoring enabled they put the recorded sound on the track *where it was heard* relative to the backing tracks during the recording (that is with the computers latency), if you disable software monitoring they put it *where it was played*. So if you indiscriminately play "on beat" in "real time" you end up with tracks that are shifted when you have software monitoring on.. So DAWs expect you to listen and many people just don't

Wow...... I will have to think a little more about what you're saying here, and set up some tests to see/hear this for myself. Or.. perhaps you're approaching this in the way I described with my Pentium P4 guitar experiences ? Yes / No ?

[-008']

Here lies the problem. Electronic musicians seem more attuned to the latency issues, because they are spending countless hours poring over a track. 'Pure' guitar players seem interested only in laying down the tune, with little regard to post production outcomes.

I doubt it's because of the time poured over a track. You had pointed it out just before.... Electronic music is mostly all quantized, so they will notice un-quantized and/or latent tracks easily and quickly. OTOH, If the guitar-based track is lazy and/or grooving with every other element also being un-quantized and or/latent in its own way.... Things might just "feel off" at some point...or even not be noticeable at all.

Then again I could be latent and off with that line of thinking. Was just an observation, nothing actualy helpful to the topic. Carry on!

[joeyluck]

I don't think anyone is arguing whether or not latency will exist with pitch shifting. Its just the Polar seems to have a much higher level of latency (noticeable latency) than someone experienced with pitch shifting effects would expect (especially at the quality it produces). Its not the most sophisticated sounding pitch shifting. So its just confusing why Polar's latency is so great...

[normen]

The big issue, in the way I see it, is that whether in the case of your Steve Morse example, or that of mine with not listening to my playing while recording due to latency, is that the 'feel' can often be lost, because the player is never playing WITH 'the band' (the band being the virtual backing).

Some quite proficient guitarists even extend upon the importance of 'feel', and claim that all VST amp sims lack that 'real' feeling of a cabinet and tube head. Many go on to say that hardware sims such as the AVID Eleven rack device almost perfectly emulate the 'feel' of a real rig set up, and they are highly favoured. Why is that ? Perhaps Eleven has dialled in the specific latency between the player's ears and a real rig, whereas the VST sims are too pure. In this instance, it might me that in a (human perception) latency-free digital format isn't imperfect enough.

a) Thing is, what are the others in the band hearing? Referring to what I said about DAWs (and what you said you still have to think about ) They might think they play dead on the beat while in fact everybody else hears them with a delay. So in this case they should also *listen* to themselves.

b) Yeah, its mostly because these DSP based systems can produce the sound in less than 2 or 4 milliseconds and hence feel like a real amp (1 meter sound through air = 3ms "latency", even the best drummers can only hear/feel down to 3 or 4 ms of latency). A CPU isn't made to process at low latency, they are made for high overall throughput, if you add USB to the mix theres basically no way to get below 8ms (no matter what lies the ASIO panel of your audio interface displays).

As for the question on how to approach the delay issue when recording, I resorted to only do hardware monitoring and do live FX with analog devices or DSP systems (ProTools or UAD Apollo). If you have pro musicians most won't tolerate native systems latency anyway.

[Namahs Amrak]

I don't think anyone is arguing whether or not latency will exist with pitch shifting. Its just the Polar seems to have a much higher level of latency (noticeable latency) than someone experienced with pitch shifting effects would expect (especially at the quality it produces). Its not the most sophisticated sounding pitch shifting. So its just confusing why Polar's latency is so great...

OK. In the PUF thread I started late last year, the impression I got was that ALL pitch shifting devices, including both VST and hardware, was similar in latency to Polar. Your post tends to suggest that Polar is one of the worse ones, in your experience. Are you saying this about other VST's, or just hardware?

There is a pedal that I seriously have my eye on, but it costs a small fortune, for a pedal, in my opinion (I have only ever bought one pedal in my life though, and that was at a discount store !). Have you seen this:


Ravish Sitar


Clearly this pedal is shifting pitches, right ? Pay attention not only to the lovely auto- drones, but the 'mode' selection'. (may not be covered in that video though). I also see a pitch input alongside the expression input he speaks of.

In this video, we can see the playing and hear the music. It is, for all intents and purposes, latency-free. I doubt an indie reviewer, (even as a retailer), would be pushing the audio back to align with the session player's video example.

[Kazz]

I don't know about the latency on VST pitch shifters, but pedals and other hardware will (as far as I know) almost always give you a significantly lower latency than an equivalent software-only effect.

[chimp_spanner]

Yeah, the POG2 is a great low latency polyphonic octave generator. If you really must have the lowest latency, I think I'd just go that route. I've also tried (to no avail) to find a decent equivalent in software.

[submonsterz]

I find the best way to deal with it is to not use it lol.
(I do use it really but all ways for funky effects that latency does not matter)

[chimp_spanner]

Not really related to Reason but the best plugin I've found so far that can come close is Pitchwheel. Seems to handle chords well, has formant shifting and not *too* much latency live. It's latency free on playback, though.

[KEVMOVE02]

Just going to throw this out there... some studio effects are really hard to pull off in a live setting, such as harmonic layering of a guitar sound through a software plugin. The people I play with have settled on having the studio version of a song (all the bells and whistles included) and a "live set" version that relies on effects pedals. The beauty of the solution is you offer different versions of the same song that will appeal to a much larger audience. As far as recording goes, I find it easier to record a dry signal on one track, a processed signal (through a pedal) on another, then add additional signal processing during mixing. The less quantizing that has to be applied after the fact, the better.