Phasey sound on PA4X Pro

[kowakorg9]

When building up keysets using strings I find that when I press the keys and have layered strings, the sound becomes 'Phasey' at times. I think the strings of the PA4xPro are exceptionally good individually, and have layered the strings successfully, but in the keysets I make the phaseyness is there, but not always.

It might be some interaction between the effects but should not be there.

I think Korg need to look at this.
I have had many different Korgs over the years , from M1 upwards, also Tyros, Kurzweil, and a Yamaha XF7, and have not had this issue before.

I also think the french horns and saxophones are far better on my Yamaha XF7 , and have used both programs that came with the instrument as well as others imported and sampled from instruments like the Korg M3 and Kurzweil.
If you spend time tweaking the effects then you can improve the sounds, but always better to start off from a really good sample. I don't regret buying the Pa4X Pro, and it has really good drums and percussion, very realistic.
Anyone else had phasey sounds on layered strings ?
Finally, the Pa4X pro and XF7 work very well together and you can build up some amazing sounds and performances, particularly with the arppegiator patterns.

[karmathanever]

Create an example of this for us to try (i.e. SET)

[mintjamman]

Nope !

[karmathanever]

[Sharp]

When building up keysets using strings I find that when I press the keys and have layered strings, the sound becomes 'Phasey' at times. I think the strings of the PA4xPro are exceptionally good individually, and have layered the strings successfully, but in the keysets I make the phaseyness is there, but not always.

It might be some interaction between the effects but should not be there.

No this is perfectly normal.

If you layer sounds together that use the same multisample, then yes, you will most certainly run the risk of having phase issues as that's how audio behaves. This is a well known fact of audio engineering.

I have had many different Korgs over the years , from M1 upwards, also Tyros, Kurzweil, and a Yamaha XF7, and have not had this issue before.

This has always existed, but given the choice of sounds on the Pa4X, you maybe unknowingly be favoring the same multisample set when your making up your layers.

Regards
Sharp

[kowakorg9]

I don't think this is 'Normal' at all.

I can layer strings (with effects) on my Kurzweil and my yamaha XF7 and there is no phasiness on strings detected.

I am convinced this is a Pa4X pro design issue. Having said that, the strings used alone or in limited layering are very good indeed. Although, I tried out the Yamaha Montage a couple of days ago on an extended listening session, and the strings and brass and orchestral sounds (and control of them) was amazing. Extremely expressive, which is what you need when composing music and playing it.

As a trained classical musician ( in my younger years), I know what strings and brass and orchestral instruments sound like, having played in several orchestras. There is no phasiness I ever heard in real life in a string section. A slight detune between instruments can broaden the sound, but it's not the same thing as a "wishy-washy" phase issue.

Also, don't assume that all computer sample packs of orchestral instruments out there are "accurate". I accept this is a subjective issue to a degree, but have heard some really poor samples and with all sorts of artefacts which detract from the sound.

In an instrument such as an arranger keyboard, the accuracy and expressiveness of certain sounds may not be seen as critical by the manufacturer, but when you get to the level of say the Kronos, or the Montage, it certainly should be extremely high quality and accurate.

[AntonySharmman]

kowakorg9 it's very well explained by james that when using 2 sounds with at least one common Rom multisample (OSC)
then phase effect or sounds same phase interaction or even common frequencies cancellation will definately occur , that's
a rule in Physics and sound engineering !
Try factory Keyboard Sets of strings and see the obvious , do not speculate issues that do not exist and be more carefull
when combining Sounds there , that must not use the same Rom multisamples (sound edit/OSC) and all will be fine !

[kowakorg9]

Antony, I am not doubting the phase issue between using common multisamples-That is for me , a technicality.

My point is that I do not have this problem with a range of other keyboards and string sounds, and some of those will probably have the same multisamples layered too.

In driving a car I don't really care how the fuel gets to the engine, I just want to drive it. It's for the clever engineers and designers which have far more knowledge than I in design to sort out the issues, so it just works.

So, as a musician, I don't accept that a phasiness occurs in some of the layering, irrespective of multisample issues. It's up to the manufacturer / designer to ensure that when the customer presses the keys to later certain sounds such as strings, that a phasey effect does not occur. If that means there needs to be more, or different multisamples, so be it.

Too many compromises seem to be made in design of a whole range of goods. I don't mind this on low budget items, but when the customer pays a lot of money for the goods they expect the main bugs to be sorted.

Operating systems on computers are a case in point. They are complex, for sure, but some have been a nightmare for customers, ( Microsoft please take note.)

So in summary, when I layer sounds of an acoustic nature, ie orchestral sounds, I want them to sound realistic. If they don't, the manufacturer has not got it right, and it needs more work.

[AntonySharmman]

That is for me , a technicality.

Then do not edit sounds and just use factory keyboard set otherwise you have to take care of those phase issues !
Me unlike you as a sound developer and owning all your keyboards and a bunch of more , I always face this issue
in any brand or VST if I work deeper with Multisamples .
As a sound developer , when I provide StringsA & StringsB sounds in a workstation then those two sounds are entirely
different , but if you try to layer twice StringsA or StringsB in a Combi or create your own sound using common samples
with those then the blame is yours if you face phase issues when combining them !

Hope you've got my point here !

[Sharp]

Hi kowakorg9.

The only person who is able to fix this issue is God himself. Your opinion on this subject is totally irrelevant because there is only one single conclusion in all this. Phase exists and has always existed.

The only thing you need to focus on is knowing what causes Phase and what you can do to try best avoid it or reduce it. If you really must mix sounds together that use the same multisamples, then you need to modify the filters of each OSC so that the multisamples sound different from each other leaving the sound engine. This will help in reducing phase issues, but I still think it's better to simply avoid the issue in the first place by not layering the same samples together.

Another trick is to transpose a layer up 3 semi tones and the other down 3, then use the tune function of the OSC's to bring them back in tune with each other. This stretching of the samples makes them sound different from each other, and helps in reducing phase. Best to play with this to see what works best though because personally, as above, I would rather simply avoid the issue entirely by simply not layering the same samples together.

You will personally need to find the happy balance between the two that works best for your ears. Antony and I can certainly advise you on tips and tricks, but asking KORG to solve this for you is a complete waste of time. KORG didn't create the universe.

Regards
Sharp.

[ausser]

Another trick is to transpose a layer up 3 semi tones and the other down 3, then use the tune function of the OSC's to bring them back in tune with each other.

Must remember that one.

Aus.

[kowakorg9]

Hi Antony and Sharp.

I accept that the technical issues are based on physics and the limitations of the systems used. Perhaps in the future we will look back and wonder why there was a problem here, because someone else has sorted the system.

I will use layering factory sounds more now, but still find it strange that when layering similar multisamples on my XF7, I don't seem to have the problem. Or maybe I do have it, but it is not nearly so obvious, listening to it.

There was a time when manufacturers and designers said it was not possible to overlay sounds, ie so that one could switch between voices or programs and have the last sound still dying away. That was frustrating.
Well Korg and others worked out how to do it on the Kronos, as did Yamaha on their recent Montage. Cost comes into it too, probably.

I accept that musicians and sound developers may look at the problems from different viewpoints. I once had a real problem working closely with a programmer on designing some computer software, until we stood back and looked at it from each other's viewpoints. Once we did that, we progressed much better together.

[AntonySharmman]

It seems that you don't want to understand our tips ...
I own Motif XF7 and I can replicate this in a couple of seconds layering the same sound , the same in Kronos as in any
audio file when have the same or closer phase , so do not speculate of what could suite to you and you'd better ask for an
advice in forum if you ever have a real problem with your arranger ...

[kowakorg9]

Even I am able to create a phasey sound on my XF7 and other keyboards and modules, many which are not listed.

But that's not the point. I think we have exhausted the common multisample argument, and accept your explanation.
Don't understand your last sentence at all, but I expect you do.
Thanks for your input.

[Sharp]

Hi kowakorg9.
The only way this will be resolve in the future with technology is if KORG write the code that checks the names of the multisamples your using, and if it finds your using the same one more than once layered in sounds, then a script will need to run that applies the tips / tricks for reducing phase for you.

This doesn't solve the problem, it just uses code to detect a possible chance of phase, and then automatically applies the tips and tricks explained above for you.

So yes, you have a point. There is a solution, but the solution is to simply do what we have been saying all along, only this time you want the keyboard to do it for you.

Regards
Sharp