It might really happen: the PC3 correction

[Theo Verelst]

I decided the main correction that makes all the built in (ROM) sounds of the PC3 at least bearable, and a lot more interesting IMO, more neutral, less distorting (again, IMO) more pro-mix balanced with neater sounding mids and more big listening speaker suitable low-mids and some other things, freely available when it's done to a certain satisfaction. I'm not there yet, but I can play most ROM sounds well enough such that they don't apply suction to my rear end anymore with some of the corrections I've made.

 

A particular setup I want to use is a globally overridden AUX2 effect in Program mode of the Kurzweil by a sequence of effect blocks that corrects all layers with distortion and mix components at a relatively low send level (probably it's normal settings make it -20dB or less most of the time), combined with specific and precise setting of the Master Effects, consisting of EQ and 3 band multi-compressor.

 

Want to hear a test result with pure ROM sounds or some sounds I'd prepared (for other purposes), here are two songs (i.e. song mode of the PC3) being played into analog mixer inputs connedted from the main Kurz outputs, digitized at 192kHz/24bits, and encoded by ffmpeg (on Linux) to the standard 44.1kS/s stereo mp3 format many are used to:

 

kurz_ana_songts0.mp3 2min 26s (copyright of music: me, only free to use for non-commercial demonstration purposes)

 

There isn't all perfection, which sometimes starts to work when fine tuning individual patches, but it's not longer that particular sound I disliked the whole time while using the PC3(-76) where these tracks were played into. No software, hardware, external sequencing or midi or effects were used, and of course a higher quality recording would do more justice.

 

Also, there are 4 tracks in both songs, which made some insert effects and send effects of some not work, I'm working on getting the "correction" effect unit count down.

 

T

[harmonizer]

I am assuming the mp3 you posted is the "after" result, and that you have done some tweaks to remove some undesirable things from the sound. The scientific thing to do would be to post two clips of you playing the exact same thing, one with the correction and the other without the correction.

[Al Coda]

The scientific thing to do would be to post two clips of you playing the exact same thing, one with the correction and the other without the correction.

 

 

This !

 

Plus some PC3 screenshots showing the improvements made in regards of (custom) algorithms, that mysterious AUX2 FX chain and "improved" master FX.

 

A.C.

[MurMan]

We may be witnessing history. The development of the distortion correction algorithm will make TV eligible for the Nobel Prize in Information Theory.

[marczellm]

they don't apply suction to my rear end

theoretically speaking, at least.

[Theo Verelst]

For those who like to try out the sounds from the first sequence above which I recorded on the PC3, (I used ROM sounds: 18, 114, 105 and 771), here are the exact settings and effect I used. There's the graphics for those who want to fill these parameters in (I don't know how to save Master Eq & Multicompr), and there's a download for the large number of 4 chains effect parameters on the override AUX2 effect chain.

 

This work isn't all there where I've tried to go: less AUX2 effect units, and a little shorter response of the correction filtering, and more accuracy and even volume leveling over (most, meaning at least half or more) which I've gotten closer to with another, similar except for the multi effect chain, setup, but it's not done yet, and I suspect there are more things I can do with this, because the machine starts to respond normal, and is quite powerful, audio quality-wise.

 

Here are the settings you need to imitate the exact settings I recorded above. I've verified the digital output is ok, too.

 

The override effect chain in it's memory location:

http://www.theover.org/Keybdmg/Kurzweil/Globcor/globeff2_2475.gif

 

The effect chain, with me the number in memory is 2475:

http://www.theover.org/Keybdmg/Kurzweil/Globcor/eff2475.gif

 

The master EQ and m-compressor:

http://www.theover.org/Keybdmg/Kurzweil/Globcor/eqmast1.gif

 

The global AUX(only "2") settings:

http://www.theover.org/Keybdmg/Kurzweil/Globcor/glaux2.gif

 

For the sake of open synthesizer/workstation patch work, here's the download file containing the above effect 2475: DSTV4GLOBOtst0i.pc3

 

So for those who aren't used to the idea of working with the Kurzweil global effects, or maybe want to try out the idea to improve more than a few sounds from more recent machines like the Forte: I use the global effects EQ (a 5 band parametric equalizer built in the PC under front panel button "effect"-Master-block->EQ) and Cmpr (multi compressor). It is an exact match with the override AUX2 effect that should be applied to all the channels in the Effect->AUXFX2 global page at -6 dB or maybe a little different if you feel that works better for certain patches. That's the 5th lin from the top with either [p] for program levels of effect two, or a number between -96 or +24 (dB) for effect send levels for each channel, corresponding to each layer in the mode of the keyboard. Here's where Override must be set to "Yes", Chain to point at the above effect (name as specified, location depends on how you loaded the .pc3 file into the Kurz), output Pri. and Mod Override to ON

 

I think most people that know their ROM sounds will "get it" what improvements I prefer. It might be that your favority guitar amp for the exact live show you've run for years doesn't sound recognizable with the AUX2 override and the global EQ/MUCO setting, but going over the ROm sounds in program mode should be easy, putting all unused (i.e. 2..16) channels to "0" (chan/Layer "up", type numerical 0 on keypad, press enter, etc until back at "CH:1" in the upper right corner of the screen, first so that all possible effect units are available.

 

The main setting to play with if you feel like it is the "AUX2 override "send level", from -6 to a little less or more.

 

T.V.

[harmonizer]

You provided lots of details on how someone who owns the same gear as you can use the same settings.

 

For those of us who don't own the same gear as you do, can you share two sound clips of you playing the same phrases of music, with one clip using generic settings, and the other clip using your modified settings?

[Theo Verelst]

no.

[Sven Golly]

http://www.hangtogetherblog.com/wp-content/uploads/2015/03/Shenanigans.jpg

[Delaware Dave]

You provided lots of details on how someone who owns the same gear as you can use the same settings.

 

For those of us who don't own the same gear as you do, can you share two sound clips of you playing the same phrases of music, with one clip using generic settings, and the other clip using your modified settings?

just play any two clips back to back. They'll sound the same, same as Theo's adjustment will.

[Al Coda]

http://www.hangtogetherblog.com/wp-content/uploads/2015/03/Shenanigans.jpg

 

 

 

A.C.

[Al Coda]

no.

 

O.k., so I´ll never find out ´cause the master FX on my PC361 are always off and my 2 auxes are occupied w/ 2 Aux-FX which I,- up to now,- always need.

I fear changes, you know ...

 

Man, I know, I´m a picky a##hole, even since I heared rumour KURZ wants you as a specialist for already and soon to be discontinued products now.

 

Sorry ...

 

A.C.

[DulceLabs.com]

http://s2.quickmeme.com/img/55/55385a77fc49e908935f17c21adf548f5d36527ef30989ab71ef502bd51e1f48.jpg

[Hoochie]

Sorry Theo.

Unless you can do the scientific thing here and give us:

What your trying to achieve

Control samples.

(This should be the untouched sound recorded in the best clean way you have)

How and what you are doing (in clear and concise language)

The results( recorded in the exact same best way manner)

All this in a repeatable manner.

Hopefully it will show a marked improvement.

 

If what your trying to get across is strongly relevant and does something outstanding then hooray!

 

If it's sonic minutiae it's meh.

You may be chasing ghosts that others don't see or hear.

 

So as others are saying here. Prove your point please.

Give us controls to compare to the final results.

That's pure science and math AND engineering and you should know this.

If your going to argue semantics, there are people here that will call you out.

 

Carry on.

 

John

 

[Mighty Ferguson]

they don't apply suction to my rear end

theoretically speaking, at least.

Technically, this is Theo rectally speaking.

 

 

 

 

 

I'll see myself out.

[Theo Verelst]

Very late 90s I did some work in an art gallery, among other things making digital images of (canvas) paintings, with works from a significant portion of the Dutch visuals arts people of name. Luckily, usually the purpose, with early digital camera's was to get a relatively true-to-nature rendering of the paintings on the available computer monitors, so to speak High Fidelity.

 

However, a lot of the interest and attention also seem to wander to the idea of *improving* the digital renderings of a (oil) painting, not so much along the lines of HiFi, but more getting nicer colors, improving lines, removing artifacts, and getting details to appeal to people, without necessarily doing justice to the original work.

 

I get it people aren't willing to try to make their digital instruments perfect, but it *is* normal for advanced musicians to work on their "sound", and seriously so.

 

T.

[Al Coda]

I get it people aren't willing to try to make their digital instruments perfect, but it *is* normal for advanced musicians to work on their "sound", and seriously so.

 

T.

 

I dunno what an advanced musician is for you, but my understanding and experience is, the advanced musician, especially being a pro, is going out performing on stage and in studios using his talent and time to make money.

When there´s time in between, there are the hrs for practising and spend time w/ family and friends etc..

 

It´s true they also make some patches, edit presets and/or program their live-rigs in regards of audio-routing, MIDI and all kind of stuff related to organization,- but tinkering w/ sound like you do is at best for the sound designers.

 

I don´t see a sound designer has to be an "advanced musician" though.

Too many one finger players have good ears and make excellent patches for hardware and software synths.

 

Since how many years do you own your damn KURZ PC3 and tinker w/ it, trying to reinvent the wheel ?

When you dislike so many details on your digital instrument(s),- sell ´em or leave ´em as they are and buy analog gear.

 

A.C.

 

[Outkaster]

Advanced musician? Then you should have been practicing for years.

[Al Coda]

Advanced musician? Then you should have been practicing for years.

 

That´s what I did before I started making a living from music in the late 70s.

But, I more or less always practise, but not that much as I did in the past,- and I program patches (also on the PC3 !) ...

 

B.t.w., I don´t call myself an "advanced musician" because I practised and still do.

I don´t know what that term means.

You´re a musician or not and you´re sounddesigner or not,- or whatelse, maybe both and so on ...

 

But that babbling about the PC3 and "corrections" is insane and stupid as well IMO.

 

A.C.

[xKnuckles]

Dyslexic PC3:

 

http://4.bp.blogspot.com/-Bww5DhuzpXg/TlfmVbrmfqI/AAAAAAAABcs/kDeb9yUEguE/s1600/starwars+c3po+headscreenshot1.jpg

[MurMan]

Dyslexic PC3:

 

http://4.bp.blogspot.com/-Bww5DhuzpXg/TlfmVbrmfqI/AAAAAAAABcs/kDeb9yUEguE/s1600/starwars+c3po+headscreenshot1.jpg

How ironic and insightful. The OP would benefit more from a translation and protocol droid than a PC3.

[Hoochie]

they don't apply suction to my rear end

theoretically speaking, at least.

Technically, this is Theo rectally speaking.

 

 

 

I'll see myself out.

 

Hahaha

[Hoochie]

You know all this correction speak might have some merit.

But not in the way it's presented here.

Take a Chandler or Summit Audio

stereo tube DI preamp and

a couple of self powered L'Acoustics.

That is going to make a real difference.

Of course it's gonna cost you $15,000 or more.

 

And if you start reading up on FIR that you will give you a headache too.

Lots of correction talk. Phase talk. Null talk. Delays. Steering sound. Shading.

But they prove their points with results.

The best place to go is soundforums.net.

 

John

 

 

[Theo Verelst]

For the people who want to do the relatively little effort to try this out, the improvement is good, even if the result responds perhaps a tad slow (i.e. the correction effect, not the key-to-sound, which becomes a bit better in fact) and isn't automatically perfect for all ROM patches, however maybe half of them are immediately quite more usable and restful, automatically, without changing any settings, just scrolling through them. If you want to try the example music built in the machine, set the MIDI sensitivity for the main channel (I presume in program mode you work on midi ch1) to "hard2" to get a better match from the ridiculously hard played example song per sound to the sensitive sounds.

 

To get the exact sounds of the example music I made with the ROM sounds + corrections, you need to use the exact effect as supplied (from the screen dump of the sound tower knobs or the provided download for the PC3), and the other settings for the effect sends and so on and the master effects are very sensitive: they must be set accurately to the exact values specified, in some cases even a parameter change of +1 or -1 is enough to spoil the sound equilibrium or however you want to put it.

 

T.

[Theo Verelst]

 

For those interested pretty much this (a little later iteration for a only slightly different purpose) set of effects is demonstrated in an easy to access Youtube video in this thread.

 

The latest news from my side is that I've started to get a successful combination of (Linux, PC) software together that acts as comparable correction, give or take a few secondary subjects. I'll probably report about this on the internet, so that people can try that our, too, if they have some of the important Linux Audio packages (Alsa-tools, jack-rack+Ladspa plugs, Carla+recent Calf plugs(LV2 or Ladspa or standalone), Jamin mastering effect) and preferably a digital input and a quality DAC on their computer, and it has to be a bit (but not overly) potent.

 

So in short I reasonably accurately imitated the equalizers at the beginning of the PC# AUX2 chain, added a chorus/flanger followed by distortion effect and some additional details, which I applied to the digitally interfaced PC3 output, up-sampled to 192kHz (alsa_in -quality 3). Then, an imitations of the Mastering EQ and the Mastering Multi Compressor (from the Calf effect suite), pretty much the settings copied from the corresponding PC3 effects, and tata: it works!

 

T

 

[Bobadohshe]

I'll take the blare control and subtlety of the original samples over these

tweaked sounds anytime.

[Theo Verelst]

well, they are preserved, and you could be more clear.

[Hoochie]

Theo,

 

 

I've looked at your chain.

There is a lot eq.

Boosts and cuts like your drawing an eq curve with the tools available in the pc3.

Where those boosts and cuts are will effect how the energy of sound in those particular bands (that are divided by the crossover) going in to the multi band compressor will begin squashing your sound.

Correct so far?

 

The chorus delay is just an effect.

The amp sim is also a sound shaper as in it try's to emulate an amp and the saturation that give you.

It smooths over any spikes in a hopefully harmonically pleasing way.

Almost another compressor that folds harmonics into the original sound

as energy is increased into it.

 

Are you trying to get a smooth response with limited aliasing?

A smooth response I can understand. I am under the impression

that the PC3 has some issues between types of sounds but not

something that's very noticeable. I think they did a better job of

this than anybody else.

 

If your playing back through a very high quality system.

And I'm talking Skywalker Ranch, Dolby labs, Meyer, etc.,

then yes, errors are likely noticeable.

But come on. Who has that?

 

Aliasing is not going to be fixed by eq where we can get at it.

It's in the converters and or the synths sample rate.

 

If your getting the digital feed out of a keyboard and using your own convertors

at 1.21 jigahertz, then yes, that might help.

Nyquist figured all this stuff out a long time ago.

Here is a piece that clearly explains some of this.

It's a good read. Short, concise, to the point.

http://www.trustmeimascientist.com/2013/02/04/the-science-of-sample-rates-when-higher-is-better-and-when-it-isnt/

 

That audio you posted.

Did the tracks not get matched up?

Were you trying to use an avant garde bed of things

to bring undivided attention to something in there? The Epiano?

Your leaving me wondering " what the H E double hockey sticks"

is going on here.

 

 

I don't know if I should try your improvements.

Prove to me it's worth my time.

 

John