KP3 or Kaossilator Pro

[lodger]

another option (if you have a nintendo ds) is the ds-10 software with a kaoss pad to accompany it. you get pretty deep programming (emulation ms-10) with the effects of a kaoss pad.

[MadTrucker]

I just bought the Kaossilator Pro last week, and I'm about to sell my KP3.

While I would normally tell anyone to go with the Kaossilator Pro rather than the KP3, you are totally different since you have an ample synth.

The problem with the KP3 is that it wets all the synth and drum sounds with reverb to cover up problems, and you can only partially remove the reverb (shift+slider). It's drum loops also have subtle seams at the start point...little pops that you won't hear until you spend time recording it and stuff. I am just talking about the drum and synth tones though. If these are of much interest to you, then the Kaossilator Pro posesses a much more professional-sounding and vast repetoir of sounds in it's tone bank that have no seams/pops or extra reverb.

On the other hand, the KP3 is great for synthetic-sounding filters like stuttering DJ loopers, a descent bit crusher and at least one descent vocoder. The KOP only have vocoder filters, and no other category at all. The rest of the differences don't metter to me. I know people love the way the KP3 can slice up samples, but for me It just made my loops off-beat and was not useful at all. Actually the features I miss from the KP3 are the hold button (the gesture changes considerably when using the shift button), the fake-reverse on the DJ looper (no scratch-like effects on the KOP), the little riff memorizer (whatever that is called), the auto-BPM, and the mute button.

However, it took me about an hour with the KOP, before I started writing the eulogy to sell my KP3, because I was willing to forfeit those losses for the new ground gained in the land of the KOP. That's because the KP3 has a character where you should clearly be able to detect it's sound when you hear it on a recording. It's highly distinctive and recognizable as a synth or filter, with very few exceptions. After taking some time, I feel the tone of the KP3 is cheap-sounding. It does not focus on sounds that are universally useful, except for it's DJ Looper, which although you find this feature on the SP404 & SP404SX, it is FAR superior on the KP3, and if you handle it just right you can almost make a convincing fake scratch technique. However, once again we experience that pesky reverb, at times almost seeming to mock the inclusion of the mute button (reverberated tone rings out beyond the mute, unless you sample your tone) for any real transforming. Thankfully both units have an input switch that can also double as a transformer in many situations.

So, if you plan to layer Kaoss sounds in drum-machine style loops, then the KOP gives a much more pro sound texture and tone color, but if you plan to filter (such as in an effects loop on your mixer) and you are not worried about the tone being very synthetic (much like my criticism of the SP404 filters), then the KP3 should suit your needs. They both do fine with sampling. I don't use MIDI right now though, so I can't comment on that aspect.

[LoneStar]

Just a quick question besides the topic, how can a synthetic filter (they all are!) NOT sound synthetic? Or do you mean "bad" like harsh, "digital", corny as opposed to creamy, lush, "warm" ?
I always wondered whether some Kaoss Pad model would be something for me, but all sound examples I ever heard kind of put me off. Especially the filters sound NOTHING like in a proper synthesizer, which is a shame because Korg DOES know how to code some.

[PHeMoX]

That's because the KP3 has a character where you should clearly be able to detect it's sound when you hear it on a recording. It's highly distinctive and recognizable as a synth or filter, with very few exceptions. After taking some time, I feel the tone of the KP3 is cheap-sounding.

Well said and I entirely agree. It has been the ultimate reason why I had to sell my KP3 as well. I got annoyed by the cheapness as well.

Just a quick question besides the topic, how can a synthetic filter (they all are!) NOT sound synthetic? Or do you mean "bad" like harsh, "digital", corny as opposed to creamy, lush, "warm" ?
I always wondered whether some Kaoss Pad model would be something for me, but all sound examples I ever heard kind of put me off. Especially the filters sound NOTHING like in a proper synthesizer, which is a shame because Korg DOES know how to code some.

Quite true, however it's the mix of basic additional FX that adds a way too distinct flavor to the sound. You can hear very easily whether a sound has been processed with a KP3. Synthetic is good, but it shouldn't make gear used sound over obvious. It also lacks a serious warm reverb to begin with, it sounds pretty shallow and almost metallic. Not sure how to describe it, but it's not that great for people less into pure electro stuff.

Still one of the best FX boxes on the market today, but you'll have to accept the pretty huge sound coloration.

[LoneStar]

Thanks for clearing it up for me.
In addition, you certainly can't say any of the SP-404SX effects sound top notch, it's too limited in terms of processing power for that. Yet I like using most of them, mainly because when I do quick and dirty stuff (which is all I do these days), it doesn't matter nearly as much as the push-button-receive-sonic-outrage fun factor.
So, I guess an additional Kaoss Pad wouldn't give me that much personally.

[PHeMoX]

Yeah, you're right, but the SP-404 doesn't color the sound. I'm not even that much of an audiophile though, so don't get me wrong,

...but the worst you'll be able to notice is probably just caused by a single resample run of your original sound. Most if not all FX are pretty clean and anonymous when it comes to their overall uhm 'flavor'.

[LoneStar]

Yeah, you're right, but the SP-404 doesn't color the sound. I'm not even that much of an audiophile though, so don't get me wrong,

...but the worst you'll be able to notice is probably just caused by a single resample run of your original sound. Most if not all FX are pretty clean and anonymous when it comes to their overall uhm 'flavor'.

Hm, I didn't notice any degrading in sound quality when using resample. People keep saying it's more obvious on the old SP-404, but then that used lossy sample compression, the SX doesn't. Theoretically there shouldn't be any change in the sound at all if you just resample without FX.

And yeah, now that you mention it, most of the FX sound pretty neutral and not "box-flavoured" at all. Good thing if you ask me.

[ellaguru]

Theoretically there shouldn't be any change in the sound at all if you just resample without FX.

theoretically no, but it's like if you re-record something, so even inside a software on a pc the sound change a litlle bit when you re-record (or resample) something...

[LoneStar]

theoretically no, but it's like if you re-record something, so even inside a software on a pc the sound change a litlle bit when you re-record (or resample) something...

That's the point of internal resampling, it is NOT like re-recording; that would mean the signal is going through converters at least twice, which certainly degrades the sound. With resampling internally, entirely within the digital domain, you don't have that. Rather, a digital data stream is stored. No change.

If I remember, I'll test that on weekend, would sure be interesting after how many times resampling any degrading becomes obvious.

[PHeMoX]

Theoretically there shouldn't be any change in the sound at all if you just resample without FX.

theoretically no, but it's like if you re-record something, so even inside a software on a pc the sound change a litlle bit when you re-record (or resample) something...

It's not a digital resample in the SP-404 series, nor is it in most hardware samplers actually. So you will notice a sound degradation after a certain amount of resampling.

It might not be super obvious when having resampled only a few times and especially when a sound ends up in an entire beat, but anyone who's ever made micro-wave resampled beats on an SP, knows it definitely will degrade the sound at some point.

Even digital resampling is not without changes, as although it might not look like it, you're cluttering the sound data stream literally in each area where multiple sounds share the same frequency ranges in a similar way as you would with analogue resampling / re-recording.

It's also why people love the fact that modern day hardware is polyphonic! Oh, and it's why in software resampling a second sound over an original one won't actually double the file size as if it would indeed not 'mix' anything 'down'.

[LoneStar]

No digital resample? Are you saying the signal is going through the D/A, then A/D again and is (re)sampled that way? Come on, I bet you don't have any more evidence for that than I have for the contrary ;P Seriously, how do you know? It doesn't make sense technically at least if you could (as the engineer) just as well loop back the signal BEFORE it goes through the converter, hence, digitally. Easier to implement, cheaper (digitally controlled analog switches for instance are quite costy, digital switches are cheap) and much better in the first place.

And remember, the 404SX is the only SP so far which doesn't compress samples. Or are other models (555, 606 perhaps) storing samples lossless too?
Anyways even if resampling is purely digital, the sound would still get decrompressed and compressed again --> degrade. Not the case on the SX.

But again, I'll test that on weekend, and we'll know more then.

[PHeMoX]

I and others already did resampling tests (on SP404 older version), but sure go ahead and try.

Even in high end samplers resampling isn't (necessarily) going to be digital or virtual! Most will use an internal analogue sampling unit.

And remember, the 404SX is the only SP so far which doesn't compress samples.

That's a good point, but it still doesn't mean it isn't using analogue resampling methods.

There's no doubt in my mind that it will improve overall quality when compared to the old SP-404 though.

[LoneStar]

Even in high end samplers resampling isn't (necessarily) going to be digital or virtual! Most will use an internal analogue sampling unit.

My question still stands: How do you know? And what the heck is an "internal analogue sampling unit"? Additional converters used for sampling only? I can't imagine why any engineer would implement something like that if he can simply put a software-controlled loopback between the already existing converters. You do realise that converters are some of the most costy parts on a sampler, right? Also, do you realise that a sampler is basically just a kind of processor (mostly specialised DSP) with not much more than converters and interface/storage periphery around it?

[PHeMoX]

Oops I meant to say 'sampling engine', not unit. But the comment basically still remains the same.

Perhaps the whole distinction we try to make here is a bit off though, as even hardware needs to be controlled through commands, data structures and all that. But I do think sound does get routed towards the sampling engine in an analogue way.

Also, do you realise that a sampler is basically just a kind of processor (mostly specialised DSP) with not much more than converters and interface/storage periphery around it?

Well while my actual technical knowledge isn't much when it comes to the guts of samplers, yes I am aware of that.

[LoneStar]

So you're saying, a sampler like an E-MU Ultra or Kurzweil K2600, both having digital inputs too, resample through the converters? Makes no sense, really. Also as I already said, a digital loopback is easier and cheaper to do than an analog one.
AGAIN, what makes you think that on most samplers, even the good ones, resampling goes the analogue way?

EDIT:
Come to think of it, maybe the loopback of the signal is even done within the DSP. Even easier.
Anyways I quickly made a VERY schematic and simplified drawing to make clear what I mean; how digital resampling is easier and makes more sense anyways. For the time being, until we know for sure let's just assume the most likely option, eh? And you'll have to excuse the bad jokes in there too

http://www.box.net/shared/863as6f9h0

EDIT2:
Look what I found in the E4K manual:

http://www.box.net/shared/nka61ok2zs

Why is it that, as soon as you select resample as source, the ADC gain setting is greyed out?
Because the ADC isn't used. Likewise, the manual says you can't use the ADC gain setting when sampling from the digital input. Clear signs of a digitally implemented resampling path if you ask me.