Sample Editor - Input and Output Do Not Sound The Same

[analogika]

Wherever possible, record samples (of any instrument) dry and add effects later. Your Piano 2 has overdrive (various amp sims, try several and see which works), so use that. Your sample sounds to me a lot like those early 90s attempts to sample a distorted guitar on home keyboards - it just doesn't work very well. Did you normalise (or maximise) every one of your samples first as well, not just the whole file?

I would also suggest you turn down/off the key click, as it only occurs when a key isn't held down on the original, but will occur on every sample on the Nord and frankly sounds better not there.

This is incorrect. Key click is always there whenever a key is pressed. You would be correct if you were referring to the percussion feature. That is only triggered when a key is pressed while no other key is being held down (and when the percussion switch is turned on, but most Hammond clones sadly omit that feature).

[RedLeo]

My latest project has been trying to sample one of my organ patches from my software out to the Nord.

I'd just like to be clear what you're actually trying to do. I understand you to be saying that the distorted organ patch is created in some piece of software (VST, whatever) and you're trying to sample this software-generated sound into your Nord Piano. Is that correct?

[RedLeo]

@Cornopean

The Nord amp overdrives are broadly similar to the overdrive in the original sample, so there's not really any improvement to be made there. The sharp attack to the sound isn't really organ keyclick, it's just an artifact arising from the extreme amount of overdrive being used. Anyway, if the organ sound is being generated in software - hopefully the poster will clear up exactly where the original sound is coming from - then I don't believe that the huge difference being heard is anything to do with the sampling process at all.

The problem the poster is having is nothing to do with trying to get a "better" organ sound. It's to do with the fact that the distorted organ sound he's trying to sample into his Nord sounds completely different when he plays back the results of his sampling session. As nothing in the sampling process can remove the distortion from the original sample or add a release segment to the sound, it seems pretty obvious that the "input render 001" and "output render 001" are just two completely different sounds.

[Cornopean]

I would also suggest you turn down/off the key click, as it only occurs when a key isn't held down on the original, but will occur on every sample on the Nord and frankly sounds better not there.

This is incorrect. Key click is always there whenever a key is pressed. You would be correct if you were referring to the percussion feature. That is only triggered when a key is pressed while no other key is being held down (and when the percussion switch is turned on, but most Hammond clones sadly omit that feature).

You have misunderstood me. The "original" I refer to is the huge, clicky sound on his original sample, which is only there (I think) when another note is not held down, or possibly some other combination of factors, but is there on every note on the subsequent Nord sample. On the real Hammond, key click is actually slightly less perceptible when a note is held down but still there (or at least it was on my C3), I presume due to less discontinuity in the waveforms, or maybe it's just a perceptual thing.

Al

[GaryA]

My latest project has been trying to sample one of my organ patches from my software out to the Nord.

I'd just like to be clear what you're actually trying to do. I understand you to be saying that the distorted organ patch is created in some piece of software (VST, whatever) and you're trying to sample this software-generated sound into your Nord Piano. Is that correct?

Correct.

[Cornopean]

@Cornopean

The Nord amp overdrives are broadly similar to the overdrive in the original sample, so there's not really any improvement to be made there. The sharp attack to the sound isn't really organ keyclick, it's just an artifact arising from the extreme amount of overdrive being used. Anyway, if the organ sound is being generated in software - hopefully the poster will clear up exactly where the original sound is coming from - then I don't believe that the huge difference being heard is anything to do with the sampling process at all.

I disagree. Overdrive is a summative process, for a given amount of overdrive, a single note will sound relatively pure, but multiple notes will increase the perceived amount of overdrive significantly, depending also on whether the notes in question are harmonically related or not. The amount of overdrive will also increase as the volume increases, and as more drawbars are pulled out (obviously this last effect cannot be simulated with a single sample)

Therefore sampling a note with overdrive and not using an amp effect will always result in the same amount of overdrive, whether one note or many are held down, and whether it is played quietly or loudly. This is not the effect of playing, for example, a Hammond through a Leslie amp, or increasing the volume to get more overdrive, or playing anything else through an overdriven amp. To get an authentic overdriven sound, a clean sample plus a decent overdrive or amp effect is much superior to a sampled sound.

On the keyclick you may be right, I've often seen sounds such as this click described as keyclick on simulations, although it bears little resemblance to that aspect of a real Hammond sound.

Al

[RedLeo]

@ Cornopean Your point about how distortion interacts with notes played is quite valid, but nothing to do with the problem at hand. Perhaps we could help him solve his actual problem first and debate the finer points of hammond emulation later?

The poster still seems convinced that he's losing his distortion because of the sampling process itself, or that the distortion on the original sample is somehow affecting the sampling process.

@GaryA You said that by removing the distortion from the original sample the results were better, but still there was not a one-to-one correspondence between "input" and "output".

A 16 bit sample at 44.1k will sound effectively identical - that's CD quality sampling, so the fact that there is still a noticeable difference in sound still suggests to me that you're not hearing the same sound. It would be very useful if you could post the "before" and 'after' sounds of this new sampling attempt to give us a better idea of what's going on.

[analogika]

Okay. Having finally listened to the originally posted examples, it seems obvious that if these are the same sound (and it's quite possible), the first one has massively distorted amp simulation going (a guitar amp, not a Leslie - more like a Marshall or so), while the second has none at all.

Amp distortion isn't something you can sensibly sample for polyphonic instruments, as it is a direct result of how loud the input is, i.e. how many notes are played, and consonant waveforms (octaves, fifths, fourths) add up very differently from dissonant ones.

[neolithic]

Clearly these are *not* the same sounds! . I disagree that these tracks could even be the same 'patch': the sounds appear to have completely different amplitude envelopes apart from anything else!

[Cornopean]

I have been resisting this, but my comments (and belief that this is the same patch) are based on my experience teaching sampling and digital audio to university students at a good university. I set them a sampling assignment: to take an instrument, and create a professional level sampled plug-in. The best had excellent source recordings, key switching (if appropriate), multiple velocities, careful stretching, suggested (but not imposed) effects, beautiful demos and manuals etc. The worst? They sounded far less like the source material than this, assuming I could get them to work at all - sometimes I literally had to reconstruct the instrument, in order to give some marks for the effort. And I have experienced hundreds of both good and bad submissions.

I have, therefore, developed quite an ear from relating alleged samples of things back to the original, and working out what went wrong along the way, and then explaining this to the students clearly, to help them improve or understand their poor mark.

And I believe that those are both the same sound, as the poster says, and based on my previous experience, he is perfectly capable of getting the hang of this process if we can all give him some clear advice!