|
Post by psionic11 on Jan 22, 2010 13:53:48 GMT
Aha, that's very useful info. I guess the Fusion just prefers to store/import audio tracks as 24bit .wav, but for use as samples/multisamples you can choose whatever bitrate you want. In my first 'industrial' band (more like indie industrial) we used a Mirage, and that was fun triggering all kinds of sampled sounds, although I personally thought many of the sounds were cheesy because of that 12bit sound. My DW8000 was much more present because of its digiwaves and resonant filters, and I also had the chip that made it bi-timbral as well.
Anywho, if 16 bit only affects quality, and very minimally at that, and NOT dynamic headroom (I want pianoforte for the C7!), then 16bit it is! I know I can re-create loudness with Fusion parameters, but just as I could also re-shape the sound using ENVs, I want to stay as close to the original's contours as possible. There are subtle but very important differences, especially when everything is stacked up and you're in performance mode, and using a limited set of ENVs (no matter if you scale shapes/lengths/levels using keytrack and velocity, etc) can only mimic the real thing so far.
Ack, that was a terrible sentence. In other words, just as you can take a piano multi-sample, and re-do the velocity and ENV response so that it's a non-sustaining, non-responsive patch that doesn't resemble how an actual piano sounds, despite the samples, then the converse is also true. You can try to re-create actual piano ENV curves and responses that change with pitch and velocity, but chances are that you'll intrude upon the sound more than you'll help it. After all, with one-shot samples, most of the relevant ENV information is already imbedded into the samples anyway.... no need to use artificial ENVs... a simple gate envelope will do (open immediate attack, full sustain, release dependent on velocity, pitch, the stage of the actual sample)... A 127 velocity A0 has a decay much different than a 37 velocity A0.
But I'm getting ahead of myself. First comes the creation of some manageable multisamples. Fusion-specific parameters later.
|
|
|
Post by Shreddie on Jan 22, 2010 16:22:12 GMT
Anywho, if 16 bit only affects quality, and very minimally at that, and NOT dynamic headroom (I want pianoforte for the C7!), then 16bit it is! 16 bit does affect headroom, but only minimally... The biggest problem with different bit rates is your noise floor. These figures are from memory so they may not be bang on but here goes! The theoretical maximum dynamic range for 16 bit systems is 96db. I reality, due to the noise floor, 16 bit audio systems usually achieve between 88 and 93db dynamic range above the noise floor. This is called the signal to noise ratio or SNR. The theoretical maximum dynamic range of a 24 bit system is about 130db IIRC but the real figures usually vary between 105 and 120db SNR depending on the quality of the components. However, you do need to be listening quite loudly to be able to detect any noise in a well designed 16 bit system, let alone a 24 bit system... And to detect that reliably you need to have an extremely well designed (very low noise) analogue amplification system... The digital signals must be converted to analogue then amplified in that domain if you are to hear it after all! Unfortunately, such high quality systems are out of reach of most of us mere mortals. Also bare in mind that audio CDs are 16 bit, and that very few have ever complained about their lack of dynamic range. Another thing to think about is what the playback system is. For example, the EX7/5/5r is a 16bit synth that plays 16bit samples but it achieves an SNR of around 96db because its converters are actually 18bit. The Fusion has 24 bit converters with an SNR of about 110db IIRC so you could get a performance gain by using 24bit samples over 16 bit but that gain would be so minimal it would be next to pointless except in the most demanding of circumstances. Yet another thing to bare in mind is that with converters like those in the Fusion, the oversampling when converting to analogue is very good. Without it a 16 bit waveform could look very lumpy in comparison to a 24 bit waveform but the oversampling smooths it so well that if you were to compare the two after oversampling, you'd be hard pushed to tell the difference. Of course, there are better converters out there but those in the Fusion are of a higher quality than those in many low-mid range audio interfaces... And even some more expensive ones so it's really not a problem. Anyways, we're beginning to get into very technical territory now... Kinda theoretical stuff that's useful to know but isn't going to help your sampling much! My advice would be to do your sample editing/looping in 24bit format, save those on your computer should you want the 24 bit samples at a later date, then convert them to 16 bit before sending them across to your Fusion... That's what I do regularly anyway. Which is why some people do like large sample libraries. But you can go a very long way to recreating the dynamics of the real thing using looped samples... So far in fact that only those with a trained ear will be able to tell the difference... And if it's in the context of a mix, even those with a trained ear may struggle if the samples are looped and synthesised well.
|
|
|
Post by markone on Jan 28, 2010 14:46:39 GMT
We don't know we're born do we?
The first pro studio I ever went to had an old turn of the century upright mic'd up with some 70s shure condenser mics, running into an ex-bbc valve (tube) base board, with next to no outboard processing, their reverb was a massive spring reverb mounted in a big concrete tube to isolate it from the live room, all running to a vintage 3M 8 track, with no dolby. If that signal chain had a noise floor much below -60dB I would really be surprised. Yet the results sounded magical, and the house engineer recorded a number of hits, that are really well regarded as top class recordings. The same guy, and much of the same equipment moved to a new facility in the 80s and there recorded both Portishead and Massive Attack's early albums.
Just another perspective...
|
|
|
Post by oliv928 on Aug 31, 2010 22:53:25 GMT
woo ! this is great. i was looking for a really nice piano sound for months. thanx a lot. i will use it in Kontakt first. Can it fit to the fusion ? Did anyone make a fusion program or bank of it ? cheers all
|
|
falcon
Junior Member
Posts: 130
|
Post by falcon on Sept 1, 2010 17:36:09 GMT
In theory, it would be possible to get somewhere near to what the K5000 does using samples of sections of harmonics (samples from my K5000s perhaps? Hmmm!) but it would be lacking in some respects. Using individual samples of sine waves and spreading them across multiple patches in mix mode would indeed be possible with the Fusion and it may well be capable of far more that the K5000s but it would be terribly labourious just to get one patch together... I know... I've tried it! Yes I do believe the amount of work to get just one patch set up this way . As for phase control in fusion sample programs: you can do this but not in a very friendly fashion. The trick is to set up an Env as a trigger pulse (0 attack, very short decay, 0 sustain) and then feed it to one of the Osc pitch inputs. You can get comb effects this way, with two Osc playing the same sample. It can also be used to define a start phase in a detuned osc flanger effect. But to get an exact 45% shift would be trial and error. And it possibly can't be made consistent over different keys, would have to try that. You have osc phase control in the fusion FM model though (EDIt: no you haven't -- I was mistaken with LFO ). And since Mix Mode was mentioned, you could in theory combine 11 mix-channels to build a 66 part additive synth. One pitfall may be that the groups of 6 particles may be out of phase, I haven't tried it, because it would take a *lot* of time to create one patch in this way. But perhaps it's a nice Frankenstein project one day. Thanks for the elaboration on the K5000. Very nice reading.
|
|
|
Post by djnorythm on Sept 2, 2010 5:33:39 GMT
I might be wrong, but I remember asking Alesis about the Fusion's sampler bit rate when it first came out and they were still active on the other forum. The Fusion saves it's sampler data as 16 bit. Everything else is 24 bit... audio tracks, fx signal path etc. Funny how they deleted that thread from August 2005.
|
|
|
Post by Shreddie on Sept 18, 2010 3:16:00 GMT
As for phase control in fusion sample programs: you can do this but not in a very friendly fashion. The trick is to set up an Env as a trigger pulse (0 attack, very short decay, 0 sustain) and then feed it to one of the Osc pitch inputs. You can get comb effects this way, with two Osc playing the same sample. It can also be used to define a start phase in a detuned osc flanger effect. But to get an exact 45% shift would be trial and error. And it possibly can't be made consistent over different keys, would have to try that. You have osc phase control in the fusion FM model though (EDIt: no you haven't -- I was mistaken with LFO ). And since Mix Mode was mentioned, you could in theory combine 11 mix-channels to build a 66 part additive synth. One pitfall may be that the groups of 6 particles may be out of phase, I haven't tried it, because it would take a *lot* of time to create one patch in this way. But perhaps it's a nice Frankenstein project one day. One day? Don't you mean one week?! Although the Fusion may well be capable of additive synthesis and may even be very good at it, the fact remains that it's not designed for it. Functions one would find useful do not exist so unless you have a great deal of time I can't see why anyone would want to do this on the Fusion. This brings us on the the K5000 series, even though they are designed for additive synthesis and have many functions to help make programming it easier and faster than it might otherwise be, it's still very time consuming in comparison to other synths. In fact, I'd go as far to say that programming a K5000 patch takes up to ten times longer than programming a patch on another synth... You can easily spend an entire day (or longer) making just one sound... That said, the results are worth it. It may take alot of effort and time to create sounds on it but as long as you have the patience it's a most rewarding synth to use. It's certainly one of my favourites anyway! Anytime! I'm glad you enjoyed reading it.
|
|
|
Post by Jesse on Dec 8, 2010 17:23:38 GMT
Any luck with the conversion psionic?
|
|
|
Post by oliv928 on Dec 13, 2010 12:38:02 GMT
I have also tried the conversion from original kontakt library but i was not lucky, and not enough qualified. I will soon try again.
|
|
|
Post by riffbeeper on Dec 13, 2010 17:01:11 GMT
I used the ESX24 sampler in Logic to play the "Free Yamaha C7/Kontakt piano and it has some crapy spoiler samples at B4 to D4( I think ). I Moved some samples around without too much chipmonking (Is that a word?) too make it better, but it doesn't seem to be a pristine, complete piano sample. I think it's purpose is to show potential customers, the basic depth and quality of their product. As far as converting it to the Fusion, I'm no help.
Riff
|
|
|
Post by Shreddie on Dec 16, 2010 9:03:20 GMT
I Moved some samples around without too much chipmonking (Is that a word?) I believe that 'munchkinisation' is the correct term... And I'm not kidding btw!
|
|
|
Post by riffbeeper on Dec 16, 2010 15:33:29 GMT
Rules Rules Rules. Thats great!
|
|