The other thing to bare in mind is that samplers will become more clever. We talk a lot about the 2gig limit. Personally, I believe with clever programming a lazy sample loading, that limit would be largely not a problem anyway.
Halion (my choice) already has pre-load down to 0.2 seconds, which means I can quite happily run all my orchestral samples in real-time on a single pentium 4 machine. As disk/memory becomes more efficient, that pre-load will drop. Eventually the size of the samples will become largely irrelevant, especially if samplers can just pre-load the first time that a sample is used, thus avoiding loading unused samples.
That said, the future is not in sampling... but in synthesising the samples. It has to be. We are already seeing successful results in modelling older synths. The same will happen for orchestral instruments. Synful is just the start. It is important to bare in mind that synful still has samples at its heart too... it just used them differently.

@paynterr said:

Halion (my choice) already has pre-load down to 0.2 seconds, which means I can quite happily run all my orchestral samples in real-time on a single pentium 4 machine.

I think you are quite right in what you say about some things, however you have to remember that whilst you may be able to run all your orchestral samples on one machine, you certainly couldn't get close to running all mine.

DG

I don't know - lower the pre-load from 0.2 seconds to 0.0001 seconds and suddenly you have 1000s of gigs of samples available.
The real challenge for sampler developers is to see how low they can get the pre-load. Of course, hardware manufactuers will do a lot of their job for them.
RAM is just fast disk space. In an ideal world, we wouldn't need RAM at all. RAM and Disk would be the same thing. As they will eventually be. We are getting there quite quickly if you look at memory sticks for cameras and the like, which are already exceeding 1gig.
I would like to see samplers with 0.0 seconds pre-load right now and take the hit of the slightly latency in pulling the samples from disk.
This would allow:
a) My projects to load immediately.
b) A huge orchestral template.
with the downside:
c) I could not play in real-time...
I could live with this, since invariably I draw my projects with a mouse rather than play them in...

Just wanted to thank Mike for the original post that started this thread. An excellent, carefully phrased question, which prompted this great discussion about the future of the tools we work with. I'm sure VSL is enjoying this discussion, and possibly benefiting from the pool of ideas.

I agree with that Hoff, and wonder about the far future of "sampling" or "synthesis." When the ideal system evolves, freed from technical limitations, what will it be? Obviously some perfected emulation or original presentation of any timber including orchestral, with no effort on the part of a composer other than a purely artistic one. Which is as it should be to begin with but isn't.

Thank you, Hoff and William.

I have learnt a great deal from this forum - we are fortunate indeed to have such a depth of experienced musicians who take the time to share their knowledge.

Regards,
Mike.

@paynterr said:

As they will eventually be. We are getting there quite quickly if you look at memory sticks for cameras and the like, which are already exceeding 1gig.

I'm just quoting this, because I hope you know there's a difference between the RAM in your computer and the memory stick in cameras. These are different memory technologies with different capabilities.

However there are new developments made as always. Not only the size gets bigger but also the RAM is possible to remember it's data without power (MRAM technology). That means e.g. no booting of anything at all anymore. Power the computer on, and whoops your template is loaded. Not hitting the market soon though as it seems.

What are we getting if technology limits aren't there anymore? A thing like the founders of GM Midi have dreamed of: Pass on little (midi) files and get a CD quality recording instantly...

PolarBear

@paynterr said:

I would like to see samplers with 0.0 seconds pre-load right now and take the hit of the slightly latency in pulling the samples from disk.
This would allow:
a) My projects to load immediately.
b) A huge orchestral template.
with the downside:
c) I could not play in real-time...
I could live with this, since invariably I draw my projects with a mouse rather than play them in...

And that downside becomes irrelevant to those who work from notation.
But even for sequencer folk, the ability to have virtually limitless loading capacity would be to big a deal to pass up... besides, just have one channel with low latency playing capacity and the problem disappears.

I think it's great that Synful has taken a different approach from VSL for their transitions. Isn't that how innovation takes place - with competing concepts and techniques? Synful is a lot more than marketing.

Some people like William take "purity" of samples as the only measure. WHich I find hilarious because only a few years ago samples were villified as fake and artificial rip offs of real musicians and the threat to all that was good and pure. But that is his taste and having heard his demos on this board, who can argue that it works for him? Not me!

Having used synful a bit, I do see that there are timbral issues that don't sound as natural. But the point of the plugin is to do what it does well, not what it doesn't. The runs and legatos are quite wonderful. I don't know if future instruments will have carrier waves or be purely sample based, but more power to anyone who will further a particular technology to see where it leads.

And as always, the attitude of Dietz on this board about their competition is so fantastic that these discussions only enhance their company's appeal, imho.

By sheer effort and torture I've been developing a program for sample selection (yes, some may remember me saying this as long as a year ago... still working...). It takes a 1-second (actually, user definable) delay to analyze and buffer samples. The program basically works, but it's buggy. It does however, use very little memory, and instruments load almost immediately. And, it uses pure VSL, with no messing about!

Beyond that, I had a thought about Synful the other day... I think that, because it uses a highly "fragmented" sample source, perhaps part of the "synthy" sound of it has more to do with the resonances of the instrument's body being chopped up... After all, hasn't just about everyone who's tried it noticed that it sounds WAAAAY better with even the smallest bit of reverb? That is, it sounds different in a very essential way, with reverb -- VSL sounds better, but not entirely different...

So, I was thinking that what the developer should do is incorporate resonance models (IR-based) into the program itself, selecting the resonance model with the instrument. Just a thought... in case he's reading!

Otherwise, for the end-user, it might make sense to try creating a violin-body IR (in Altiverb, or whatever) at home, then applying that to Synful first, at 100% "wet".

This could make Synful sound much better, even for "dry" playback (yeah, pseudo-dry, but you know what I mean...)

J.

This sounds very very interesting... tell me more... I am also an IT architect/developer in my day job, so would be very interested to hear more about this... this sort of intelligent tool is the way ahead I feel.
Does it work on a PC? Can I try it out?
Regards

I have a question for you technological wizards.

Have you ever considered building a small, simple, multiple use, EXS player that can be used outside Logic, either as a VST/AU instrument, or standalone?

I'm fairly sure there would be a lot of interest.

Regards,


Alex.