So whats next?

[MaestroX]

Whats next for samples? Every year developers make new sample libraries larger and larger, with more and more articulations. But when will it end, or what will be the next new thing? How far can sampling really go?

[Silhouette]

Well it probably will never end, at the end of the day the people selling samples want to make some money, so they're going to keep missing articulations out so they can add new ones the next time they upgrade to make more money! If they run out of things to upgrade they'll probably just make up some articulations or some weird dynamic!! hehe If I was in their position i'd do the same, more profit that way, but its still really annoying! :?

[VicDiesel]

Search for the "synful orchestra". A clever programming job with some samples, but a lot of synthesis to get attack and transitions between notes right. Some of the demos are very impressive.

Victor.

[Thorolf]

The next step in sampling will occur when MIDI is succeded with a system that allows more fine tuned dynamics (more than MIDI’s 127 steps) and more control over the tone after start. MIDI is very limited when it comes to doing all the stuff that non-keyboardists do all the time: Adding guitar bend and vibrato, changing tone quality over time with crescendi/diminuendi (strings & winds), on the spot flexible portamento time.

There are successors to MIDI who can do all that MIDI can’t. Yamaha’s mLAN and OpenSound Conrol is the most important of these. When sample manufacturers start to tailormake their samples for these platforms, it will surely be the next step for sampling.

As for current MIDI sampling, I say it has already reached its zenith in all important ways.



Regards

[MaestroX]

Originally Posted by VicDiesel

Search for the "synful orchestra". A clever programming job with some samples, but a lot of synthesis to get attack and transitions between notes right. Some of the demos are very impressive.

I got the demo of synful orchestra but I could never get a decent sound out of it. I must be doing somthing wrong cause as you said they are impressive.

[Mrsix]

I think as long as you lay down orchestral tracks in a sequencer in a realistic manner, and then heavily process the files by EQing, etc.. you can achieve some decent results without samples, etc... my two bobs!

[MaestroX]

Its alot of effort though. There are loads of products that claim to be an orchestra out of the box. Its never the case. has anyoone ever used a "orchestra out the box" software and it actually does what it says?

[reith]

I tend to think of even moderately complete orchestral samples as a holy grail job. Quite aside from the space required, the amount of work to use them them becomes astronomic.
But does anyone want to capture every nuance of a live orchestra? I find them useful as a set of timbres that sound more human/less sterile than synthesisers whether actual hardware or Vsti, because they were originally produced by people on acoustic instruments which inherently contain imperfections.

Samples may be fine for some music but we're miles off producing the Rite of Spring, Daphnis and Chloe - or even Beethoven's 5th. I found a demo of Debussy's Syrinx on the VSL site and...no further comment!

[gravehill]

Originally Posted by Thorolf

The next step in sampling will occur when MIDI is succeded with a system that allows more fine tuned dynamics (more than MIDI’s 127 steps) and more control over the tone after start.

This already exists. It's called OSC and it's a protocol aimed at replacing MIDI. The only problem is that it won't get widely used overnight, if ever. There's just too much gear equipped with MIDI on the market. I suppose (well, hope at least) that nect generation of hard and software will support both MIDI and OSC. After that we'll either move 100% to OSC or alternatively it will die away due to lack of support. We'll see what will happen.

On the actual topic, I'd suspect that the next stage is to dynamically control different articulations. Also I guess eventually pitch wheel will be used more for CC control as outright pitchbends are usually not that useful/realistic sounding in the context of samples.

I suspect that the commercial pro-level sample libraries will still continue getting bigger and bigger for 2-3 years. At that point, around 2010 or so, physical modelling algorithms should start to be good enough that they'll start to replace sample libraries. Thus the instruments themselves will get smaller in terms of harddisk needs but will require LOTS more processing power than is available for a mediocre pro user today. OTOH, at that point 20-30GHz processors should be in wide use already.

Or something.

[tawei82]

any reverse cymbal sample that is very good?
I find it difficult to search for a more natural sound.
string esemble is OK. but for only 1 orchestra instrument, midi seems not enough real.