What's new

Literature, Scientific Papers and Analysis of Virtual Instruments

I fear this too and as far as release dates go, I’m creating a chronology as I go and may publish that as an appendix. (I’ll likely put it online as well but it has the same issue of likely disappearing.) The release dates for major libraries at least have a lot of redundancy, with reviews, press releases, and chatter on forums and social media that you can usually track it down even if the companies themselves are coy about the information, but each library takes googling and sorting through the hits, which is fine until you hit a library with a common name that makes googling challenging.


Its not that different than releases for big instrument manufacturers. NAMM events, AES, etc events or close to those industry specific where there will be more eyes on new release news.
Also Per business quarters and for some odd reason Tuesdays at 11am local times. But I've seen Friday work very well for the unfinished and a few others. One month intro sale. Or its "announcement tusdays build on wed and release on thursd.
Also stupid mail chimp newsletter reminding you every few days its the last week, last day, last 2 hours of the intro sale... untill.. hmmm... its extended... (surprise! :cautious:)
I think you follow sample logics history of releases itll become more aparent. Here is even a nice chart showing when they've made big releases over the past 15years
Its skews heavily on the start of the second quarter and late sept or october.
And no new prodcuts during black friday or holiday season as thats for mega sales and no one expect full price. Plus the intro sale for fall can be extended.
Summer is a dead zone in europe. We all forget europeans have like 2 months off or soemthing crazy like that.
But most importantly...

Is the library done?

cuz fuck man

it takes FOREVER!

PD: if you are really looking into releasing a book maybe reach out to maufaturers for discount codes. Affiliate links etc to make some dough or cross promote the book.
 
Hey again guys,
I'm currently in the middle of writing my bachelors thesis on virtual instruments and I'm wondering if anybody here is willing to give feedback as an expert with experience either on the theoretical or practical side of the topic.

It is a rather practical approach to a bachelor thesis with a mixed method, as I am constructing a prototype of a virtual instrument on the basis of theoretical discussions.
Theoretical frameworks on realism and expressivity are discussed in order to create a realistic and expressive virtual violin instrument.

1. I am hoping to find someone who would give me practical feedback on the virtual instrument itself:
- the script
- the sample order
- the user friendlinessetc.

2. I am hoping to find someone who can give me feedback on a scientific level

I know everybody's time is valuable and I can't expect help without anything in return, but I'm sure we can find a way to make the project beneficial for everybody involved.

Time frame: Now (June) until august.

All the best,
Sebastian

The thesis is constructed more or less like this:

1. Theoretical Frameworks
1.1 Realism
1.2 Hyperrealism (as coined by Cassannelles)
1.3. Emotion and Music
1.4 Expressivity through perfomance
1.4.3. expressvity and the violin
1.4.4 gradual use of techniques
1.5 Expressivity through production
1.5.1 Timbre
1.6 Discussion of sampling-synthesis and physical modeling
1.6.1 theoretical approaches to sampling synthesis
1.7 dimensions of a virtual instrument
1.8 sampler
1.9 aesthetic context
1.9.1 Neo-classical aestehtics of string instruments
1.9.2 analysis of "only the winds" by Olafur Arnald

2. Analysis of virtual string instruments in a Neo-classical context
2.1 Olafur Arnalds Evolution - Spitfire Audio
2.2. Tina Guo Cello - Cinesamples

3. Method
3.1 Theoretical derivation
3.2 Analysis
3.3 Prototyping
3.4 Evaluation

4. Prototype development
4.1 first iteration
4.1 first evaluation
4.2 Second iteration

5. Final Evaluation

6. Discussion

7. Conclusion
 
Indeed. I am a mere Universitair Docent (university lecturer). Internationally that’s an assistant-professor. I usually just call myself teacher. I did not intend to brag or anything. Actually, on the contrary… I only mentioned it because I figured it could help the OP gauge my credentials / experience.
 
Indeed. I am a mere Universitair Docent (university lecturer). Internationally that’s an assistant-professor. I usually just call myself teacher. I did not intend to brag or anything. Actually, on the contrary… I only mentioned it because I figured it could help the OP gauge my credentials / experience.
It is a very generous offer indeed.
 
It's not polite to speak loosely about the professional titles of academics. In Doc's case, maybe he doesn't like to brag; plus there are a whole different set of titles in Dutch.
Hmm I always thought Doctor came from the medical term not so much as an academic title. I mean, not that I think @doctoremmet is an actual medical man or anything.

I never thought he was bragging either...

I guess my stupid joke was interpreted very differently than what I intended. I'm sorry Temme. I will delete my post.
 
Hmm I always thought Doctor came from the medical term not so much as an academic title. I mean, not that I think @doctoremmet is an actual medical man or anything.

I never thought he was bragging either...

I guess my stupid joke was interpreted very differently than what I intended. I'm sorry Temme. I will delete my post.
I took it as a joke! No need to delete anything!
 
Hey,

I didn't really know which category was the right to post this thread, so I'm just going with the Welcome to VI one.

I'm currently doing a lot of research on the matter of Virtual Instruments for my bachelor thesis, and what I can find is very limited or outdated.
There is lots of literature on analogue or software synthesizers, most a bit outdated, but overall very little I can find regarding sampled, hybrid and virtual orchestral instruments.

Does anyone know any literature or good databases for extensive research?
Hope to hear from you guys.

All the best,
Sebastian
To be honest, very little progress has been made in the last 20 years regarding virtual instruments underlying technologies. (I still maintain that my Korg Z1 was the most advanced synth I ever owned, and I currently own a Kronos ... ) Just because, say MODO drum and bass didn't exist 20 years ago doesn't mean that their underlying code didn't [in this actual example, it was very old code released to the public, by steinberg, that very few were making use of, that a company [IK] decided to finally make use of. That's how this typically works].

Companies decide to pay the programmers for the work to actually do something (turn into a product) with innovations that stopped short of having their papers published -- or they don't, and that's how you get thousands and thousands of different kontakt libraries instead of innovation.

I've read physical modeling papers going back to the 60's, they're literally written on typewriters, and I can guarantee you way more people were active in the field of trailblazing on that side of the 20th century. In my opinion you'd have an easier time writing about how much progress has stagnated since the late 90's on anything 'remotely' novel in the field of digital audio and audio DSP.

I seriously challenge anyone to show examples of any modern trailblazers today in terms of DSP and the innovation of processes which have already for years now been far out of the uncanny valley.

The cost of getting away from samples in terms of research and development is huge, and none of these audio companies really want to foot the bill for that level of R&D.
 
Last edited:
To be honest, very little progress has been made in the last 20 years regarding virtual instruments underlying technologies. (I still maintain that my Korg Z1 was the most advanced synth I ever owned, and I currently own a Kronos ... ) Just because, say MODO drum and bass didn't exist 20 years ago doesn't mean that their underlying code didn't [in this actual example, it was very old code released to the public, by steinberg, that very few were making use of, that a company [IK] decided to finally make use of. That's how this typically works].

Companies decide to pay the programmers for the work to actually do something (turn into a product) with innovations that stopped short of having their papers published -- or they don't, and that's how you get thousands and thousands of different kontakt libraries instead of innovation.

I've read physical modeling papers going back to the 60's, they're literally written on typewriters, and I can guarantee you way more people were active in the field of trailblazing on that side of the 20th century. In my opinion you'd have an easier time writing about how much progress has stagnated since the late 90's on anything 'remotely' novel in the field of digital audio and audio DSP.

I seriously challenge anyone to show examples of any modern trailblazers today in terms of DSP and the innovation of processes which have already for years now been far out of the uncanny valley.

The cost of getting away from samples in terms of research and development is huge, and none of these audio companies really want to foot the bill for that level of R&D.
Maybe another factor is academic institutions don't want to license their stuff?

For example I think UVI is the only private company using stuff from IRCAM like their timestretching. Pretty much all other companies are using the Elastique Pro algo by Zplane.
 
Last edited:
I'm learning percussive sound synthesis, particularly the FM synthesis. These 2 papers by the same authors have helped me a lot. They expained how FM synthesis works so well and proposed an improved model called loopback FM. I'm trying to implement it in python.

 
Last edited:
I'm learning percussive sound synthesis, particularly the FM synthesis. These 2 papers by the same authors have helped me a lot. They expained how FM synthesis works so well and proposed an improved model called loopback FM. I'm trying to implement it in python.

Why is python your language of choice for implementing this? It's really not tailored to synthesis or realtime audio. I recommend C++ and JUCE for anything to do with realtime audio dsp and getting off the ground quickly.
 
Why is python your language of choice for implementing this? It's really not tailored to synthesis or realtime audio. I recommend C++ and JUCE for anything to do with realtime audio dsp and getting off the ground quickly.
Well, I'm not a computer student and I learn python because my accounting professor needed me to do so. Thanks for your recommendation. So I want to learn C++ when I have more time in the future.

I've uploaded my python implementation code that can run now on github.

 
Last edited:
Hey,

I didn't really know which category was the right to post this thread, so I'm just going with the Welcome to VI one.

I'm currently doing a lot of research on the matter of Virtual Instruments for my bachelor thesis, and what I can find is very limited or outdated.
There is lots of literature on analogue or software synthesizers, most a bit outdated, but overall very little I can find regarding sampled, hybrid and virtual orchestral instruments.

Does anyone know any literature or good databases for extensive research?
Hope to hear from you guys.

All the best,
Sebastian
The literature may be outdated, but are books and databases really all you need? There are some excellent producers here with a lot of experience. Some of them started long time ago and dealt with the first synthesizers. You could benefit from their knowledge.

Just an idea.
 
Well, I'm not a computer student and I learn python because my accounting professor needed me to do so. Thanks for your recommendation. So I want to learn C++ when I have more time in the future.

I've uploaded my python implementation code that can run now on github.

probably a good idea to check out some of the speakers at their conference. it might give some ideas.

 
To be honest, very little progress has been made in the last 20 years regarding virtual instruments underlying technologies. (I still maintain that my Korg Z1 was the most advanced synth I ever owned, and I currently own a Kronos ... ) Just because, say MODO drum and bass didn't exist 20 years ago doesn't mean that their underlying code didn't [in this actual example, it was very old code released to the public, by steinberg, that very few were making use of, that a company [IK] decided to finally make use of. That's how this typically works].

Companies decide to pay the programmers for the work to actually do something (turn into a product) with innovations that stopped short of having their papers published -- or they don't, and that's how you get thousands and thousands of different kontakt libraries instead of innovation.

I've read physical modeling papers going back to the 60's, they're literally written on typewriters, and I can guarantee you way more people were active in the field of trailblazing on that side of the 20th century. In my opinion you'd have an easier time writing about how much progress has stagnated since the late 90's on anything 'remotely' novel in the field of digital audio and audio DSP.

I seriously challenge anyone to show examples of any modern trailblazers today in terms of DSP and the innovation of processes which have already for years now been far out of the uncanny valley.

The cost of getting away from samples in terms of research and development is huge, and none of these audio companies really want to foot the bill for that level of R&D.
Hear, hear.
 
Top Bottom