BasicSynth (Anglais) Broché – 13 janvier 2009
|Neuf à partir de||Occasion à partir de|
Les clients ayant acheté cet article ont également acheté
Aucun appareil Kindle n'est requis. Téléchargez l'une des applis Kindle gratuites et commencez à lire les livres Kindle sur votre smartphone, tablette ou ordinateur.
Pour obtenir l'appli gratuite, saisissez votre adresse e-mail ou numéro de téléphone mobile.
Détails sur le produit
Dans ce livre(En savoir plus)
Commentaires en ligne
Commentaires client les plus utiles sur Amazon.com (beta)
I have not used the book to its full potential, since I'm still a novice with both the equations and concepts of actual DSP as well as programming. However it has helped tremendously to understand some of the basic concepts of DSP without unnecessary jargon as well as given me a clearer look into how a computer program actually represents some of the seemingly complex equations.
The later chapters can get extremely complex in subject, but keep in mind this book is not meant as a replacement for DSP books, but rather a supplement. For a good place to start I recommend Richard G. Lyons "Understanding Digital Signal Processing".
One of the really great aspects of the book is that the code is presented in a very clear fashion such that it is easy to identify what the functions in the excerpts are doing. Not only that but frequently Mitchell provides example code that is a literal representation of the equation, followed often by multiple versions of that same code in some simplified fashion. This can either be code that is more computationally efficient or other times it's just presented in a way that's easier to read and understand for the would-be programmer. Since programs are not always written with diligent attention to useful comments and formatting, browsing existing code may not be that helpful.
As the other review by Dan Mitchell explains, the book covers many topics including the most common forms of computer generated sound, including FM, subtractive and additive synthesis, as well as filters and common processors like reverb and chorus. There are numerous other topics in the book that are definitely worth your while if the subject matter interests you.
My solitary complaint about the book is that it is currently only fully available in print format. While there is nothing inherently wrong with that, for a field that is as fast moving as programming software instruments and effects, it is helpful to be able to update code and concepts if necessary. In defense of the book however, most changes in C++ are minor at this point and unlikely to affect the content of the book and second the subjects the book covers are well-trodden and "old" enough that new innovations in the field will also have little or no effect on what are tried and true methods of sound synthesis.
All in all a fantastic book and make sure to get the example code at [...] when working with the book.
Since I am a system software developer trying to understand how instruments can be synthesized, I had no previous exposure, even though very advanced algorithms, math and physics understanding. This book will tell a software developer anything he needs to know ;), so it doesn't throw you in cold water. But beaware that it will only go through the basics. I read through it in a few hours and mostly its a lot of templates to generate various stuffs and mix it together. Its good to have such a bird-view feeling over topic. After that I would suggest what I am doing now, reading "Designing Sound, by Andy Farnell" which goes into the great detail of how the physics work behind the scenes and how to simulate it.
BUT, I am a bit annoyed by the outdated optimzation hints. Especially in 2008... If that was a book from 2000 OKAY! But its not.
Your compiler, especially a C++ compiler knows all the tricks. You don't need to pull out loop invariants, you don't need to do some bit-shifting ( x*=2 will actually be done as x << 1 whenever possible by any "good" compiler, so don't obfuscate your code) magic and floating point vodoo... In fact doing this will actually give you worse performance in some cases because the compiler doesn't understand what you intended to do and can't apply the "common-path" optimizations anymore. Floating point is definitly NOT slower than integers on latest processors (think of AVX, which wasn't available 2008 I think), just utilizing the full power of parallel floating point processing can be a bit convoluted at times. But switching to integer-forth-and-back-conversion-vodoo is definitely the road to slow code.
Optimize for asymptotic times and write clean code. Then the profiler will tell you WHAT and WHERE to optimize. Premature optimization is a very bad idea given the enormous advancements in compiler technology and processors.
If you're looking to create your own software synthesizer or are just interested in how a software synth works, give this a shot, I cannot recommend this book highly enough.
I recommend the book to anyone who is interested in exploring the code and concepts that are used to create digital synthesizers.