- 第 40 页 -
本帖最后由 himhui 于 16-5-2 18:38 编辑
給有興趣了解一下的朋友看看.....
I don't want to move this topic to the geeky corner: simply they are two different approaches/categories.
infinite impulse response IIR
finite impulse reponse FIR
I'll be very quick and not accurate, but at least you have a rough idea.
Normally those approaches are mixed in DSP, for example the FIR approach is used sometimes in oversampling even in plug-ins based on IIR filters. At the end of the day the system impulse response is finite or infinite
For what I know we are the only company which is adopting a FIR approach to the whole modeling process. For example you can find FIR equalizers around, but not a FIR equalizer which is also modeling harmonic distortion using a FIR technique, and dynamically. Sometimes a FIR equalizer calculates the filter at run-time, while our one is based always on sampling
Acustica is a strong believer of the FIR approach (a bit like the tube vs solid state fight).
Our approach is close to a sampler vs an analog modeled virtual instrument. I'm just trying to simplify the concept to something you already know. As you know, a sampler requires to load huge amounts of things in memory, which is not good and not very simple to do. The closest thing to our engine is a rendering engine, the one you could find in 3d videogames or virtual reality (like oculus-rift). So our tech is based on different engine releases: a bit like "id tech 3" vs "id tech 5" used in quake or doom3 videogames.
This could not be so obvious at first sight, but we use similar data structures (we have large trees to calculate in realtime, exactly like it happens in doom video game for handling the partition of the space). Same things, same solutions (while an audio sampler just can stream things from the hard drive or it caches them and it just plays the sample you want).
給有興趣了解一下的朋友看看.....
I don't want to move this topic to the geeky corner: simply they are two different approaches/categories.
infinite impulse response IIR
finite impulse reponse FIR
I'll be very quick and not accurate, but at least you have a rough idea.
Normally those approaches are mixed in DSP, for example the FIR approach is used sometimes in oversampling even in plug-ins based on IIR filters. At the end of the day the system impulse response is finite or infinite
For what I know we are the only company which is adopting a FIR approach to the whole modeling process. For example you can find FIR equalizers around, but not a FIR equalizer which is also modeling harmonic distortion using a FIR technique, and dynamically. Sometimes a FIR equalizer calculates the filter at run-time, while our one is based always on sampling
Acustica is a strong believer of the FIR approach (a bit like the tube vs solid state fight).
Our approach is close to a sampler vs an analog modeled virtual instrument. I'm just trying to simplify the concept to something you already know. As you know, a sampler requires to load huge amounts of things in memory, which is not good and not very simple to do. The closest thing to our engine is a rendering engine, the one you could find in 3d videogames or virtual reality (like oculus-rift). So our tech is based on different engine releases: a bit like "id tech 3" vs "id tech 5" used in quake or doom3 videogames.
This could not be so obvious at first sight, but we use similar data structures (we have large trees to calculate in realtime, exactly like it happens in doom video game for handling the partition of the space). Same things, same solutions (while an audio sampler just can stream things from the hard drive or it caches them and it just plays the sample you want).
himhui 发表于 16-5-7 23:11
據說下周有機會.....先聽著吧, 否則會睡不著
下周就会来？这么快？？？ 看来下一周都要失眠了...
，提示我说下载有问题换个浏览器试试，我去重新装个浏览器～有些时候苹果就是麻烦～
sdzqbzx 发表于 16-5-10 19:50
studio one我装了redEQ，显示授权无效，studio one直接强退，logic也不行。。。我去找找原因
要先登記的, 看這裡
http://www.audiobar.net/forum.ph ... age%3D1&page=27
观众反应