Hi guys, the question is like this :
Lets say I have a VSTi connected to a VST effect connected to the master.
Like this :
VSTi ---> VST ---> Master
The question is this , lets say that the VSTi transfers to the VST an audio which is more than 0db so had it been connected directly to the master clipping should have occured. Now, lets say that the VST simply reduced the input it receives from VSTi so if the VSTi sends +3db to the VST effect and the VST input gain is set to -3 the effect will send 0db to the master and *apparently* no clipping and no data loss have occured.
Thats what I thought until now... but I started learning abit VST programming using the VST sdk and read that the data signal a VSTi outputs is a float number... a number between -1 and 1 . I Assume clipping is a range which isn't in the range of -1 and 1 ?
if thats the case there's data loss between the send of the VSTi and the VST because the VSTi sends +3db...
I think I'm wrong about this because had that been the case there would have been no logic to limiters if there's always data loss.
Where am I wrong ?

its too early you fried my brain

Quote:

its too early you fried my brain

Nope, no clipping... because all audio is handled as floating-point numbers. That means that internally, audio can go above 0dBFS without clipping. Just make sure it's legal when it hits your DAC.

Forever,




Kim.

Quote:

Nope, no clipping... because all audio is handled as floating-point numbers. That means that internally, audio can go above 0dBFS without clipping. Just make sure it's legal when it hits your DAC. Forever, Kim.

So what you're saying is that db above 0 got a digital presentation and when converted to analog they dont have a presentation so analog causes the clipping ?

Anyone ?

Scr1pt3r wrote:

So what you're saying is that db above 0 got a digital presentation and when converted to analog they dont have a presentation so analog causes the clipping ?

Something like that, though it's not "the analogue" that causes clipping - it's the conversion process from floatingpoint->integer (soundcard drivers only accept integer signals, primarily because signals >0dBFS can't be recreated post DAC).

Forever,




Kim.

vurt wrote:

its too early you fried my brain

me too, i just got up 10 minutes ago