• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Some questions about EQ and impulse response with foobar2000 foo_convolve and Equalizer APO

jaytrinitron

Member
Joined
Jan 27, 2018
Messages
40
Likes
4
So I've been messing about with equalizer APO and foobar2000. I am using a Schiit Modi 3+ DAC and a JDS Labs Atom amp, and I have a dt990 pro 250ohm, a dt990 special edition 600ohm (basically a dt990 premium 600 ohm), and a Senn HD600. I had some questions about how all the processing works:
1. With foobar2000's foo_convolve plugin, does it matter what the sample rate and bit depth of the impulse response file is? The way I understand it, the sample rate has to match (44.1khz for 44.1khz music, which is the way I have it setup now. And for music that isn't 44.1khz, I resample to 44.1 which I'm fine with). But with bit depth, does it matter whether the impulse response is 16 bit int, 24 bit int, or 32 bit float, or something else? I'm particularly confused about this because my foobar2000 (using the foo_wasapi plugin) is set to output over wasapi exclusive 32 bit int to my DAC (the highest bit depth available for my DAC) which is what a thread here recommended. But the way I understand it, foobar2000 processing is in 32 bit float, and my input files are 16bit int and sometimes very rarely 24 bit int, so I'm not sure what to pick.
2. I don't really understand what it is, but when applying a graphic eq with equalizerAPO or an impulse response eq with foo_convolve, do I need to be worried about pre-ringing? If I do, is there anything I can do to mitigate it?
3. Finally, I occasionally output 44.1khz audio to windows, and my windows sample rate is set to 48khz. From what I understand, when this gets upsampled to 48khz any artefacts should be far below the range of audibility, even with the windows audio stack. Is this true?
4. I've been trying to make sense of the thread here and I'm doing everything except: setting a preamp gain of -4db (I don't hear any clipping or issues with the music I tested, and if I do I can just enable the preamp then) and disabling windows sounds and enhancements (I believe this is required for equalizer APO to work). I think I should be fine, unless I am mussing something major? Then again, I don't think I've ever heard clipping in my stack even before I got a separate DAC and amp or used equalizerAPO, except for a few rare instances where the tracks were pre-clipped.
FWIW, I did add -2dB in foobar2000, because that doesn't seem as significant a loss in terms of volume and because apparently my AKM dac does have 2dB overhead already. Not sure if I should add a preamp gain of -2dB in equalizer APO or not, since I've never heard any issues anyway with intersampling or what not.
 
Last edited:

Patrick1958

Senior Member
Forum Donor
Joined
Jun 28, 2018
Messages
498
Likes
412
Location
Belgium
4. I've been trying to make sense of the thread here and I'm doing everything except: setting a preamp gain of -4db (I don't hear any clipping or issues with the music I tested, and if I do I can just enable the preamp then) and disabling windows sounds and enhancements (I believe this is required for equalizer APO to work). I think I should be fine, unless I am mussing something major? Then again, I don't think I've ever heard clipping in my stack even before I got a separate DAC and amp or used equalizerAPO, except for a few rare instances where the tracks were pre-clipped.
FWIW, I did add -2dB in foobar2000, because that doesn't seem as significant a loss in terms of volume and because apparently my AKM dac does have 2dB overhead already. Not sure if I should add a preamp gain of -2dB in equalizer APO or not, since I've never heard any issues anyway with intersampling or what not.

Should you have the song from Thomas Dolby - I scare myself, play it with and without (negative pregain). If you don't hear clipping/distortions you should be fine, if you do, adjust the pregain until clipping is no longer audible and use that setting in future settings.
ps : by default convolution files have an -6 dB pregain embedded.
 
OP
J

jaytrinitron

Member
Joined
Jan 27, 2018
Messages
40
Likes
4
Thank you for your reply. I tested the song (good song btw), but I used 320 vorbis from spotify premium. Not sure how much of a difference that makes, especially with things like intersample distortions and such, but I didn't hear any clipping/distortions as far as I know.
ps : by default convolution files have an -6 dB pregain embedded.
Interesting, how do you know this? Although it makes sense because my level appears to drop by about that much in foobar2000 when using an impulse response file.
 

Patrick1958

Senior Member
Forum Donor
Joined
Jun 28, 2018
Messages
498
Likes
412
Location
Belgium
Thank you for your reply. I tested the song (good song btw), but I used 320 vorbis from spotify premium. Not sure how much of a difference that makes, especially with things like intersample distortions and such, but I didn't hear any clipping/distortions as far as I know.

Interesting, how do you know this? Although it makes sense because my level appears to drop by about that much in foobar2000 when using an impulse response file.
Wiki Jriver.
 
OP
J

jaytrinitron

Member
Joined
Jan 27, 2018
Messages
40
Likes
4
I see, thanks. But all I could find there is that "normalizing filter volume" in JRiver applies a fixed -6dB pregain, not that all convolution filters in any program have it. My filter is just an eq and it's in foobar2000, so I don't understand how this applies.
 
Top Bottom