All times are UTC+01:00




Post new topic  Reply to topic  [ 133 posts ]  Go to page Previous 110 11 12 13 14 Next
Author Message
PostPosted: Tue Apr 06, 2021 5:32 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
Quote:
Quote:
Quote:

Ok, that's the complete opposite of what I see. Can you send me your settings and test audio?
sent.
Ok, issue confirmed. And I know the cause: "BAD!" was designed to improved IMD for bass tones, which is at least 95% of all the bass in music, and it works very well on that. But on random bass (bass noise, basically), it makes things slightly worse. (The improvements on bass tones is still much bigger than the detrimental effect on bass noise). I'll see what I can do to improve this, but even with this, you really should still keep it turned on for music.
I think I found a good compromise for the behavior. I've added a setting for it ("Allow distortion on thunder bass (threshold)"), which I'm going to hide in Extreme Tweaker mode. Basically, we do indeed want this filter ("BAD!") to be active on bass tones and not on bass noise ("thunder"), because in the latter case the IMD that's caused by this filter is actually worse than the IMD it's cleaning up. It looks like highs IMD is mainly annoying when it has a pattern in it. "0" is equal to the old behavior, "50" (the maximum) is getting close to turning "BAD!" off. 30 seems to be a good compromise, I haven't been able to detect any added distortion in music so far and the distortion added to pink noise is largely gone. With higher values (50) I do have the impression that there's more distortion - I haven't done a blind test though. The default is currently set to 25, 30 is probably better but I think the risk of adding IMD in music is bigger than any gains from this filter, so without testing a lot more I want to err on the safe side.

On average, a bit more highs get through with this setting > 0 (nearly no effect on other frequencies), which is probably a good thing as long as those extra highs aren't distorted.

I'll run a new build tonight.


Edit: At 50 there's clearly more distortion. 25, 30 and 35 seem to be ok. I'm not too sure when I test it (even in a loop, I think I hear different things every time even with the same settings), but I measured it and - at least for the music that I tested with, 35 seems to be still ok. 50 clearly isn't, both based on listening tests and measurements.


Top
   
PostPosted: Wed Apr 07, 2021 8:07 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
BETA012 posted. There's a new slider, "Allow distortion on thunder bass (threshold)", probably not a great name for what it does... But that setting controls how strongly it tries to reduce IMD, which is bad for random (thunder) bass sounds, good for tones. 50 is bad. 30 and 35 are probably still ok, but I'm not really sure - see my previous post. 0 is old behavior ("BAD!" on in the previous few beta's).


Top
   
PostPosted: Wed Apr 07, 2021 3:10 pm 
User avatar

Joined: Wed Nov 08, 2017 3:16 pm
Posts: 191
Quote:
BETA012 posted. There's a new slider, "Allow distortion on thunder bass (threshold)", probably not a great name for what it does... But that setting controls how strongly it tries to reduce IMD, which is bad for random (thunder) bass sounds, good for tones. 50 is bad. 30 and 35 are probably still ok, but I'm not really sure - see my previous post. 0 is old behavior ("BAD!" on in the previous few beta's).
maybe its just me but i prefer to use white noise to calibrate ST for music since its flat freqs all across.
can you imagine if a different type of white noise was used that emphasized upper freqs instead of lower freqs ?


Top
   
PostPosted: Wed Apr 07, 2021 3:27 pm 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4151
Quote:
Quote:
BETA012 posted. There's a new slider, "Allow distortion on thunder bass (threshold)", probably not a great name for what it does... But that setting controls how strongly it tries to reduce IMD, which is bad for random (thunder) bass sounds, good for tones. 50 is bad. 30 and 35 are probably still ok, but I'm not really sure - see my previous post. 0 is old behavior ("BAD!" on in the previous few beta's).
maybe its just me but i prefer to use white noise to calibrate ST for music since its flat freqs all across.
can you imagine if a different type of white noise was used that emphasized upper freqs instead of lower freqs ?
I use it too, but also many other tones like sine sweep.

_________________
control point
control point2


Top
   
PostPosted: Wed Apr 07, 2021 4:02 pm 
User avatar

Joined: Wed Nov 08, 2017 3:16 pm
Posts: 191
Quote:
Quote:
Quote:
BETA012 posted. There's a new slider, "Allow distortion on thunder bass (threshold)", probably not a great name for what it does... But that setting controls how strongly it tries to reduce IMD, which is bad for random (thunder) bass sounds, good for tones. 50 is bad. 30 and 35 are probably still ok, but I'm not really sure - see my previous post. 0 is old behavior ("BAD!" on in the previous few beta's).
maybe its just me but i prefer to use white noise to calibrate ST for music since its flat freqs all across.
can you imagine if a different type of white noise was used that emphasized upper freqs instead of lower freqs ?
I use it too, but also many other tones like sine sweep.
the problem is, music is completely random freqs at different amplitudes and waveform shapes.
i have different methods of testing and calibrating, such as injecting a sinewave based tone at the frequencies the MB is set to with white noise at the same dB level and observing the drops at certain frequencies and their bandwidth occupied by the drop.
the same type of testing applies to pre-emphasis, to hear no audible drop differences with pre-emphasis turned on/off but not applied on the output (pre-emph enabled but not to output).
so turning it on and off should produce virtually no difference, thats when pre-emph is calibrated properly, it produces the emphasis needed for AM/FM but without any push-downs on the lows from the mids/highs.


Top
   
PostPosted: Wed Apr 07, 2021 8:08 pm 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4151
Quote:
Quote:
Quote:

maybe its just me but i prefer to use white noise to calibrate ST for music since its flat freqs all across.
can you imagine if a different type of white noise was used that emphasized upper freqs instead of lower freqs ?
I use it too, but also many other tones like sine sweep.
the problem is, music is completely random freqs at different amplitudes and waveform shapes.
i have different methods of testing and calibrating, such as injecting a sinewave based tone at the frequencies the MB is set to with white noise at the same dB level and observing the drops at certain frequencies and their bandwidth occupied by the drop.
the same type of testing applies to pre-emphasis, to hear no audible drop differences with pre-emphasis turned on/off but not applied on the output (pre-emph enabled but not to output).
so turning it on and off should produce virtually no difference, thats when pre-emph is calibrated properly, it produces the emphasis needed for AM/FM but without any push-downs on the lows from the mids/highs.
That's nice. However how about areas between bands? :)

_________________
control point
control point2


Top
   
PostPosted: Thu Apr 08, 2021 12:25 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11185
BETA012:
Mac build that asks for microphone access uploaded
WatchCat build with fix for beeps in unregistered version uploaded


Top
   
PostPosted: Thu Apr 08, 2021 3:27 am 
User avatar

Joined: Wed Nov 08, 2017 3:16 pm
Posts: 191
Quote:
Quote:
Quote:

I use it too, but also many other tones like sine sweep.
the problem is, music is completely random freqs at different amplitudes and waveform shapes.
i have different methods of testing and calibrating, such as injecting a sinewave based tone at the frequencies the MB is set to with white noise at the same dB level and observing the drops at certain frequencies and their bandwidth occupied by the drop.
the same type of testing applies to pre-emphasis, to hear no audible drop differences with pre-emphasis turned on/off but not applied on the output (pre-emph enabled but not to output).
so turning it on and off should produce virtually no difference, thats when pre-emph is calibrated properly, it produces the emphasis needed for AM/FM but without any push-downs on the lows from the mids/highs.
That's nice. However how about areas between bands? :)
well that too of course but it depends on the level of slope for each band and how many bands are being used and the respective levels of them (as well as other additional parameters).
my comment was more or less "generalized" to a target regarding calibrating ST and the use of pre-emph.
id rather not make it an "off topic" discussion in this thread though.


Top
   
PostPosted: Thu Apr 08, 2021 1:59 pm 
User avatar

Joined: Sun Dec 23, 2018 7:44 pm
Posts: 792
Location: Texas, USA
Watchcat seems to be functioning properly now.
Also, thanks for making the input gain of the plugin menu work when bypassed. 8-)


Top
   
PostPosted: Fri Apr 09, 2021 3:19 am 
User avatar

Joined: Tue Mar 17, 2009 2:56 pm
Posts: 4151
Something is wrong with "Soft Clipping" per MB bands.
I set it to +35dB, so basicly should do nothing, but it rises band level by around 1dB.
--
Some strange bug when changing CPU quality. Claimbigf: In Use!

_________________
control point
control point2


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 133 posts ]  Go to page Previous 110 11 12 13 14 Next

All times are UTC+01:00


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Limited