All times are UTC+02:00




Post new topic  Reply to topic  [ 707 posts ]  Go to page Previous 154 55 56 57 5871 Next
Author Message
PostPosted: Mon Aug 21, 2023 10:36 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11425
Quote:
Quote:
Just in case it is useful to anyone, here are some sample settings utilizing the new compressor. This is *not* a finished product. But it's most of the way there.

Note that I am one of the people who is just using MB1. No AGC, no MB2. As a result, you're going to frequently see the MB meters get pegged. We need a bigger UI range for that as an option now. But we've had other priorities thus far.
Hm, okay. So testing out Run the Jewels - Walking in the Snow (https://www.youtube.com/watch?v=G52lnwP8uX0) between 15 and 30 seconds, the loudness increases substantially. I'll keep looking for stress songs that sound 'off' with this new compressor.
I checked this with my own test preset, Wes' preset and some presets that use the older compressor modes. Wes' preset misbehaves a lot more than mine, most older compressor presets that I tested were clearly a lot worse - and you can also hear them move a lot for example when the hi-hats kick in just before 0:15.

My own test preset was never made to sound good, just to test the behavior of the compressor, so I'm a bit hesitant to post it. But the basic settings are: Attack 200, release 2000, release speedup set to 4x (the behavior is clearly better at 2.5x), caspergate at -12 (default), Intellilimiter at +12 dB.

Quote:
Just brainstorming here. Would it be reasonable to include a Threshold Offset slider in the new compressor architecture (so we can adjust the overall threshold as extremely as 12 db below/above what each band is set at)? BreakawayOne (and presumably Omnia tech) has a type of slider that adjusts multiple sliders at a time and returns to the center as you let go of the mouse button, adjusting each slider by however many db you tried to offset by. I'll capture video of it because it's hard to explain: https://imgur.com/nMoT8Ae

Image
What would the goal of this be? We already have "Density drive" which basically adjusts both the thresholds and the injection levels to go deeper or less deep into the compressors without affecting the output level.


Top
   
PostPosted: Mon Aug 21, 2023 1:01 pm 
User avatar

Joined: Sun Dec 23, 2018 7:44 pm
Posts: 981
Location: Texas, USA
Quote:
What would the goal of this be? We already have "Density drive" which basically adjusts both the thresholds and the injection levels to go deeper or less deep into the compressors without affecting the output level.
Ah, I guess it was to have a single slider directly affect other sliders rather than relying on a density offset/modifier that doesn't "return to center." I suppose this would be for designing presets and setting up all the modifiers/offsets to be neutral without going deep and adjusting the individual density or threshold sliders per band.

I sorta posted without enough context, as a sleepy me does. And now I can't remember exactly what I had in mind :lol: :oops:

EDIT: So, I recently learned that the scopes, goniometer, and AZIMUTH rely on legacy code where it's unscalable, and one pixel is 1 pixel. This is becoming an issue at 4k. The AZIMUTH correction line is extremely thin at 4k and doesn't scale with the DPI. Notice how the bottom of the AZIMUTH correction line doesn't draw all the way (circled in green). And the goniometer could use a faint antialiased line connecting the dots. Do I need to send my skin?
Attachment:
File comment: resized 50% filtered, resized 50% linear, and sized 100%.
goniometer.png
goniometer.png [ 83.95 KiB | Viewed 1961 times ]
Also would it be possible to antialias the scopes, or is that overkill? It just looks a little incongruous with how smooth the rest of the new UI is.


Top
   
PostPosted: Mon Aug 21, 2023 6:53 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11425
Quote:
Quote:
What would the goal of this be? We already have "Density drive" which basically adjusts both the thresholds and the injection levels to go deeper or less deep into the compressors without affecting the output level.
Ah, I guess it was to have a single slider directly affect other sliders rather than relying on a density offset/modifier that doesn't "return to center." I suppose this would be for designing presets and setting up all the modifiers/offsets to be neutral without going deep and adjusting the individual density or threshold sliders per band.

I sorta posted without enough context, as a sleepy me does. And now I can't remember exactly what I had in mind :lol: :oops:

EDIT: So, I recently learned that the scopes, goniometer, and AZIMUTH rely on legacy code where it's unscalable, and one pixel is 1 pixel. This is becoming an issue at 4k. The AZIMUTH correction line is extremely thin at 4k and doesn't scale with the DPI. Notice how the bottom of the AZIMUTH correction line doesn't draw all the way (circled in green). And the goniometer could use a faint antialiased line connecting the dots. Do I need to send my skin?

goniometer.png

Also would it be possible to antialias the scopes, or is that overkill? It just looks a little incongruous with how smooth the rest of the new UI is.
That's odd. I have a 4K screen and AZIMUTH does draw to the bottom. What's your scaling in Windows set to? 150% (default) or something else?

Connecting the dots in the goniometer can quickly get really heavy in CPU power, because we would need to draw lines instead of separate dots. For the normal scope we can combine things vertically (there's a limiter number of pixels), for the goniometer that will likely take a massive amount of CPU, which would also depend completely on how loud the input is and how many high frequencies are in it.


Top
   
PostPosted: Mon Aug 21, 2023 7:05 pm 
User avatar

Joined: Sun Dec 23, 2018 7:44 pm
Posts: 981
Location: Texas, USA
Quote:
That's odd. I have a 4K screen and AZIMUTH does draw to the bottom. What's your scaling in Windows set to? 150% (default) or something else?

Connecting the dots in the goniometer can quickly get really heavy in CPU power, because we would need to draw lines instead of separate dots. For the normal scope we can combine things vertically (there's a limiter number of pixels), for the goniometer that will likely take a massive amount of CPU, which would also depend completely on how loud the input is and how many high frequencies are in it.
I see. Gotcha. So combining things vertically would be like a large vertical resolution downscaled to look smoother? Or am I completely missing the point?
(edit: oh I see, you mean drawing lines vertically between the points on a waveform? So does that mean antialiasing is out of the picture for now?)

4k @ 200% DPI (detected 192), scaling of 125% in ST with mostly small fonts and 50% compactness. I'll include the skin section in an sts. Winamp Modern colors by Valiant.
Attachment:
skin001.sts [4.28 KiB]
Downloaded 134 times


Top
   
PostPosted: Mon Aug 21, 2023 10:14 pm 

Joined: Mon Aug 25, 2014 3:22 pm
Posts: 283
It came to my attention that my prior preset had some old settings in it which worked great for build 97, but not 98 or newer. So here's an updated file. The main problem was if you loaded my version 46 preset in build 98 or newer, you were not hearing the fast limiters engaging before Intellilimit, and therefore you were missing a little polishing density as well as overdriving the clipper beyond what I designed.

This updated copy is also built around some SDP fixes found in build 100 or newer.


Attachments:
Aquinas-0.48.sts [207.63 KiB]
Downloaded 139 times
Top
   
PostPosted: Tue Aug 22, 2023 1:05 am 

Joined: Wed Jun 03, 2015 7:03 pm
Posts: 199
I like new stuff, i checked latest beta and this preset from Wes. First I thought something was wrong with my output settings, then i checked and all was fine.
Sound is so agressive and almost every little dynamics is not there anymore.
Also on next louder track change, it need great amount time to stabilize and while doing that, distorts even more.
How can i fix that?

_________________
https://fmfeeling.com


Top
   
PostPosted: Tue Aug 22, 2023 1:28 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11425
Quote:
I like new stuff, i checked latest beta and this preset from Wes. First I thought something was wrong with my output settings, then i checked and all was fine.
Sound is so agressive and almost every little dynamics is not there anymore.
Also on next louder track change, it need great amount time to stabilize and while doing that, distorts even more.
How can i fix that?
I have the impression that something is wrong on some systems, On my pc, if I lower the clipper drive at least, this preset is very dynamc and absolutely not aggressive at all, and it appears to fix almost everything that I ever heard which was wrong in other presets. Bojcha sent me a screenshot earlier based on a recording with test tones that looked completely bizarre. So I wonder if something is broken on some systems - that's possibe because different CPU types each get their own code, and we've had weird compiler bugs before.

Can you try lowering the clipper drive, to see what it sounds like?

What CPU do you have?

I'll test the SSE2 version on my end, we have seen issues with that verson in the past - it's the version that you get with an older CPU or with Windows XP.


Top
   
PostPosted: Tue Aug 22, 2023 1:58 am 

Joined: Mon May 08, 2023 7:33 am
Posts: 37
Quote:
It came to my attention that my prior preset had some old settings in it which worked great for build 97, but not 98 or newer. So here's an updated file. The main problem was if you loaded my version 46 preset in build 98 or newer, you were not hearing the fast limiters engaging before Intellilimit, and therefore you were missing a little polishing density as well as overdriving the clipper beyond what I designed.

This updated copy is also built around some SDP fixes found in build 100 or newer.
I've checked this preset with one of my favourite test-tracks: Caroline Polachek - Fly To You. Unfortunately it doesn't sound how it should, I can hear how it goes up and down, I don't think this is the way it is supposed to be (f.e. 1:35 - 2:00 u hear it).

CPU settings: 6 cores (parallel threads), Intel i7-6700HQ


Last edited by Ek95 on Tue Aug 22, 2023 2:20 am, edited 2 times in total.

Top
   
PostPosted: Tue Aug 22, 2023 2:10 am 

Joined: Wed Jun 03, 2015 7:03 pm
Posts: 199
Quote:
Quote:
I like new stuff, i checked latest beta and this preset from Wes. First I thought something was wrong with my output settings, then i checked and all was fine.
Sound is so agressive and almost every little dynamics is not there anymore.
Also on next louder track change, it need great amount time to stabilize and while doing that, distorts even more.
How can i fix that?
I have the impression that something is wrong on some systems, On my pc, if I lower the clipper drive at least, this preset is very dynamc and absolutely not aggressive at all, and it appears to fix almost everything that I ever heard which was wrong in other presets. Bojcha sent me a screenshot earlier based on a recording with test tones that looked completely bizarre. So I wonder if something is broken on some systems - that's possibe because different CPU types each get their own code, and we've had weird compiler bugs before.

Can you try lowering the clipper drive, to see what it sounds like?

What CPU do you have?

I'll test the SSE2 version on my end, we have seen issues with that verson in the past - it's the version that you get with an older CPU or with Windows XP.
My CPU is and old I5 3er Gen running win 10

_________________
https://fmfeeling.com


Top
   
PostPosted: Tue Aug 22, 2023 2:27 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 11425
Quote:
Quote:
It came to my attention that my prior preset had some old settings in it which worked great for build 97, but not 98 or newer. So here's an updated file. The main problem was if you loaded my version 46 preset in build 98 or newer, you were not hearing the fast limiters engaging before Intellilimit, and therefore you were missing a little polishing density as well as overdriving the clipper beyond what I designed.

This updated copy is also built around some SDP fixes found in build 100 or newer.
I've checked this preset with one of my favourite test-tracks: Caroline Polachek - Fly To You. Unfortunately it doesn't sound how it should, I can hear how it goes up and down, I don't think this is the way it is supposed to be (f.e. 1:35 - 2:00 u hear it).

CPU settings: 6 cores (parallel threads), Intel i7-6700HQ
https://we.tl/t-ZTAEMchfbs

I just processed it here, with the clipper drive reduced by 10 dB so we only hear the compressor. Can you check my recording?

I'll also upload version BETA100, maybe the Sudden Drop Protection fix in that version fixed something. I'll create an official post for it tomorrow morning (the build isn't completely finished yet, the ARM targets aren't there yet). But you can download it now.


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 707 posts ]  Go to page Previous 154 55 56 57 5871 Next

All times are UTC+02:00


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