All times are UTC+01:00




Post new topic  Reply to topic  [ 89 posts ]  Go to page Previous 1 2 3 4 59 Next
Author Message
PostPosted: Mon Aug 03, 2020 12:54 am 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 10068
@Lynx_TWO: The delay thing should be solved by the VST host. There is a VST call to query the amount of latency, and it's the host's job to compensate for that.

@about block sizes: Yeah, I know... that's the main reason why we currently don't support very high sample rates. Internally it currently downsamples to something below 64 kHz. Actually, the Declipper plugin doesn't do that - but it would actually need bigger block sizes, which it doesn't have either. Without that, the CPU usage would greatly increase at higher sample rates (double for 96 kHz, quadruple for 192), and the memory usage would as well. For FM and streaming it doesn't really make sense, but for mastering it would. So I guess it would make sense for the VST plugins and for WatchCat.


Top
   
PostPosted: Wed Aug 05, 2020 10:11 am 

Joined: Sun Nov 17, 2019 3:51 pm
Posts: 2
Hello Hans, tell please, what exactly do function Limit CPU ?
There is
Very low (pi 3)
Low (pi 4)
Medium
Sane maximum
Limitless, follow preset

I was tried to switch from Sane maximum to Low (pi 4), but I didn't hear the difference in the preset I am using.
I just noticed that the CPU consumption has decreased.


Top
   
PostPosted: Wed Aug 05, 2020 10:23 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 10068
It limits some settings to lower (or in case of "Sane", sane) values. For example, the final clipper "Strictness" can be raised all the way to 20, but that takes a lot of CPU power. In Pi 3 mode, it's limited to 3 or so, and in Sane mode it's probably around 10. Similarly, there are limitations to settings that control Stokkemask strictness, Declipper precision etc.

Higher *is* better, but how much effect it has on the actual audio typically depends on how aggressive and loud your preset is.


Top
   
PostPosted: Thu Aug 06, 2020 11:48 am 

Joined: Mon Jan 27, 2014 1:55 pm
Posts: 68
Hans, can you make the Release gate on AGC separately for all 4 bands?

Whenever i play vinyl records where the bass is suddenly silent for a couple of seconds or especially when the vinyl record is fading out at the end of an song, the bass is annoyingly fed back (feedback?) through the needle and back into Stereotool. This leads to loud constant bass humm untill a next track is starting.

With release gate threshold at somewhere around -20 dB, i can eliminate this problem for 99% of the cases. But, as you know, the gate works on all bands at the same threshold. The -20 dB threshold is way too much for the highest 2 bands in my case so i would love to be able to set them for every band separately.

Sure, i can also lower the volume of the studio monitors, but they're not even that loud actually when this happens.
Place the speakers somewhere else or moving the record players is also not an option because of a custum made studio furniture.

Record players are the Audio Technica LP120 with Tonar Banana element + needles. Not the cheapest, certainly not the most expensive but they are fairly sturdy. I have tried putting them on floating pads but that did'nt do the trick.


One other thing:

I drive the Advanced Clipper to around -4 dB and i think i have a very consistent sound, but when i switch on ABDP, i notice that it removes a lot of deep bass and lowers the bass overall volume. If i turn off all the bass clipping settings, the bass is far better than with the settings at standard values.

Does it matter that i have the bass settings switched off? Does it still remove distortion? For example the always clip deep bass and lower bass dynamic further settings are 2 settings that flatten out the bass in my case...

Oh, when i was typing another thing came to mind...

I noticed that the Noise Gate takes a lot of highs out with the standard settings. Does it just cut off audio at the set level or is it somewhat more intelligent? When i press "Difference", i hear a lot of highs that are cut out of the sound. Lowering the thresholds have good effect for sound, but then the noise is not faded away properly. So i tried configuring it to my needs, but i always end up shutting it off and accept the noise that's there.
At the moment i would say i'm close to Skyradio NL with my sound (in my studio on my equipment) but on the radio there's that silence in the music. I don't know if you understand that, but you can hear that the background is silent. There's only music. I'd like to create that. I know on FM it's due to pre-emphasis that helps lowering the noise in high frequencies and i still need to try my FM transmitter with Stereotool but if i could do something similar with the noise gate without cutting out highs, that would be wonderfull. For streaming purposes...

Thanks in advance for the effort you all put in this software. Keep up the good work!


Top
   
PostPosted: Thu Aug 06, 2020 5:58 pm 
User avatar

Joined: Sun Dec 23, 2018 7:44 pm
Posts: 155
Location: Texas, USA
I have to say between the 4-band AGC and BIYF booster, my sound has never been more consistent. This is indeed an interesting beta.

[EDIT1] Potentially regarding synchronizing samples, would it be possible to add a wet/dry mix? Trying to use the wet/dry fader in Reaper, you can hear the un-delayed/un-buffered aspect mix with the slightly delayed/buffered one. So perhaps an in-app/plugin wet/dry mix would fix that.

The bypass is great for disabling all processing and freeing up cycles. However for A/B-ing and finding what might be the right levels for all settings at a time, wet/dry might be extremely valuable. Especially if it was possible to take a 'snapshot' of the settings at any (let's say... 60% wet) value and export the altered settings. This would probably be a new GUI type wish.

[EDIT2] Better grammar and punctuation.


Last edited by MrKlorox on Mon Aug 10, 2020 6:18 pm, edited 1 time in total.

Top
   
PostPosted: Sun Aug 09, 2020 8:34 am 

Joined: Sat Feb 21, 2015 10:53 pm
Posts: 18
Quote:
@Lynx_TWO: The delay thing should be solved by the VST host. There is a VST call to query the amount of latency, and it's the host's job to compensate for that.

@about block sizes: For FM and streaming it doesn't really make sense, but for mastering it would. So I guess it would make sense for the VST plugins and for WatchCat.
Yes after some more tests I think it was the DAW that was randomly allowing the audio to shift, since it seems hit-or-miss, so it must be an issue with the software.

Looking forward to having larger block sizes for mastering use! I agree it wouldn't make any sense for FM and Streaming, too much latency; but hey I'm running this on dual Xeon system with 20 cores, 40 threads, and 192GB of DDR4 ECC RAM for restoration and mastering purposes, so bring on the memory usage! :D

Thanks and keep up the great work! I do love the fact I can also use this on at least 10 channels in REAPER and it seems to just work. Had no idea the VST3 plugin had multichannel support, but that's fantastic! It's amazing to me how many plugins, even brand new ones, don't add multichannel support.

EDIT: - OK, very strange - while in does appear that StereoTool VST3 in REAPER supports multichannel, only the first two channels are output at the same volume. The rest are muted quite a bit...


Attachments:
File comment: Beta 9.53 - This is in REAPER - 8 channels in, 8 channels out. All seems well until...
StereoTool in REAPER.png
StereoTool in REAPER.png [ 8.16 KiB | Viewed 1981 times ]
File comment: Whoops! Levels of the other 6 channels are MUCH lower but still seem to be there relative to the original level?
StereoTool Multichanel.PNG
StereoTool Multichanel.PNG [ 19.12 KiB | Viewed 1981 times ]
Top
   
PostPosted: Mon Aug 10, 2020 8:15 am 

Joined: Mon Jan 27, 2014 1:55 pm
Posts: 68
In the latest Stand-Alone 64 bit Beta i noticed that under Configuration > CPU and Latency > Limit CPU the standard setting is Very Low (PI 3)

I had some cracking/creaking sounds whenever a very deep and long bass came trough. The kind of crackling Hans explained to me in a earlier topic, had something to do with buffer settings . But in the previous Beta i did not have this crackling/creaking, so i started checking all the latency settings...

The CPU on my quad-core pc was running 14% CPU with the setting at Very Low. I've set it to Limitless and the cracking/creaking noises are gone. Also, but i'm not 100% sure, i sounds to me that the overall sound quality was improved a lot after setting it to Limitless.
CPU usage increased with a staggering 2% 8-) , so i guess Limitless is actually the "Normal" setting?


And, before the setting change i could sometimes hear the AGC work and after the change this was a lot less. Does this CPU setting have effect on the precision of the AGC/Compressors? It can be my settings though, but i surely noticed improvement in the AGC part...

Regards..


Top
   
PostPosted: Tue Aug 11, 2020 2:27 pm 
Site Admin
User avatar

Joined: Mon Mar 17, 2008 1:40 am
Posts: 10068
Very low should *NOT* be the standard (unless you're running it on an ARM device), Sane should be the default...

Note: It's a bit quiet here because we're working on several issues that aren't suitable yet for beta versions. We will continue with the bass and 4-band AGC stuff when this is finished.

@Lynx_TWO: Multi-channel (upto 8 channels) should work, and the output should be processed on all channels, not just the first 2. Can you try what happens if you set Pre Amp or Post Amp at 0%? So verify if those other channels are being processed at all?


Top
   
PostPosted: Tue Aug 11, 2020 7:08 pm 

Joined: Sat Nov 12, 2011 7:46 pm
Posts: 260
Unfortunately I can only say the same thing about multichannel as @Lynx_TWO Here in Audition I can send a 5.1 signal through the plugin. But only the first two channels are processed or changed. The others remain unchanged. And on the other hand this results in a big offset between the original recording and the processed audio.


Attachments:
2020-08-11 19_10_03-Adobe Audition.jpg
2020-08-11 19_10_03-Adobe Audition.jpg [ 117.12 KiB | Viewed 1716 times ]
Top
   
PostPosted: Thu Aug 20, 2020 10:10 pm 

Joined: Wed Nov 08, 2017 3:16 pm
Posts: 77
bug found with 9.53 beta 008
using legacy agc, the bass correction has the opposite effect as described.
tested with 30Hz sinewave @ -18dB input.
please, dont troll me for using the legacy AGC, it just works better at keeping variable input audio levels at a consistent "vanilla" level with -6dB headroom on the output, cant seem to replicate that correctly with the "new" AGC.
confirmed this bug also exists on the release version 9.52.


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 89 posts ]  Go to page Previous 1 2 3 4 59 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:  
cron
Powered by phpBB® Forum Software © phpBB Limited