| Stereo Tool https://forums.stereotool.com/ |
|
| Stereo Tool 5.02 https://forums.stereotool.com/viewtopic.php?t=2661 |
Page 75 of 99 |
| Author: | eldoradofm [ Sun Nov 28, 2010 4:43 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
I'm using beta58. I will check with allow louder highs later today. |
|
| Author: | Bojcha [ Sun Nov 28, 2010 4:55 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Hi, Eldoradofm.. I checked that track, btw it's one of track for my tests, I hear what you mean, High tone like small bells. Fix for Blakan preset = Set 5,5kHz Clipper in multiband to 1.59. @Hans Nice filter "Punch Clip" - But i am fighting with it, to push "punch" more and set "punch clip" lower, or Less punch and "punch clip" higher. Somewhare at default 1.00 (095) is best and punch can be set bit higher, nice. i guess i should't ask anymore for strict clip (0-300/350Hz) ? 
|
|
| Author: | hvz [ Sun Nov 28, 2010 5:54 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Bojcha: You might want to turn on the 'Very deep bass protection" again, it can give some artifacts but if it has little effect (which is the case if you use not too high Loudness settings) they should not be noticeable. And it does really reduce distortion. |
|
| Author: | hvz [ Sun Nov 28, 2010 6:12 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Quote: At 1:39 and 2:57 the sound gets squashed because of the highs.
I see very loud frequency spikes upto 12 kHz there! Which is very unusual for normal tracks. I hope one of the items on my todo list will fix this:
Quote: - Fix Allow louder highs behavior. Attempt to replace it by a highs AGC. And rename to de-esser.
Currently, Stereo Tool lowers loud highs if their PEAK level is very high - but in case of pure sine waves, the RMS level (assuming the same peak level) is much higher than when it contains noise, and Loudness responds to RMS level instead of peak level. I wrote the current filter over a year ago to protect against noise, before I introduced RMS *anywhere* in Stereo Tool. So I need to replace it by an RMS-aware de-esser. (Which I happened to be planning to start with tonight  )
|
|
| Author: | Bojcha [ Sun Nov 28, 2010 6:19 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Quote: Quote:
from ~5kHz and up.- Fix Allow louder highs behavior. Attempt to replace it by a highs AGC. And rename to de-esser. Currently, Stereo Tool lowers loud highs if their PEAK level is very high - but in case of pure sine waves, the RMS level (assuming the same peak level) is much higher than when it contains noise, and Loudness responds to RMS level instead of peak level. I wrote the current filter over a year ago to protect against noise, before I introduced RMS *anywhere* in Stereo Tool. So I need to replace it by an RMS-aware de-esser. (Which I happened to be planning to start with tonight )
|
|
| Author: | hvz [ Sun Nov 28, 2010 6:24 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Yup. I usually use 4-7 kHz for separation between normal (voice) and highs (hiss) sounds. Also, I'm going to place this BEFORE Loudness instead of (the current) INSIDE Loudness for better results. |
|
| Author: | Bojcha [ Sun Nov 28, 2010 6:30 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
heh, few days ago i ask how to "join" last 3 bands in multiband 
|
|
| Author: | hvz [ Sun Nov 28, 2010 10:10 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Highs AGC added. I'm still checking what the correct default value should be. (Feedback is welcome!) Behavior of 'Allow louder highs' has changed: It's now a de-esser that limits the highs (above 4000-8000 Hz) below a certain RMS value. Default value = 65536 (I still need to add the corresponding volume in dB, but it corresponds to 65.53 in other RMS displays, or -2.46 dB). If you set it to 0 nearly all the highs are removed, if you set it to the maximum value (almost?) no limiting is done anymore. Important change w.r.t. earlier versions: Before I automatically increased the 'base' Allow higher highs' level if you selected 75 us pre-emphasis or if you increased Loudness. That was actually a very bad idea, because it made it impossible to find a single 'always good' value. There is now no 'base' value left, so the behavior does not change if you increase Loudness (which does mean that at higher Loudness settings it will remove more highs, because there just are more highs). Winamp DSP plugin: http://www.stereotool.com/download/dsp_ ... 510-59.exe Stand alone version: http://www.stereotool.com/download/ster ... 510-59.exe VST version: http://www.stereotool.com/download/vst_ ... 510-59.dll VST version (No SSE2): http://www.stereotool.com/download/vst_ ... 510-59.dll Command line version: http://www.stereotool.com/download/ster ... 510-59.exe Linux command line version: http://www.stereotool.com/download/ster ... BETA510-59 (not available) Linux GUI version: http://www.stereotool.com/download/ster ... BETA510-59 (not available) Edit: For the Eminem song I have to set the value to below 40000 (around 32768 sounds good), but that damages the highs too much imo. I think it can be solved by making the AGC behave a bit differently. To be continued... (At least the slider is now working again). Edit #2: There was still a loud 'S' handling bug in the code, I'll fix that first and from there see what's needed. #Edit #3: Fix for both issues is coming up, Eminem song now sounds good upto 70000 instead of 35000. (Slightly reduced punch in the highs in some cases). |
|
| Author: | Bojcha [ Sun Nov 28, 2010 11:33 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
This promises! is it possible to have some indicator .. when and how much this de-essing is active. edit: Found something is not good.. - "Up Speed" is too slow for De-Essing. - End With lower values whole audio knows to drop. |
|
| Author: | Chris [ Sun Nov 28, 2010 11:41 pm ] |
| Post subject: | Re: Stereo Tool 5.02 |
Beta58 is ok! I am waiting Beta 60 :p
|
|
| Page 75 of 99 | All times are UTC+02:00 |
| Powered by phpBB® Forum Software © phpBB Limited https://www.phpbb.com/ |
|