Stereo Tool

Ok, to avoid having to browse through this page every time, and because the list is getting too long for me to remember, I'm posting some things that I want to add in the near future:

GREEN Done
RED Won't do that

Short term stuff
e. Login/password with instances
f. Console version: Read sample rate and bit size from WAV header - 1 day
g. Installers: Detect more programs and automatically install in those programs - 1-2 days
h. Cloud file upload/process
i. MicroMPX add hole punchng (or TCP stream?) - 1-2 weeks, includes adding some sort of protection and kicking out existing targets

Fixed items
0. Declipper bug found by Leif - 0 dB effect DONE
1. Add controlled amount of distortion, but not IMD Hm, maybe I should turn it off when there's no bass to make voices sound fuller... Will check that later.
2. Multiband surround issue http://forums.stereotool.com/ucp.php?i= ... f=0&p=2091 FIXED
3. Improve Stokkemask - very positive results! Waiting for feedback
4. Use RDBS names when using 75 us pre-emphasis http://forums.stereotool.com/viewtopic.php?f=16&t=4869
5. GUI closing warning http://forums.stereotool.com/viewtopic.php?f=16&t=4708 DONE
6. Full screen mode (mainly for small displays)
7. Fix AGC problem (release doesn't respond the same at very high input levels) DONE
7a. AGC reduce smoothing over time - causes level to be controlled LESS well. Use 1024. DONE PERFORMANCE OPTIMIZATION NEEDED (PRE-CALC SINUSES) DONE
7b. Same for SB and MB compressors. Use 512-1024 DONE PERFORMANCE OPTIMIZATION NEEDED (PRE-CALC SINUSES) DONE More performance optimizations are possible here!!! DONE
8. Add labels to meters (-6 dB, -12 dB etc.) - problem is where to put them? http://forums.stereotool.com/viewtopic.php?f=16&t=4928 DONE
9. Output MPX in command line version (useful for Linux version), https://www.facebook.com/messages/christian.milling DONE
10. Waveform zoom in/out DONE
11. Improve multi-touch screen handling DONE
12. New version notification: Remove from GML plugin
13. Add R128 normalization to GML plugin version DONE
14. Improve VMware detection (add param, and let it be set to VMware, Native, Unknown - where Unknown can be switched to either if it turns out to be that.
15. Metering: MPX power, maybe R128? (but that takes memory and CPU load) MPX Power DONE First optimize BS412 memory usage (group 128 samples into 1), so we can always measure it, then add this. DONE
16. Use different names and scales for the new compressor settings.
17. Add MULTIPATH protection page. DONE, changed MP clipper behavior, now similar to Stokkemask but with some settings not configurable. Tests needed...
18. Add hysterisis (delay, 50-100 ms?) when changing ASIO buffer size to avoid Juli@ driver crashes. 250 ms added, waiting for test results
19. Make sound card handling more robust against driver errors http://forums.stereotool.com/viewtopic. ... 023#p23023 ASIO debug version still needs to be tested on an ASIO pc (and I need to do that myself) DONE
20. Receiving streams. Let receiving wait for about 30 seconds before it forces a restart. Better: Make it configurable (via input buffer size?). Make GetVlc() PARAM version: WantsVlcInput().
21. Bass AGC to make bass really constant. Currently bass can only get BELOW midhigh level, but it must be allowed to get ABOVE it. Add some settings to control maximum level above that of normal audio. And add metering, both to AGC and bottom panel
22. Add compressor 'Continuous Release' mode, plus optimizations. Release is probably good now, but I think I should not touch attack (at least for now) - so restore the old code. Not sure if this is thing is useful?
23. Check effect of new RMS compression/limiting. What to do with block sizes? Check old presets... Also check what to do with better limiter settings
24. Fix R128 normalization in Carbon Coder plugin version - complaint from Fox TV
48. Power Highs like Power Bass. Also moved it to before the AGC for better level control.
52. 10 to go - Check CPU load. Start with checking if there's anything left that uses the 'unnecessary steps'. Sevdah Web preset: Data still gets converted 58 times... I think I need to do this one first, it should have some effect on the CPU load. 28 removed - next convert the 2 IIR filters so they can be optimized and the merge/split around it can be removed. I'm not measuring any effect from this though (but it makes the code simpler which is also good)
53. Noise Gate/Stereo Boost: Pre-calculate 1-cos() and sqrt() values.
55. Check MemoryPool behavior for cache improvements -> No effect measured, and might make behavior less constant.
56. Check if we can go in opposite direction for each next step to improve cache.
57. Check if lazy reverse FFT is an option. -> No, difficult and gain does not even seem to be measureable.
58. Created a separate class that performs the processing chain. Currently the same code is repeated twice (once for normal processing, once for low latency processing) - which means that a lot of code is duplicated and it's difficult to add extra chains. Most, not all, of that code is now moved elsewhere. (although the low latency chain is not yet modified, but that should now be relatively easy) -> Done!
59. Add composite clipper for stations that add stereo and RDS later! -> No, I was wrong, this isn't possible.
62. Improve multi-threading by making a single function that performs all the threaded stuff.
63. Optimize Final Clipper!
72. Make Normal Output with either FM-preemphasis (without composite clipper) or completely separate preemphasis.
69. Add proper AM panel.
66. Improve AM, check if I can do something with c-QUAM stereo
73. Add AM C-QUAM stereo clipper/limiter mode.
28. sending out streams.
54. New Stereo Boost: Increase L-R, measure |L+R| and compare against old, adjust band level to stay the same DONE This is done in Stereo Delay
46. Find and use template optimizations, and rename current code to _T_ names.
31. New stereo widener. (Lol that should have been on top. See, I start forgetting things )
83. Check what goes wrong with Hyperthreading -> Nothing (seems to have been a hardware issue on an unstable system). Added CPU core affinity settings to control the behavior.
77. Feedback compressor mode: Take LIMITER action into account! So act as if limiter is in front of compressor.
67. Cache issue when copying data on other cores. Does not seem to be an issue, faster memory has almost no effect on performance (a few percent).
47. Duplicate some settings (lowpass etc.) to enable making 1 preset for both web and FM. Partially done, clipper drive should also be split.
36. 64 bit version performance improvement (requires newer Intel compiler, performance issue with that is now reduced to 5% loss instead of 25% earlier, but this 5% still needs to be fixed). Problem is solved with XE2015 compiler.
37. Unique preset numbers for GML plugin (and myself)
78. Try using new limiter method (blending instead of using levels) for clipper as well! - Tried, result was worse.
32. Add noise removal filter based on noise spectrum.
75. Clipper protection must be checked again with current knowledge, to make more punchy bass kicks. See also these images: http://i.imgur.com/42fKt7v.png , http://i.imgur.com/8HkC6gu.png .
84. Multiband band coupling matrix.
79. Use Dirty Mids/Highs on final steps but not on preclipping steps. Might reduce pumping/make sound more consistent.
85. Redesign Advanced Bass Distortion Protection.
68. Is this something? http://www.sweetwater.com/insync/what-s ... mpressors/ - Never mind, we have water based now
43. Do something with alternative attack mode - No, new compressor instead.
88. Freeze slider for MB, which stops all bands simultaneously. Good for voice.
91. New compressor (ask Joop for name). Pushing too hard in any direction should increase R. Which kinda makes the maximum speed controlled.
89. Try out 2-band compressor after current MB. - new MB is there, configurable number of bands.
94. Change RDS default settings (check first!) - received this: "Flags Artificial head and Compressed are archaic and should be kept cleared unless you have a real reason for their use." - yes, Compressed links to "CCIR study programme 46A/10, Dubrovnik 1986" - I assumed it means compression is used on the audio (!) but that doesn't seem to be the case.
72. Gating in MB does not work when using Continuous Release. Maybe also a hard gate? Already fixed in Analog redesign, skipping this one.
41. Optimize performance with no processing enabled - Done as part of Omnia.9sg development
60. ProcessBlocks_Mus2Ch and cleanup code can be optimized (SSE2 etc.) - Only used in Stereo Boost. And already using #pragma's.
26. Bass compressor http://forums.stereotool.com/viewtopic.php?f=16&t=4976 - Do we still need this? - Probably not now we have True Bass. Removing it for now.
110. BASS PRE-RINGING FILTER IS AUDIBLE IN MULTIBAND OUTPUT!!!!!!!!
123. Add backup sound card support (sound card 1 switches to 2 after some silence).
121. Automatic "Ignore high frequencies above" frequency selection. 19200 for composite, maximum lowpass freq + something for others.
153. Try making "True Bass" for highs
51. Add wideband (maybe 2 bands?) limiters for streaming. Doesn't seem needed anymore.
40. Optimize GUI - Not really relevant, pretty optimized already and running in low priority thread
42. Add some AoIP thing (LiveWire, Ravenna, ...) - can this be done in software??
71. Delossifier: http://www.facebook.com/l.php?u=http%3A ... =xAQEctnCM see FB chat with Alexandre Quessada 14-may-2014
95. RDS: Make US AM AF freqs available in 75us mode.
101. Add clipping at 75 while BS412 limiting at 50, for stations that stream using a lossy codec to their transmitter site.
102. Add Generic plugin function to set parameter values.
103. Bug: Mac: After closing About box, main GUI remains unusable. And it's too small. Suggested fix: Disable for now, make a better About box later. Should be part of the planned Mac native GUI design.
104. Bug: Mac: If you click the Help Button it will close Reaper. Not sure if this is supposed to happen. Also the Help-URL is not existing. Also part of the planned Mac native GUI design.
105. Upsample limiters (high bands) to avoid distortion
112. Restore previous value. http://forums.stereotool.com/viewtopic. ... 06&p=32564
128. Also allow processing MB separately for Normal Output, probably only with Density/Aggressiveness sliders relative to FM.- Doesn't make sense, better to run a complete separate copy
131. Hide deprecated settings when not default.
155. MB: Overview window with the most important settings sounds like a good idea! I will make that.
92. Make a stand alone application that can process files, one file or batch mode, with many different input file types. To make it no longer needed to use a VST host.
150. Graceful switching between presets, without the current glitches. At least when the same filters are used.
90. Save PNR data in VST plugin!!!
99. Fix version with Thimeo engine latency in Winamp (50% higher on normal mode, 75 in Thimeo mode - before replacing readReached event by a switch it was 50% and 50%, I should be able to restore that - caused by looking at the wrong [][] values). Also, the Chain() thread should wait for the calculations to finish. Stand alone latency is ok.
120. Weird bug @ TeamFM: Setting buffer size LOW causes hiccups. Increasing the buffer size does not help, until RESTART is pressed or switching between sound cards. Problem also only occurred at lower sample rates (44.1, 48, 176.4 was ok). Latency 0.1 seconds. Above 0.14 it seemed to be ok. When using normal sound card instead of VAC as input the problem was gone as well.
122. Make stand alone file based version, maybe using BASS for file I/O?
161. Try running Absolute Highs on N=8 and N=10 (instead of the current 9) and take the minimum. - TRIED
178. Input Balance issue (see 2 posts below BETA054), cause known (reallocalte big buffer for waveform display, then buffer size doesn't match, and for some reason this causes a shift in time. However, calculated values appear to be correct. Check in more detail after NAB.)
179. LW: Show Local IP(s)?
180. New Sync code with ASIO & PrecisionClock
181. Make Linux "Generic" plugin
183. Use 4 (or 8) cores for processing, Raspberry Pi etc. Or Clipper on separate CPU?
74. AGC must use more bands, in a smarter way than now. Not just a different crossover behavior but also different linking etc. - redesigned instead
25. More AGC stuff http://forums.stereotool.com/viewtopic.php?f=16&t=4798 , http://forums.stereotool.com/viewtopic. ... 018#p23018 Not too sure how important this is with the latest MB improvements.
33. It would be good to have a field in the presets dedicated to describing what makes the preset and / or author information, date, type of music that is being sent and so on. http://forums.stereotool.com/viewtopic.php?f=16&t=4227 - Done, SST only
76. New stereo widener may need to be improved after I receive some 222a recordings.
83. Read current sample rate of sound cards as in http://i.imgur.com/5NqQBsV.png
97. Add load/save presets to remote GUI.
106. AGC gating: Add switch to "Gate Based on Level after AGC"
109. Add/change 2 and 3 band AGC with crossover similar to "Power bass" crossover.
124. Wet voice detection? Ask Leif what it is used for.
126. Possibly merge Stereo Boost into MB1. Should be before compression but after noise gate. No AutoEQ first.
129. VLC input, ASIO output -> Samplerate is locked to VLC sample rate, so MPX out usually won't work. Should be fixed as part of sound card redesign.
132. GUI bug: Weird jump in scopes when using Ignore High Frequencies and view size is 10 seconds when moving in/out of scope panel. -> Ah, caused by difference between full redraw (on move over) and scrolling to the left (when normally functioning).
157. New AGC design might be as simple as using the current multiband with a mapping to make it act like it has a window. As I have actually been planning (also for the MB's) for quite a while now. Either 1 band or 4 separate bands, which still should probably support matrix mode for the top 3 bands then. In the MB, if we do this, each band should still be allowed to be displaced (not so much moved!) slightly inside the window to better fit in between the surrounding bands, that might make the sound somewhat more natural.
175. Add HTTP commands and feedback - viewtopic.php?f=16&t=7701&p=34436#p34436
182. Get rid of Jack for Mac, possibly Linux too
184. Stand alone program & VST with demo option to do Declipper, Delossifier, AZIMUTH, Dequantizer, PNR, Natural Dynamics. (No noise gate!). Somewhere in next 2-3 weeks. ProTools plugin without demo. Maybe run limited time and stop. Or announcement every 25 seconds.
186. Add RT+ and Arctic Palm-like formats to help page
189. Add 9sg link - create extra Normal Output mode
191. Also show tooltips for blocks of parameters.
- Continue with finding parameters to hide
200. PortAudio MATHIJS
201. Entering license in Mac version VST crashes MATHIJS
202. WatchCat close issue on Mac MATHIJS
207. Continue with AA Ratio, ARS Mode, Max Ratio, check which are needed and adjust GUI to make everything fit again.
35. VST other param standard (JesseG)
100. Check Delayed Bass Clipping: Relative sensitivity. Doesn't work according to Bojcha. It does work though, but maybe doesn't achieve its goal.
114. RDS: UECP AF Method B
115. RDS: UECP: ODA groups for TMC, RT+
116. RDS: UECP: Linkage (LSN). Links one station to traffic info on another. Only TA on on one station means switch to the other (EON).
117. RDS: PTYN (both via UECP and directly)
118. File info in DLL (version number, copyright - standard resource can be linked in)
119. Watchdog with Windows messages, ask on startup list of processes and get watchdog process id PID - . Start if not running. Define own messages. Will work around hangups in VLC!
123. Update Hard Limit, using new limiter should be better!
141. Make oversampled limiters for very fast "limiter speed" (under 1ms already some distortion)
160. Improve MP3 detection (very low bitrate, Lay Down)
187. Add Stokkemask auto control (quickly move up, slowly move down)
194. Make DIFFERENCE work in Natural Dynamics for separate bands
195. SPA "This sound is processed by"?
197. Adobe Audition CC shows VST plugin at wrong size - bottom and right side (big chunk) are cut off.
198. MAC version VST 64 doesn't work in Audio Hijack and Nicecast -> No errors logged, it just doesn't open it (even on explicit loads) -> ask Rogue Amouba for info. - both are deprecated
205. "Stuck" 1 dB difference multiband compressor mode before AGC
206. Thread priority overview
208. Stokkemask redesign, more accurate and faster.
209. New stereo widener based on AutoEQ
212. RDS RT end with hex code 0D (Carriage Return) instead of padding with spaces, then stop sending - faster and no unnecessary scrolling.
213. RDS RT increase priority on changes like with PS
218. MicroMPX NIC selection broken in SST web interface - Koek
220. SST: Show warnings
221. Website rename Products to Buy
223. VST3 support DAVID
227. WatchCat: Use "Simple Clipper" for R128 normalization if there's no license for Advanced Clipper.
228. WatchCat/Generic plugin: Respond to latency request with different number for step 1 and 2
230. "Stereo Width" in Stereo Image is broken - it causes phase problems if you're using it to go to mono with silence on the other channel. This can be made smarter!
234. Check out Butt
29. Parameter scheduling http://forums.stereotool.com/viewtopic.php?f=16&t=4760 , including for RDS http://forums.stereotool.com/viewtopic.php?f=16&t=6037 .
30. Just read this in a Facebook discussion: "reduced low bass in the L-R reduces perceived IM distortion under multipath conditions." https://www.facebook.com/groups/transmi ... ment_reply
65. Smooth volume up on startup (for after power outages etc.). Use 1-2 seconds to reach max volume. Could be done on input...
108. Add new (fixed) ITU1770 to AGC too. Would be also good to be seen in PEQ graph.
169. Add filter to optimized processed audio for lossy codecs? See comments (//#warning) in Clip1Ch::clean1ch()
229. AF method B via UECP
34. 7" less crowded interface https://www.facebook.com/photo.php?fbid ... =1&theater
98. Dequantizer: Make the effect configurable for the lib build. And make it more configurable (final filter vs. first filter).
111. Combine filters into a single band split/merge (ND, AGC, Stereo Boost, MB1, MB2, True Bass, more?). Requires AGC redesign, start with MB1+MB2.
135. ASIO number of channels should be determined dynamically.
158. Replace all pop-up windows by internal displayed windows (Declipper, Delossifier, About).
- Add AF method B via UECP!
172. Show band frequencies in GUI overview (Natural Dynamics, MB).
177. Check Dirty Voices bass IMD problem. Maybe also rename to Phat Bass?
204. Maybe add common logging?
211. Kantar
214. Optimize DrawWave in Server.cpp similar to built-in GUI
216. When using SCA as a second audio input and in combine/mix mode (ie not in backup mode), LQLL output does not contain the SCA audio input, only the normal input, whereas both FM out and Normal Out outputs has them both mixed as expected.
222. Use top part for extra displays (declipper, numbers for MPX level etc)
232. Declipper check L-R effects
236. WatchCat website
239. GUI: Hide disabled stuff, far less confusing / much cleaner interface.
245. Koek update interface for MicroMPX
246. L-R reverb
247. Crash ProppFrexx
248. Update Omnia.SST presets
256. Fix Declipper L/R noise
27. Easy way to enter numerical values http://forums.stereotool.com/viewtopic.php?f=16&t=4929
152. FM Output bypass mode
188. Add auto detection of certain sound cards, like Marian Trace Alpha.
226. Multi-instance version
235. Loudness meter (also for streaming)
241. Delay RDS? For stations where the audio comes in with a delay, RDS text updates too soon.
244. Dynamic adjustment of priorities for HW
259. AES67 thread affinities and priorities
170. Absolute Highs: Add auto-adjust HPF frequency - Not doing this anymore, Highs Exciter sounds much better anyway
251. Change clipper defaults (Dutch Chocolate Moose has some improvements that use the new features)
a. MicroMPX (mainly): Use optimized Fourier transform
b. MicroMPX image: In setup.sh, add selection of sound card
c. MicroMPX: Auto-fetch license key based on 8-digit code from new site
d. PIMP/STAMP: Make/fix image, and finish web interface

To be done
73. Natural Dynamics must do nothing in already dynamic tracks, such as Abba - Gimme Gimme Gimme (original version). - REDESIGN, 2 WEEKS
146. Add impulse response import function!!! Useful for both non-linearity and echo/reverb. - 2 days
Make bands with crossover frequencies - 1 week?
38. Requested extra metering: http://forums.stereotool.com/viewtopic.php?f=16&t=4977 , http://forums.stereotool.com/viewtopic.php?f=16&t=4785 .
39. Something for preset A/B comparison: http://forums.stereotool.com/viewtopic.php?f=16&t=5056 .
44. Monitoring feature for side chains etc. - 1 week, includes adding DIFF output
45. Auto-scale filters to make levels constant (turn MB on/off should not change level etc.). Would be nice if we could let it 'listen' to audio for a while and adjust settings based on that. - 1 week
64. SB amount of dynamics detection for Natural Dynamics - http://forums.stereotool.com/viewtopic.php?f=16&t=5519 / Add Natural Dynamics wideband dynamics input, and perform smoothing for very short 'crackling' sounds (Jesse Graffam preset 'Aurora beta 93'.)
80. Auto-switch to mono for mono content (configurable).
81. Simultaneously process for FM and AM. - 1 week
86. APO: http://forums.stereotool.com/viewtopic.php?f=8&t=5747
91. PNR-like frequency curve learning to create equalizer, among others to learn microphone response. - days
125. Move Noise Gate into MB1. Try whether it's good to compensate for AGC action or not (probably not - according to Bojcha, see Skype chat 23-dec-2015 22:00). Alternatively compensate for AGC, but that's not how many others do it.
130. Auto phase rotation adjustment for optimal result, none when not needed. - Not sure if this is possible. Might take a lot of work.
134. Add a warning for the negative Extra Delay value that's bigger than (or almost as big as) the buffer size? - 2 hours
147. GUI: Add BS412-like graph for levels - 2 days, can display ITU history at different scales
149. Global mode change: Hide features that you don't need (so no FM for non-FM users). http://forums.stereotool.com/viewtopic.php?f=16&t=7095 - 2 days
151. Stand alone version: Let installer create desktop shortcut
154. MB: Add voice volume protection, don't allow mid bands to drop below surrounding (to some extent), protects voices and it might help to replicate the sound of certain other processors - Do we still need this? If so, 2 days?
159. Try Delossifier on L-R??
164. Delossifier: Make Absolute Highs use normal HPF frequency if Delossifier detects no MP3. Or use combination of measurements or something... - 1 day
165. Normal Output: Don't output FM anymore! Use plugin_or_fm_output_enabled && !ALWAYS_SOUNDCARD_ACCESS. But that's not all - also check different lowpass freqs and clipping levels etc. Might be better to combine this with sound card redesign. Fixed for SST. - GOOD THING TO FIX FOR V10! 1-2 days
166. set Encryption : impossible to view or modify all the processing settings. Solution is: Password like there is now (but only on Processing/repair settings!) to unlock viewing the settings plus one to load the .sts file (in the .ini it should load automatically I think...). In both cases, we could point to a separate fle that contains all the encoded settings. In case of .ini the password for unlocking must be stored in the .ini file..
167. Preset Demo : an encrypted preset with an audio message and beeps when not registred (same as your plugins)
168. Preset Licensing : encrypted preset without beeps unlocked by a license - or never visible?
171. Absolute Highs: Check how slowly the frequency should rise. Probably should only be available if Delossifier freq responds slowly? (Maybe already ok?)
173. Try: Volume difference effect with different speed at different frequencies (From Pre-ringing killer)
176. Add sound card adjustment presets for analog sound cards. - 2 days + 1 day to add auto-detection mode if that's possible
190. Try adding phase delay compensation Mathijs?
192. RDS encoder: Use Mathijs' method with interpolated 0's. Should give a nicer signal! Also I saw a weird jump bug in the raw (unfiltered) RDS output. Shouldn't matter (it gets filtered out) but still better to fix it. - 1 day
196. Put in Ben Liebrand settings for vinyl cleanup? Asked question on vinyl forum. - 2 days
198. Make Mac AU plugin BUSY - AARCH!
199. Figure out new value for "Stop declipping if texture did not increase above", turn off on no useful action (also in display).
203. Add buffer underrun warning - auto detect lowest possible latency?? - a few days for the auto-detect
210. HD delay Belar - a few days
215. Create ST Lite, either as a mode or as separate software. - A few days, probably very useful with new Pi image
219. Add MME warning
224. Auto-setup lower latency settings and slow CPU threading. - Add to high CPU usage popup; 2 days.
225. Select instance with command line option
231. 18 Hz trigger tone
233. Reduce extreme stereo
237. Thumb up/down in GUI for presets - 1 week. But we probably want to combine this with a preset website portal; then it will take more.
238. Separate product: SCA2 use as backup, feed to output without processing, bypassing normal chain. Use to have FM receiver as backup, and then just feed the full signal to the output.
240. Fardau
242. Optimize ABDP for 9sg etc
243. Simple IF version -> wait for Tom
249. AES67: Add WheatNet
250. Create file when input fails
252. Installer: Make installer aware of programs like MediaMonkey, AIMP, ..... Partially done
253. HD delay
254. Document JSON interface Done, but should also be included for ST. - A few days to auto-generate a list of all parameters (which would be useful anyway).
255. Documentation: "Processing" page contains all kinds of BS412 info - why? - Will be fixed when we redo documentation based on JSON files. 1 week.
257. Console version: Read sample rate and bit size from WAV header - 1 day
258. # bands selections should allow option to insert a band instead. Or ask what to do when you change it. Wait for Toms interface - 2 days?
260. Add input resampler - 2 days
261. MicroMPX: Create file format, and optimized (fewer packets) and TCP streaming - 2 weeks
262. Is a few % phase wobbeling a good idea to improve stereo image and flat frequency response in rooms? - 1 day
263. Try Delossifier with new Highs Exciter! Or a combination! - 4 days
+= 95 days
264. Gary Kline's MP3 fixing wobble filter which appears to make MP3's sound better. "Wobble effect is gone in HQ mode - must be made an option! - Gary Kline"





TO BE DONE - 9SG:
- Nu is er de optie bij UECP “Use internal settings on connection drop”, dit lijkt goed te werken. Echter als de verbinding weer is hersteld, komen niet direct de gegevens ingesteld via UECP terug. Pas alle settings overnieuw worden gestuurd, zijn deze weer aanwezig in de 9SG. Zou dit niet gebufferd kunnen worden en zodra de connectie terug is, dit worden weergegeven?



TO BE DONE - O11:
- .a file

Long descriptions
49. Redesign whole sound card control part. Create input and output points for each available sound card, then let users connect these to points in the program (for example one input can be sent to both normal and low latency inputs, and a patch point anywhere can be sent to a normally unused output sound card). Basically, a source and sink should be created for each input or output sound card, both ASIO and non-ASIO, and for streaming. If one input sound card is used multiple times that should work, if multiple outputs are sent to the same output sound card that should work as well. Theoretically it would even be a good idea to allow sending outputs to inputs. Outputs should be usable as sources for other programs (VAC-like). Synchronization (including FM transmitter synchronization etc.) should also move to this sound card framework. GPS clock suppost for SFN (??? Stream must have timing info as well, see **). VLC backup streams should not stay active to avoid expensive data bundle usage, and multiple ST instances must be supported in Jack. See mail Aaron Jowitt Jan 1st 2015 for another description. Joop Krauthausen: Wants to have 2 stream inputs from 2 different studio's and a way to switch between them. Turning both on would currently probably work if one is silent (but it will keep reconnecting if it's not there which is bad); also something must be done to synchronize them. Currently we synchronize on the output but this looks like we also need to do it on the input! Scheduling or switching with a command (?) must also be possible. DIFF output must be routable to a specific sound card and then overrule what's sent there normally (so in a card with many I/O's you can have 1 output for streaming and a 2nd one for monitoring, same audio, but with DIFF sound when it's turned on somewhere).

Also, this should be fixed (reported by Gareth Parker in an email):

The problem I’ve encounterd with the new ubuntu pc is stereotool automatically connects the jackd line in from pc soundcard to the stereotool input on stereotool startup. I have to manually disconnect the jackd connections from line in to stereotool and then run separate connections from the output of vlc player (also on same pc) to the input of stereotool and it works until the next reboot. I created an entry in the jackd patch settings to try and automate this and it works to a certain extent, automatically creating connections from vlc to the input of stereotool on jackd startup but steretool still creates the connections from line in to stereotool input automatically on top of the connections from vlc.

Also to be added to this: Profanity delay (drop input buffer on key press, and then slowly build it up again, probably by slightly changing sample rate, or maybe just put a jingle in, but apparently many stations don't like that).

Vlc used to also automatically create its own jackd connections from its output to the soundcard output, but there’s an option to disable this in the vlc settings, however I can’t see any similar options like this in stereotool, maybe it could be considered as a feature in the next version, if not then i’d be willing to pay you if you could and do a special build for me, so stereotool doesn’t automatically create jackd connections on startup.

So, in case of using Jack it should reconnect to the name it was connected to before, and if that name isn't present keep trying.

When the buffer size of a specific sound card is changed, the rest of the audio should not glitch - this is something that might for example be used for stations to synchronize their HD signal with the FM signal and they typically would adjust it regularly.

** Let's say that we know that a specific sample needs to be transmitted at exactly 22:00. Then we need to know how much latency this sample will get due to processing (for example 4096), and buffering, and synchronization resampling, and ASIO block sizes etc. Besides this, there might be a system-dependent constant offset (due to hardware, DAC etc). An extra issue is that the framework cannot do anything with the FM output signal, so any adjustments need to be done on the input side.

Detect and warn about bad sample rate conversions! (Windows)

----

PERFORMANCE MEASUREMENT RESULTS (values shown in GUI, so on one core, version 7.84):
- Nothing: 0.1% With Overhead (top CPU load indicator): 0.6%
- AGC 1 band: 0.4%
- AGC 2bands: 1.1%
- AGC 3bands: 1.4% - So 0.3% per band, 0.4% extra when switching on multiple bands. That 0.4% must be the cleanup filter.
- Stereo Boost: 1.6%
- ACR Stereo: 0.6%
- Stereo Image: 0.3%
- AZIMUTH: 0.4%
- ND 9 bands: 2.1%/2.2%
- Noise Gate: 1.7%
- Declipper: 2.3%
- Phase Rotation: 0.2%
- Phase Delay: 0.3%
- True Bass 2 bands: 1.6%
- MB1 9 bands digital: 4.2%
- MB1 9 bands analog: 2.5% - Weird, uses more than ND!!! That's unexpected since ND is based on Digital mode.
- MB1+2 9 band analog combined: 4.8%
- MB1+2 9 band analog split: 4.8%
- MB2 9 bands analog: 2.9% (raar)
- Bass Boost: 0.8%
- Singleband digital: 1.0%
- Singleband analog: 0.6%
- Advanced Clipper: 3.7% (met de-esser?)
- FM RDS/Stereo: 0.4%
- Advanced Clipper: 3.5%
- FM + Advanced Clipper: 3.7%
- FM + Advanced Clipper + Pre-emphasis: 3.7%
- FM + Advanced Clipper + Pre-emphasis + Stereo + RDS: 3.7%
- Same with Composite Clipper: 10.1%
- Strictness at 5 instead of 8: 7.5%
- With Stokkemask: 13.3%