INTERACT FORUM
More => Old Versions => JRiver Media Center 20 for Windows => Topic started by: lendall on July 31, 2015, 10:22:40 pm
-
Here is what I want to do: I want to export music onto my phone so that everything plays back near the same volume level. I listen to music on headphones using my phone as a source in fairly noisy environments, so I want the volume of the tracks on my phone to be "compressed" in this way from the get go, and I do not want to rely on the music app on my phone to do this for me.
(1) The analysis side of it: Based on hearing some unexpectedly very loud and very soft tracks played back on my phone, I have gone back and rechecked the analysis of these anomalous songs. What I found is that if I check a track analyzed under MC19 that has an R128 value of (e.g.) -1.28, and if I reanalize that track under MC20, I often get a wildly different value, e.g. -6.6. At what point in MC's development did the analysis settings change? Should I wait for MC21 (I've already prebought it) and then reanalyze everything? With thousands of tracks we're talking quite a bit of wear and tear on the old CPU.
(2) The Sync Handheld side of it : I always check Apply DSP effects when I run Sync Handheld, but I wonder if I am doing what I should be doing to achieve the result I want. Should I check Volume Leveling, Adaptive Volume (Night Mode?), or both? Remember that I am after a fair amount of "compression" as far as levels are concerned. (This has nothing to do with Fletcher-Munson and "volume leveling" in that sense -- i.e., I am talking strictly dB, and frequencies.)
Thanks for any enlightenment you can provide me.
-
Starting with R128 analysis: I don't know when it was changed, but I've read several times that the new way is more accurate. If you're interested in volume leveling, then you should analyze with the new method. Don't worry about "wear and tear on the CPU". There's no such thing. Computers are designed to be used.
You asked about Adaptive Volume and you used the word "compression" several times. Dynamic range compression is different than volume leveling. Volume leveling adjusts the overall volume of each track so that they have a very similar average volume level. Adaptive Volume has 3 different modes. Night Mode, applies dynamic range compression. This takes the softest parts of the song and raises their level closer to the loud parts. So the difference between the loud and the soft is smaller. That reduces the dynamic range, or "compresses" it.
You might want to apply Volume Leveling or Adaptive Volume (night mode) or both depending on what you are trying to achieve. Volume Leveling should makes songs have similar overall volume. Adaptive Volume will made the soft parts of songs louder, so they are easier to hear, but will remove dynamics.
Brian.
-
We fixed a few bugs over the lifetime of MC20 that could sometimes result in bad measurements, but that should only affect a very small amount of files, and generally the results should otherwise be the same since its introduction.
There are no changes in the analysis coming in MC21, so you can analyse them now if you want to.
-
Thanks for the very helpful responses. :) Of course the only way I can tell if an earlier analysis has been faulty (other than making note of tracks that "sound suspicious" and then reanalyzing them) is to reanalyze the entire library. Generally I find analysis more critical (and useful) for so-called classical music than for pop music or jazz, where I find less difference in levels. In classical music the dynamic range within any given track can be vast, which is why I am interested in applying some compression (I believe I am using the term correctly in this case). I am willing to give up some of the subtlety of the performance and the recording to make softer passages audible in noisy places.
Technical followup question regarding Sync Handheld function: Sync Handheld creates a cache of tracks that have been converted to match the conversion specs I set. In my case huge FLAC files (for example) are converted to "medium" OGG files and stored in the cache. My question: I set Sync Handheld to replace tracks if they have been modified. Does reanalyzing a track modify it for this purpose? In other words, if MC sees that a track has been reanalyzed, or has had tag changes, will MC do the conversion process again and replace that track when I run Sync Handheld again? This may seem like a picky question but it's kind of important, since otherwise reanalyzing tracks with "bad" analysis will not help as far as the tracks that currently play way too loud (or too soft) on my phone.
Thanks as always, and particularly for answering this "challenge question." ::)