I did use the Toneboosters Ebuloudness plugin, but I guess I don't know what I'm doing. I tried to match the integrated loudness on each song, which was around -10. I don't understand the K meter thing at all...which one is average RMS? All of them? Which should I use K12, K14? I ended up using the integrated loudness along with switching the Sonar meters to RMS only on playback, and then used my ears for final adjustments. I'll try the Voxengo plug, but I would really appreciate a little help understanding this better. Thanks, (and please play with that Limiter 6 plug some, I think you'll all like it, if you use guitars, bass and drums in your music).

Ok I think I'm getting it now. Lu k14 seems to work best for me with Toneboosters. Barricade has the K meters too. I'll google some more.

I'm trying No. 6 on a project and finding it has some interesting stuff going on. Working with a very pulsing bass track and I'm able to ramp up the limiter and get it pumping in time with the track. I find the pointer knobs a bit clunky and mostly ignore the meters. I thinks it's a cool alternative to similar gear. Thanks for building it guys.

Beware of trying to match average RMS as a way to make all your songs sound equally loud, because it will only get you into the ballpark. A recording with good song dynamics could vary by as much as -24db to -8db average RMS.

The K-system, in a nutshell, is a way to correlate average RMS to what you hear out of your speakers. It requires you to calibrate your monitors so that you're monitoring consistently. Bob Katz's website has more information, or better yet pick up his classic book "Mastering Audio".

If you settle on K-14 as your standard (a good choice), put SPAN into mastering mode ("Mode" button, upper-right) and select the K-14 meter scale ("Metering" button, beneath the display).

As you play back your song, watch the meter readings (which will bounce around, that's a good thing!). You want the bulk of your song to bounce around the 0db mark, usually going no more than 3-4db above it in the louder passages. I never let it go above +6db, and then only briefly. 

I'd recommend importing some of your favorite commercial recordings into SONAR so you can observe what SPAN does while they're playing. 

Thanks Bitflipper! Good tips, and I'll try them all. Last night I had a few minutes so I just checked each song and they are all in the +2 to 3 range. With The Toneboosters plugin Ebuloudness there are 3 measurements LUK14, true peak and loudness range ( I think). I was setting there thinking that if there was a wider range it would probably effect the LUK14 measurement some...for example. If one song is more dynamic and measures +2.5 K14 it is probably close to a song thats less dynamic that measures +3.

bitflipper


If you want to compare all your songs at once (along with any reference recordings) with a single tool, Toscanalyzer is very cool.

Cool indeed.  I'll be downloading asap.

The EBULoudness plugin is technically more accurate than SPAN, but it's intended for broadcast applications. I did not find it particularly useful for loudness assessment, and as it's a bit of a CPU hog I don't use it. For a while I was using the TT Dynamic Range Meter, but after a year I went back to SPAN. The offline TT DR Meter is still handy for scanning a bunch of files at once, but lately I've been using Toscanalyzer because it gives more detailed information.

At the end of the day, though, my ultimate tool is a combination of ears, ATH-M50s and a quality MP3 player. The player shows a numeric value for the volume setting, which, though meaningless by itself, works as a means of comparing my stuff to commercial recordings and to my other material. It's a kind of crude K-system, in that it enables auditioning songs at a consistent, predictable volume level.

My recent purchase of a set of M-50s may be the single best use of my dollars in quite a while.  They are tres fab.  Astounding, really.

My last ventures I didn't have a sub at all and used just my M50's to make bass decisions...definite improvement.

For my fellow ATH-M50 users: I have recently been applying an EQ curve to my master bus to help compensate for these particular headphones. It's a small correction, based on the frequency response published at headphone.com.



My correction curve is basically a mirror of the above plot:
-4db at 45Hz, Q = 4
-4db at 130Hz, Q = 4
+4db at 5.5KHz, Q = 16
+4db at 9KHz, Q = 16       [EDIT: should be -4db at 9KHz, not +4db; thanks cecilius2!]


Yes, the graph would seem to indicate a need for larger corrections, but I don't know how accurate their measurements are or how much variance there is between individual units so I've elected to be conservative.

Just don't forget to bypass the EQ before exporting!