What is the ideal Loudness Meter average for AudioJungle?

I will check it as well) Well, yes, i wasnt pretending, i just thought these parameters were right, since this is what Ozone 7 showed me. I am switching to FabFilter now, also need to read more about RMS in digital dynamic processors (Voxengo SPAN also has some different RMS readings by the way, shows -3dB less by default etc.). Theres always something to learn)

I am more concerned about crest factor now, sick and tired of these peaky drums and low energy body of other instuments (some mixes are fine others skinny as hell) All the best! :sunglasses::grinning:

Normalization would be awesome! ++1

oh no no, @PhreaSpirit, I was not referring to you or to this forum :slight_smile: I was more thinking about some specialized music forums where some people tend to measure the, well, loudness of their productions :wink:

So, we do get different readings depending on the meter we use… well, that sure is practical!

Getting -5dB RMS when another plug-in will show you -11db RMS is pretty messy. How can we know? What can we trust? Why would Isotope, a major player in the mastering realm, use a meter that gives inaccurate readings?

Is there also a discrepancy for LUFS values, depending on the plug-ins, or is is just affecting RMS?

i always use magic a/b for metering its easy to spot where my track goes, most of reference track i use always hit -6.5 to -5 db and because of that i always crank bus limiter and compression to reach that number but only manage -7 to -6.8 why it is so hard to reach such loudness. as far as i know buyers likes loud tracks.

I think it’s just too difficult. Imagine to go through 600 000+ .zip archives and normalizing everything. It may be over 2 000 000 files. Insane :slight_smile:

I don’t know - I tried many different meters, and they all give the same values…
(dpmeter, internal Reaper meter, youlean, melda…)

I’ll check ozone to see what the readings are and come back to you.

Wow, I’ll have to investigate, I have the same strange reading as you with Ozone, it shows insane RMS levels… with a quiet piano piece, it show -7db (instead of -14…)
Something is clearly totally wrong here!

EDIT: ok, by default, my ozone was configured to display… the max values!!, and not the current! Also, the reading is “instant”, with the classic “300ms (VU)” timing, you can increase this to get a more stable value, but if you really want the “integrated value”, then you’ll have to use one of the other tools.

When you do that, do you get values that are closer to what other meters give you?

yes, but this is difficult to compare an instant value to an integrated one… :slight_smile:

When you are measuring the real average RMS, Ozone is not the best tool… I really recommand dpMeter2 :slight_smile:

You wouldn’t have to go through the zip files, what’s done is done. Just the previews would all be normalized to a certain level and this can be done automatically with an algorithm (youtube, spotify, apple, all the major streaming services do this already).

And once people realized that their tracks were all going to be normalized at say -12 LUFS Integrated across the site, they would stop feeling pressured to absolutely SLAM everything through a limiter to compete and everything would start sound a lot better!

I think when a customer buys a track, he expects same file without watermark, but all of a sudden, the track is twice as loud. It shouln’t be that way.
Anyway, I just guessing. If this process was easy, they would’ve done it long ago.
By the way, I don’t like overcompressing my tracks too, just so you know :slight_smile:

There is a very nice read here + video: http://productionadvice.co.uk/lufs-dbfs-rms/

(skip down to “Appendix: Get your RMS right !”)

Ozone is actually correct and passes pink noise test…at least according to the article. I tested it as well, also dpmeter 2, mvmeter2 and Voxengo Span ( in AES17 +3dB mode). All are correct (roughly-11.5) . FabFilter L1 was “wrong” at about -15 dB. Dpmeter 2 can show the same numbers in “RMS averaged” mode, so probably that’s the algorithm that FabFilter uses by default and you can’t change it.

When it comes to music there’s a difference but not like -11 instead of 5.5, it is about 3 dB max difference depending on a track… Ozone tends to show higher numbers, yes (it measures an average value over very short time i guess), but i think you will be in a sweet spot by setting integration time to 1475ms in settings. By the way Span & DPmeter 2 show almost identical RMS values. So if you wanna play safe then go with dp2 or span.

Some plugins have pre filtering, others use non-sine waveforms etc. So i guess there’s always some differences when it comes to RMS… just my observations so far, hope it helps.

2 Likes

Hey People. Did anyone get a clear answer from AJ about how loud our tracks should be yet?

I saw this video today about LUFS levels for Youtube and some other streaming sites. I hope you find it of use.

2 Likes

I don’t think there’s a definitive level suggested (or applied) by Envato for Audiojungle. This video is still good for services with some kind of normalizing going on, though. :slight_smile:

On my side, I simply go by ears (and compare my levels against the competition from time to time). I try to stay relatively conservative and rarely squash my sounds flat, because I think it’s important to leave a bit of room for the clients (so they can still adjust levels to fit their needs).

Ps: I mainly do sfx, though, so maybe I’m less affected by loudness wars.

3 Likes

I go as loud as possible. I don’t care much about LUFS, I care about dynamic range (PLR). Some genres need to be loud. The most squashed my masters get is 8-7 PLR and I have observed that values on the AJ best sellers also . Less dynamic range than that is too compressed for me. Nevertheless plenty of commercial music is way more squashed than that, having a lot of success. Is better to focus more on the mixing rather than the loudness. A bad mix will always sound lifeless and over compressed when you push the limiter.

1 Like

I heard that you can make a louder amplified frequency, and then MIX will seem louder, but the rms level may even be -9 -10 that is not too compressed …

Guys, and who knows this is really true, the less tools, so the Mix will be louder? What is the reason for this?

It’s absolutely relevant issue to sfx! I’ve noticed that super loud files are kind of a fashion. Ambiences also. Which is nonsense, but hey… type of sound in relation to dinamic range, nature and character of audio visual work, new aesthetics and ways of listening dictate the RMS/LUFS levels. After all this is a stock site. Sort of a fast food approach to sound. It can be tricky tho. When your loud files are gonna sell better then i say - go for it. Serious work - keep the levels where they should be.

2 Likes

I’ve noticed that too. Pretty annoying to come across ambiances like “gentle morning rain” or something, but maxed out in the red (or even distorting in some cases!) haha.

2 Likes

Exactly :slight_smile:

1 Like