Thanks @Ghost Boy!

This is nice bro!

Ready

@Eleven 11 Studio Storage is not the problem, I guess, these days. It's a dropbox backend, and I think that's amazon in the end. You need to store the wav files anyway because of the pull feature. blends take less space because they share the same files. Still they process after you upload, so thats a hint. I bet on some looseless encoding. The bandwidth might get a problem, but its not millions of users. Does streaming count as broadcasting? Then it will be changed. If you have a way to record the stream you can find out. :)

THERE MUST BE SOM DITHERING AS IF THEY WERE TO TAKE EVERYBODY'S FULL WAV FILE, THE SITE WOULD REQUIRE TONS MORE STORAGE

@Eleven 11 Studio I think all the pictures on blend are smoothed in the first place. This is nice because it gives the page a nice look and my covers look better than the original looks. This is what you do with actresses in film as well, right :D And with the waveform, you have the problem of a lot of data point vs a limited resolution on the screen and the question arises how to reduce them, there is plenty of ways to do this. Second, it is a question of showing the integral under the curve or not, this gives it a more bulky look. And third, I am not sure how they process the audio, so what you see might be the curve after some limiting, scaling or sample rate reduction. Anyway, now I am geeking out, it is always a reduced representation, one is tempted to be trained on connection with certain graphics and what one hears. I think in the end it is more a question of which representation of your music you are conditioned and if you compare things by visual clue you should use the same method for both songs in the first place and the same song might look different in different programs.

Enjoy reading all this comments from @Davilone and the replys. Good technical conversation, I have been using Ozone 7 also. Great piece of work that is. By the way, have you ever noticed that the waveform in the Blend project causes the song to visually appear way more smashed than it sounds or than it appears in your mastering software?

So I found it, it is called encoder checker and is located in the post processing field in the master section, which I do not see until I go full screen on my old macbook alu. You can realtime preview up to 3 codecs, BUT you cannot see the stereo field reduction and because the plugin is last in chain you cannot visualize the stereo field, so you need to render to see the difference. This track looses a bit in mp3, ACC iTunes does a better job. Steinberg make a stereo field preview in the encoder checker happen ! ... you can still hear it on headphones. Try it yourself, just pull and make a mp3 or acc of the wav in iTunes and listen ;)

@Davilone thanks man, I really like your tracks! I think the other thing that gets messed up in encoding is the stereo field, so if you have a very wide track it gets chopped off. I think you can check the encoding in wavelab as well, but its a new feature I am not used to, still learning v8.5. There is this fancy new LU meter which tells me the track has 18.4 LU, the scale ends there so I guess it is officially perceived loud to the human ear :D ... btw, I check the stereo field with ddmf stereoorets, this is another thing that can be done in izotope I guess, had a demo once .... and I mono the track with that somewhere below 250 Hz.

Crazy tune man! I use Ozone 7 for mastering and at mixdown I usually leave about -6db headroom and then use the limiter/maximizer to push it up to -0.3 or 0db. As @reiswolf menitioned if you are going to encode your track you should leave some headroom like -0.3 to -0.6. in Ozone 7 there's a cool feature that lets you preview how your track sounds when encoded and different bitrates etc...

The other thing I forgot, if you go in mp3 or other compressed formats, leave some db on top, like 0.8 db. Or the encoding gets messed up. Check out recent videos by izotope on youtube, one of them mentions it, I forgot which and I don't have any izotope (yet).

Ok, that s the question, in short, in this track, I did: 1. The drums are technically clipping, as you can see on the raw samples, they run into 0db, but in a good way, I do a trick from a tutorial from RESO, who mentioned using camel crusher/ camel phat, which makes the drum transients stop at zero. Not talking about how to make them sound good here. This is just how to get fat dnb style drums. 2. Then you need to get rid of everything that sticks out in loudness, use spectrum or voxengo span and have a look. There is a clipping counter in span, useful right! EQ: roll of things you do not hear, higher 20kHz and below 30Hz or something. Just make sure that everything has its place in the spectrum and nothing sticks out or overlaps too much, like e.g. sub bass and kick. 3. Compress, to suck it up in volume, find a setting that keeps the transients alive. 4. Don't worry, the channels in ableton can go red because of the floating point precision. On the master you should have no red, so I (and now I get burned by a lot of people) put a gain plugin to get some space back, like 3-6db, and than use a stack of a couple of limiters which all work a few db off to top and keep it loud. Check ddmf NoLimits (few bugs), free W1 limiter. And then use good monitors and your ears and compare to (also clipping) commercial reference tracks. 5. In the end I use wavelab to detect and correct my sins and if it sounds good I leave it, if not I correct and do it again. Thats about it. Some people do it different, this is how it is done here.

Awesome. How do you guys get your projects so loud without clipping?!

Thanks @Monochrome Echo ! Looks like alot of people got inspired by the plugins ... good job soundtoys ;)

Fantastic track @reiswolf! Working some more on my entry today . Should be up shortly...

@Eleven 11 Studio Thanks! Glad you liked it!

Fun tune man. I like it. Very original too. I am about to load my entry

@ElectroBongo Now you know why I like resampling :D ... seriously, the drum kit is resampled and I needed to put it in another project, because of the processing, its all FM ... :P ... the idea is a lot of small processing makes a better sound than a few maxed out plugins. But whats the point in making a plugin based challenge and not seeing the plugins work. @Hartmut Schmidt ... check out ElectroBongo s stuff, it is probably louder :) ... So the project for me is actually three ideas: sound toys meets FM synth meets Loudness processing.

Holy mother load of VSTs. I counted 22. It would be fun if the Blend cred board had a number of VSTs category. You'd be #1 this week. : )

Wow, that is loud indeed! Good luck, man! :-)