People who are happy with iTunes or mp3 don't care how bad it sounds. Why wreck your mix just for them?

Soundcloud sounds pretty good to me.

You always mix in the highest uncompressed quality you deem necessary from .wav files of 16 bit 44.1 kHz through to 24-bit 192 kHz. Once you have your mix sounding superduper cool and you have created your CD ready master at 16 bit 44.1 kHz, only then should you consider compressing it to whatever formats you deem do the least amount of harm to your sound. MP3's will always have their inherent artifacting as will most any other compressed format. Some, like .WMA (Windows Media Audio) compression algorithm to me have a better quality of sound at the same compressed bit rate over 128 kb per second MP3's. Whereas MPEG 4 & Flac might even be a tad better.

Many people moan and complain about additional signal processing or volume leveling that many of the online music services seem to inject into your audio. I don't care for that either and have no idea why they do it? One should realize however that if your mix is a good mix and recording, it will still sound good regardless of what awful things highly compressed file format algorithms create. You might not like what they do but the beauty should still shine through. This sounds like a shampoo commercial from the 1970s LOL. Maybe I should produce a movie about hair and call it " Breck Back Mountain "? Simply because L'Oreal Back Mountain doesn't sound quite right.

While audiokid suggested Soundcloud (Which I too like since no other processing seems to be added), that's not necessarily a download pay for service. It's just a community streamer. Though they might have some kind of pay for service also? I haven't looked into that closely enough to know. But that still negates the question about what compression algorithm will keep your music sounding more true. The less compression the better but the larger the file will be. So a 320 kb per second MP3 will sound better than a 128 kb per second MP3 since far less data is also being removed or compressed. Although certain sites have a requested file format they want you to utilize such as Flash in some instances. Though I regularly compress my mixes down to MP3 at 128 kb per second to make sure the bulk of it holds up still. Sure, I can hear the artifacting. Most folks don't know what they're listening to however. Sometimes it makes for a cool sound and other times it destroys the sound. So some of that is based upon musical genre. I don't want some kind of phasing or flanging sound on operatic vocals for instance. But on rock 'n roll it can actually add a cool element to the sound just from the artifacts alone. Still though it's not 100% applicable to all rock 'n roll. And I don't believe that rap nor hip-hop would be hurt by that at all? Since it's pure raw Art in its basic form & genre. Everything is a series of compromises. You don't get something for nothing. And you can't take away 80% of something and still expect it to be perfectly the same. So you have to understand how bad you want your goodness to be.

There are things that can happen in compression that can cause a definite deterioration of your audio. Some of that can come from the 15 kHz fly back transformer/horizontal hold of televisions that have been on, anywhere in your household or studio that has gotten into your audio. The compression algorithm will see that and tried to give some priority to preserving that. In the process of doing that, it can cause all sorts of horrible audible audio artifacting. I'll never forget the day back in 1976 when I was the overnight disc jockey at the number one rock station in Baltimore. The Rolling Stones had come out with their album called Black & Blue. And on that vinyl album, I could hear this definite 15 kHz. In fact it was so high on the album, it was even visible on the VU meter of those crappy CCA radio broadcast console. OMG! How did that ever get released that way? Obviously the disc cutting engineer was way over his 50s. The rolling stones engineer was probably well into his 40s and all of them had screwed up hearing from being on stage with loud PA systems and couldn't hear it either? That certainly wouldn't have compressed well to MP3's without some heavy equalization to notch out that 15 kHz that was screaming quite loudly from that piece of vinyl. Which was probably on the original master tape? Of course it could have originated from the disc cutting room if the engineer had been watching a football game on TV while doing the job? Thankfully with with the demise of CRTs that problem is quickly vanishing. Long live LCD's! Actually I'm waiting for oLED's to get a hell of a lot larger. Though their lifespan seems to be much less that of our typical inorganic LED's. LCD's also did not have much of a lifespan but there have been improvements made there. But I digress while I am depressed.

Is depression a new kind of compression?
Mx. Remy Ann David

Every song I've posted on Soundcloud was looked at by a pretty thorough mastering engineer who saw a definite 13kHz to 18kHz bump, centered just over 14kHz in the same files as emailed, ftp, web transfer and SourceLive broadcast... which all 4 of the other transfers showed no increase of noise or other artifacts.

Mine might have been an exception if it had just been 1 file, but this bump existed in 5 different files mixed at different times from different artists.

For most folks recordings, that high-end bump may do them more good than harm? My hearing must be failing? I didn't hear that in my uploads/play outs. But I trust you and your Mastering Engineer are probably correct. So it wouldn't be a bad idea for us all to upload some pink noise and look at the playback on a RTA? That would be the real test. Forget those other reel tests because that's analog blah blah.

+6 DB over 250 nano webers per meter.
Mx. Remy Ann David

audiokid, post: 384133 wrote: Soundcloud sounds pretty good to me.

Soundcloud can use flac which is a huge step away from MP3.

Very interesting, thank you for the replies.
Remy, I appreciate the rambles, they're always informative, educational and funny. :-)

So, one thing that I'm learning about and would like to ask as well is; is there merit to how much compression to add at the mixing stage based on how much saturation/distortion you've added earlier in the process. Let me try and explain:

I pretty much record and mix rock/acoustic music, rarely do I touch electronic or midi. I like the analog goo/color that I can get from my preamps as well as the good ole Ramsa console (thanks Moon!). There are times where I've mixed and kept the meters in the red because I really enjoyed the saturation that I was getting and felt that they added to the song. So, when saturation is introduced to that degree, is it logical to assume that this is in fact a form of compression to be considered in the overall compression equation?

I don't like overly compressed mixes but I do like harmonic distortion and analog saturation, so it seems sometimes when I'm running the meters in the red, I can't get away with as much compression from my outboard pieces. I suppose to sum it up, what is the relation between pushing saturation and compression (via a compressor).

Does that make sense?

Oh yeah sshack. Those Ramsa mixers and consoles were decent pieces. I've used numerous ones that I believe were discrete transistor not IC chip versions. And they have a quality similar to other quality discrete transistor consoles. In some of the gain elements, their class A at the input sections and class A/B at their output sections. Much like API & Neve. So there is a magical sounding element you can get from your board. And yes, by saturating certain elements lightly you are gaining a different harmonic structure that obviously sounds more musical when you get that sound than when you don't. So your board is creating some natural limiting from the saturation. It's rich in seconds and/or third harmonics which definitely can be advantageous sounding. Just like one of those aural exciters/enhancers provide. That is without the additional element of high-frequency enhancement since it's more broadband in nature like folks get from API & Neve. Because it isn't an exciter/enhancer per se. And so the natural compression you are getting is just that already. For instance, in some mixes I generally have a little EQ with a 1176 limiter on the bass guitar. When it's not sitting right in the mix, I've been known to pull the patch cords out of the limiter, shut off the EQ and then just slightly over gain my Neve 3115 MIC/EQ. Eh, voilà. Saturation limiting with marvelous harmonic content and boom, it sounds huge and sits in the mix exactly where I wanted it to. So these are distortion elements that actually enhance the sound as opposed to sounding just like plain old unlistenable distortion of total clipping. But it's a fine line between the two. A lot of us love to bang the meters. It's a lot of fun turning off your control room monitors and just listening to the little meter thingies click away, LOL. Then you know you've got a good sounding mix. Of course this usage of saturation is not appropriate on every single source. But it is one of those cool elementary elemental sounds of rock 'n roll. You and I work very much the same way and the personal gratification from the results is second to none. Of course there are lots of people to try to overdo a good thing or are just doing it through lack of understanding and experience. And so it's great in audio when you get the right sound in the wrong way. And you also have the pleasant knowledge of knowing that doesn't work in video with images as it does with sound. There really is no video that can be analogous to audio saturation other than just being blown out over exposed and well beyond 100% where whites clip and all resolution is lost. Whites as in the color not the ethnicity of someone. Isn't it great that in English the same sentence can be misinterpreted a number of different ways yet still get the message across so I guess I've saturated my explanation?

So to sum things up (without talking about summing mixers) natural limiting and additional harmonic oriented EQ being self added from the pure understanding of what and how much to push the electronics can achieve a stellar sound without the use of any additional gobbledygook. Supercalifragilisticexpialidocious would be a good way to describe that sound. I think that technically describes all that was added through the use of simplicity. Add to that analog tape saturation and you've got yet another sound. Sometimes members of the Beatles instructed Geoffrey Emmerich to request a console oriented push of obvious distortion. Of course some of that was over the top as they wanted a particular sound to sound distorted. But that's even still a LESS IS MORE deal of a more simplistic approach in earlier more simplistic times. So everybody walks around scratching their head wondering why they can't get that sound out of their DAW software and plug-ins? You can, it just sounds fake because it is. Whereas the real thing speaks for itself, with the harmonic signature we all want.

It's fun not making sense to everyone else here. "Distortion? Why would they want distortion?"
Mx. Remy Ann David

older topic worth resurrecting.

MadMax, post: 384170, member: 1402 wrote: Every song I've posted on Soundcloud was looked at by a pretty thorough mastering engineer who saw a definite 13kHz to 18kHz bump, centered just over 14kHz in the same files as emailed, ftp, web transfer and SourceLive broadcast... which all 4 of the other transfers showed no increase of noise or other artifacts.

Mine might have been an exception if it had just been 1 file, but this bump existed in 5 different files mixed at different times from different artists.

I mix into a master using 2 DAW's and 2 independent ADDA. Both DAW's are uncoupled. One DAW is for tracking and mixing, the other is for capturing the mixdown. Regardless of what I'm doing (recording, mixing or mastering) , I am always able to gain stage, predetermine the mix from the master where I monitor off DAW2.
Regardless of what sample rate my tracking and mixing DAW is locked at, my capture DAW is preset (usually 44.1/24).
Fact or fiction of SC changing my sound, or any destination for that matter has never been a concern or surprise with this process of mine.
I never make allowances because there is no need to make an allowance. I mix into a master which always sums precisely to 44.1. At that point, Everything is relative to the bite-rate or encode its going. It is what I hear it is on my playback device as it is online. Sequoia gets it does exactly how I expect it to sound, everytime. If if sucks, its because I suck.

So, even if I am tracking at 96/32float, I am actually judging and gain staging the mix in the destination SR and final product section of the entire session.

The capture DAW is where I have mastering software and an export. This process has enabled me to process consistent 2 track mixes 100% of the time. It makes absolutely no difference what my destination product is. My 2track always sums and uploads/ burns to CD as expected.. Works great today, however, I may change that tomorrow, who knows. (y)