As I've already said, I don't rate the thing hugely *in comparison to a proper mastering job*. But I do agree with Noel that the before and after examples on the site undersell it. It's more impressive than I thought it would be when I first heard of it.

azslow3
Just joking, but I have thought the whole point to have a DAW is to fine tune all aspects from input till the result. "One knob" audio creation with "One knob" effects applets for iPad/Android are well known, and the result still can be sent to LANDR for "professional automatic mastering". Is that the future?

Az if you have tried to do mastering you will realize how many variables there are to do it properly. You have to have great ears, a properly treated room, tons of experience with mastering chains of plugins and objectivity.
Besides the aesthetics and art there is a lot of science to mastering. A lot of the tedious tasks in mastering can certainly be automated which is what Landr is doing behind the scenes. And by using machine learning algorithms its absolutely possible to pull in the kind of mindshare an experienced mastering engineer would have. 
 
There is a lot of stuff happening behind the scenes for that one knob - its not just picking a preset like you might do in Ozone. Its really "listening" to the music and adaptively making decisions on how to master the music based on the actual content. In many ways its similar to the thought process an engineer would go through. Except here its using DSP and advanced math to make the decisions rather than ears. I would argue that in some cases the math would be more accurate than someones ears depending on how old the person is :)
 
VocalSync was developed using machine learning to figure out how to take a vocal and figure out the transients and stretches for example - did you know that?
 

Noel Borthwick [Cakewalk]

Mosvalve
I listened to their examples on their website and I have to say they sound horrible to me. I hope this isn't the future. I guess in the future everything is going to be microwavable.

Don't do that - taste is very subjective. If you listened to presets on most synths you would think they were all horrible :)
Try it with your own music and start with the Landr Low setting. You might be surprised how good it actually is.
If you are willing to tweak your mix to accommodate what you get from Landr its very possible to get an excellent master.

Your right about that. I was going to try it anyway with my mixes just to see. I actually hope it works well.

Let us know how you fare. Definitely start out with the low preset and if you don't like something adjust your mix to compensate. 

John T
As I've already said, I don't rate the thing hugely *in comparison to a proper mastering job*. But I do agree with Noel that the before and after examples on the site undersell it. It's more impressive than I thought it would be when I first heard of it.

Lets just say that I tried it on my last recording and on some of the tracks it sounded better than my professional masters. It in fact revealed weaknesses in my mixes that I wasn't aware of. If I had this when I was mixing I wouldn't have made those decisions. 

I did not read anything in the "Newburyport" update bug fix list related to two similar problem reports (CWBRN-46065 & CWBRN-34347) I previously posted. But after I updated, I tried replicating the issue and I could not do it anymore. So whether by design or happenstance, I guess you could say this bug is now officially squashed. I did add a "Fixed" note to the original problem reports
 
Good job Cake! 
 
Billy Buck

Noel Borthwick [Cakewalk]

John T
As I've already said, I don't rate the thing hugely *in comparison to a proper mastering job*. But I do agree with Noel that the before and after examples on the site undersell it. It's more impressive than I thought it would be when I first heard of it.

Lets just say that I tried it on my last recording and on some of the tracks it sounded better than my professional masters. It in fact revealed weaknesses in my mixes that I wasn't aware of. If I had this when I was mixing I wouldn't have made those decisions. 

And that's another good use case for the thing.

John T
There's also an emergent middle tier of production that's not going away, I think, which is bands and artists going completely DIY, without necessarily ever becoming great engineers. Not everyone is looking for a radio hit, or even a polished sound. Some people are happy with delivering more home-spun material to a small-ish audience, and get a lot of gratification from being a bit cottage industry about it. And good luck to them.

I've mixed exactly one song in my career (from a set of poor recordings my daughter gave me that she'd recorded in her square untreated bedroom).  My mastering stage consisted of turning on the Concrete Limiter on the master buss to fatten the song up a bit and bring it close to 0db.  I'm sure that there are some common denominator things mastering engineers do that could be (often) handled by a general purpose program and that would work better than what I did.  Like translators who save some time by passing their document through a software translator as a first pass before cleaning it up by hand (but which is only worth it on the easy stuff).
 
That said, I doubt very much I'll make use of LANDR.  I'd rather pratice and learn to do it myself, for the joy of it  (and anything I do won't be going very far anyway).
 

>> The “Give all Keystrokes to Plug-in” function was not working correctly with some plug-ins (MeldaProduction)
 
Now THAT is a freakin' reason to upgrade, right there.
 
 

Noel Borthwick [Cakewalk]

azslow3
Just joking, but I have thought the whole point to have a DAW is to fine tune all aspects from input till the result. "One knob" audio creation with "One knob" effects applets for iPad/Android are well known, and the result still can be sent to LANDR for "professional automatic mastering". Is that the future?

Az if you have tried to do mastering you will realize how many variables there are to do it properly. You have to have great ears, a properly treated room, tons of experience with mastering chains of plugins and objectivity.
Besides the aesthetics and art there is a lot of science to mastering. A lot of the tedious tasks in mastering can certainly be automated which is what Landr is doing behind the scenes. And by using machine learning algorithms its absolutely possible to pull in the kind of mindshare an experienced mastering engineer would have. 
 
There is a lot of stuff happening behind the scenes for that one knob - its not just picking a preset like you might do in Ozone. Its really "listening" to the music and adaptively making decisions on how to master the music based on the actual content. In many ways its similar to the thought process an engineer would go through. Except here its using DSP and advanced math to make the decisions rather than ears. I would argue that in some cases the math would be more accurate than someones ears depending on how old the person is :)
 
VocalSync was developed using machine learning to figure out how to take a vocal and figure out the transients and stretches for example - did you know that?

Noel, I am not in position to judge the quality of this "tool". I just do not like the model it is deployed and I find a bit curious that developers of another tool which is close related to the task are promoting that "easy way".
 
Right now I am siting in the center where thousands of people working with simulations, pattern recognitions, multi-parametric optimizations, DSP algorithms and so on. I know what it is and I know that is not an easy task. I wish we could have LANDRes for some our tasks, people could sit at home (organizations) and there will be no problem to find a table during lunch time