Mixing Advice Sites & Services

Can LANDR Replace Your Mastering Engineer?

LANDR

Drag-and-drop online mastering is here, and it’s free to try. LANDR provides unlimited 192 kbps mp3 masters of your tracks in seconds.

If you like what you hear, you can pay for uncompressed 16-bit .wav masters. Pricing is very reasonable at $9 for four or $19 for unlimited masters per month. Paid users also get to select the “intensity” of the mastering: low, medium (the default), or high.

Their algorithms were refined over eight years of university research, and they even have a resident astrophysicist. An astrophysicist!

Guess this mastering engineer is out of a job, right?

To find out, I selected tracks from three recent mastering jobs, to compare my results with LANDR’s.

LANDR

Three notes before we get our hands dirty in this mano-a-microchip match-up:

1. Am I biased? Perhaps, but who’s more qualified to evaluate a mastering service than a mastering engineer? Let your ears be the judge.

2. Louder pretty much always sounds better to the human ear. While differences in volume are important for the purposes of this evaluation, you should try to match playback levels when comparing the examples for sound quality.

3. This is an apples-to-apples comparison. I paid $9 for four uncompressed 16-bit .wav masters, to compare to my uncompressed 16-bit .wav masters. All files were encoded to mp3 at 320 kbps using LAME at the highest quality setting.

Wideband Network “Reach”

Wideband Network - Reach

I was asked to give it “the full EDM treatment,” which I interpreted as, “make it loud!”

Here’s a taste of the chorus, unmastered:

Reach (unmastered)

Here’s LANDR’s master at the default (medium) intensity:

Reach (LANDR medium)

And here’s my master:

Wideband Network "Reach" (Resonance Mastering)

Is it just me, or does the LANDR version sound like it was mastered by an astrophysicist?

To my ears, the biggest problem is the lack of low end. The result is thin and narrow and just “off.”

Keep in mind, LANDR uses the same algorithm for all genres of music. This tonal balance might be perfect for folk or classical, but it doesn’t cut it for EDM, hip hop, or even pop.

On the plus side, I appreciate that LANDR doesn’t win the Loudness War by default. There’s plenty of dynamic range. Unfortunately, the track would be the quietest in any EDM playlist. It’s simply not club-ready.

Let’s turn it up to 11 and try again at “high” intensity (a paid option):

Wideband Network "Reach" (LANDR high)

Louder? Check.

Better? Not to my ears. There’s more of everything I didn’t like in the previous version – the thinness is more pronounced and the lows are even more lacking.

It’s easy to trade low frequency energy for volume. The challenge is achieving both.

Worse, this version is even more compressed (as opposed to peak limited). You can see visually how little dynamic contrast there is, compared to my master at pretty much the same volume.

Broke for Free “Summer Spliffs”

Broke for Free - Petal

Tom Cascino’s tracks feature a characteristic richness and warmth, with plenty of deep bass. Within days of mastering his album, the lead track was #1 on Hypem:

Broke for Free on Hypem

Here’s the unmastered mix:

LANDR at high intensity:

And my master:

Maybe I’m reading too much into the song title, or maybe it’s the fact that Tom and I both live in California, but to me “Summer Spliffs” captures that feeling of cruising down PCH with the top down in August.

Does that character come across in the LANDR version?

London Exchange “When Doves Cry”

London Exchange - Re-Mix/Re-Model 25YRS

This cover was pulled off an old DAT recording for a rarities and remixes release.

Unmastered:

LANDR at high intensity:

My master:

In this case, I find the LANDR version to be ridiculously bright, edgy. and essentially unlistenable.

Also note the break at 0:13 where it’s supposed to drop off in volume and build back up. The “quiet” part sounds just as loud, if not louder, than the “loud” part!

This track might have sounded better at the default medium intensity, but I ran out of credits. Regardless, LANDR has no way of knowing that it’s a ballad, and therefore doesn’t require the same RMS level as a club track on the same release.

Speaking of which, try playing the three LANDR high intensity samples one after the other. Would they sit together nicely on the same album? No way!

There are huge tonal and volume differences between masters at the same intensity level, relegating LANDR to one-off singles.

CONCLUSION

To be fair, LANDR is an incredibly ambitious project! It’s amazing that it performs as well as it does.

The algorithm will get better over time, but it can never replace a professional mastering engineer, because it lacks musical understanding.

It can’t know whether occasional high frequency bursts are vocal sibilants that demand de-essing, or cymbal crashes. It can’t tell if the excess energy at 200 Hz is the characteristic warmth of a rich fretless bass, or vocal mud that needs to be cut. It doesn’t even know what genre your track is in. One size fits all.

Most importantly, it can’t tell you to go back and fix your mix!

When I hear a problem best addressed in the mix, I ask the client for changes. That applies to everything from excessive sub bass to thin guitar tone to ultrasonic synth spikes to questionable vocal intonation. Maybe it’s coincidence, but my clients’ mixes tend to get better with every release.

Your mastering engineer can be, as Chris from London Exchange puts it, The Fifth Beatle. We are partners in releasing the best records possible, which often extends into areas beyond mixing, like promotional advice and track sequencing.

I think my job is safe for now.

Have you tried LANDR? What did you think? Let’s talk about it in the comments.

You Might Also Like

75 Comments

  • Reply
    Dan
    June 13, 2014 at 8:34 am

    Ugh, I heard about this and assumed there is no way it could work properly… Sadly now there will be 509 more mastering companies who utilize this.

    • Reply
      Brian Hazard
      June 13, 2014 at 9:14 am

      And dozens more “I will master your track for $5” offers on Fiverr.

    • Reply
      Jon Cooper
      June 24, 2014 at 4:17 pm

      I hate to say this but that is a great idea!

      • Reply
        Brian Hazard
        June 24, 2014 at 5:49 pm

        Evil! 😉

      • Reply
        Andrea-Cichecki
        August 7, 2014 at 2:44 am

        Ah yes of course this is just like a MC Donalds ‘cheap and bad’ … has absolutely nothing to do with mastering and music. Just fast and cheeeeeap bullshit whahahaha and then the ‘Provides unlimited 192 kbps mp3 masters of your tracks in seconds. “What a funking bullshit !! Everyone knows that this quality is to cry. !!!! ??? Decent mastering costs money. Consider alone that the power consumption of a mastering studio costs lots of money du the analog gear ….. we are not even talking about the engineer who work at your track etc. Think people!!!!, keep it real and the quality up!!!

        • Reply
          Brian Hazard
          August 8, 2014 at 1:45 pm

          To be fair, the 192 kbps mp3 is just to audition the master. Most people are happy to pay a little for the uncompressed version, but this way they can try before they buy.

          Haven’t been to McDonalds in years. I’m with you on that one! 🙂

  • Reply
    Eric Oehler
    June 13, 2014 at 9:08 am

    I can’t say I’m even a little surprised.

    Although I *am* a bit surprised that it wasn’t a complete trainwreck.

    Right now, this is going to be the kind of service used by “some kid who wants to make a demo.” But I say in about 10 years, these algorithms will be getting a lot smarter. It’s not a huge leap to say that some sort of intelligent software algorithm will be able to find common trouble spots like sibilance or frequency balance problems.

    And with R128 and similar loudness standards being implemented more adequately across platforms, it’s going to get easier for systems to analyze integrated loudness and adjust accordingly.

    More than likely, though, this sort of stuff will trickle down into the tools that project studios use. Gosh, why send your tracks out for mastering or learn how to balance your levels and dynamic range when you can buy a $249 iZotope OzoneAutoFix or whatever?

    (And Sonny Moore will still drive his tracks too hard)

    This is just the first shot in a looooong battle. It’s going to be up to mastering engineers to position their services as something people *need* that they can’t get with a button push.

    • Reply
      Brian Hazard
      June 13, 2014 at 9:21 am

      Spot on as always Eric!

      Professional mastering tools will require less and less input as services like LANDR offer higher degrees of control, until they converge and become standard in the render options of our DAWs.

      God help us all if there’s a Skrillex preset. 😉

      • Reply
        Almark
        June 14, 2014 at 8:15 am

        Shrillex? You mean the -3 RMS blast of RAW 0’s and 1’s, dreadful. This new technology might be amazing, but you’re right, you can’t replace us ear people, we mix and master and no software can ever do it as good as human ear.

    • Reply
      allymacd
      December 12, 2015 at 2:27 pm

      A shit sound is a Train crash!! surely??? I’m glad it was crap.. i Genuinely thought it would be a hard one (for a Landr Type Service) to use “an Algorithym ” to follow all types of “nuances” Its hard to figure some interlocked frequencies in a mix, when you know what your listening for !! No way an algorithm can sort out music, Just like Plug ins.. they “work” but don’t sound like much.. unless (i find) using a lot of work/time.. Maybe in time some Genius will crack a relatively good working system , but it will never help, finishing up, good new inventive music..its always evolving..Wheres an Algorithm for that ???

  • Reply
    Mariea Watkins
    June 13, 2014 at 1:50 pm

    Wow, I must say that it’s interesting. Yet, it’s still “robotic” to my ears. The subtleties and nuances of music, I still enjoy, and yet must realize that competitive level wars are still present. So, thank God for the mastering engineers who understand both to deliver the best of what music needs!

    • Reply
      Almark
      June 14, 2014 at 8:48 am

      If guitars sounded that bright I would turn off the music in 20 seconds, very fatiguing. There is so much more to mastering than this software, what about the strange phasing problems that occur, mid/side mastering, things that no software can correct? I would say this is basic for people who just want to get their cut on the radio, or a quick fix, but its not professional, not by any means.

      • Reply
        Brian Hazard
        June 14, 2014 at 9:59 am

        True, things like mid/side processing and multiband compression can introduce weirdness, even when used properly. As for the brightness, I just saw an interview today where the head Mix Genius said music today should be mastered for laptops and SoundCloud, not for hi-fi systems. That explains a lot!

        • Reply
          Almark
          June 15, 2014 at 11:32 am

          But what about the people who actually have better systems?

          • Brian Hazard
            June 16, 2014 at 8:17 pm

            Apparently LANDR is not designed with those people in mind.

          • Almark
            June 18, 2014 at 12:40 pm

            yep, the whole word is hearing music on laptops and phones.

  • Reply
    Jon Cooper
    June 24, 2014 at 4:12 pm

    This is the fist time i listened to these comparisons in my studio.

    I am quite impressed by the capabilities of Landr. For the quick brightening up of a track its great. And for the speed and the fact you can keep doing it over and over for very little cost is fantastic.

    That said. It does brighten up the track as a WHOLE, it still remains flat sounding and lacks some stereo fidelity, which less noticeable on headphones.(Which i use a lot)

    As for the 11 max intensity, I don’t think i would need that for a whole track i can kill it myself! But it may be a great application for mastering individual tracks within a track.
    ONE Track which may of been mastered on MAXX OVERKILL setting is Twinkle – Tu as Perdu . Which i play in a set at about 14 mins 30 if you wish to listen to the application of overkill to a track.
    https://soundcloud.com/joncoopermusic/jon-cooper-lexis-mix

    It is clear, and i don’t want to blow your trumpet Brian, but the trumpets of all sound engineers and producers. You cannot get the human touch from an algorithm on a computer.(yet)

    In your mastering its clear that there has been some mindful focus on key elements on the tracks. You have focused on the Low end drums and the rhythm and also the vocal where you can really notice the difference in stereo fidelity and clarity brought about by you i assume.

    As you know Brian, i am mastering 3 versions of a song at the moment and I’m interested to know, in your final master what is an acceptable level of clipping?

    Im at 4.5 over 6 mins ( down from 11 LOL) and it is starting to sound “ready”. At what point do you say done? I wonder if any writer, producer is fully happy with something.

    • Reply
      Brian Hazard
      June 24, 2014 at 5:54 pm

      Thanks for the kind words Jon!

      Personally, I don’t think LANDR is acceptable for professional use in any application – even a track within a track (apologies, but I didn’t click through to your link because I’m not in the studio, and because audio quality is hard to judge through SoundCloud’s encoding anyway).

      As for clipping, some mastering engineers will clip a little off the top right from the start. I don’t do that, and set a -0.1 dB ceiling in my limiter, with 4x oversampling, so I never get clipping or intersample peaks.

      Rather than counting the number of clicks, you might look at average RMS level. It’s closer to how we perceive volume, but not perfect. -12 is pretty dynamic for masters these days, but -8 or even -6 isn’t unheard of.

      • Reply
        Jon Cooper
        June 25, 2014 at 3:37 pm

        Thank you Brian,

        As ever extremely useful tips. Maybe my music is closer to the Skrillex genre where RED is GOOD! 😉

  • Reply
    stroeeralexander
    August 10, 2014 at 9:17 pm

    Take musicians with no money – either for hardware to properly master on, or to pay $$$ for pro masters. Now they have a fighting chance to produce a demo that can pass as commercial quality. Internet music is compressed to death so 192bit just worx for soundcloud and such. If we go commercial release, the uncompressed will go a long way but ideally everyone hopes to get a pro at one stage. Truth is, small artist starting out, no pro, so what can you do. 🙂 As for the sound quality: A/B-ing shows you deficits, yes, but the point is would you have sincerely noticed them WITHOUT A/B-ing, or would the everyday listener buying songs would have figured it? So, this is all about starting out, once you hit big you dont need it anymore I suppose.

    • Reply
      Brian Hazard
      August 12, 2014 at 3:55 pm

      I don’t think the examples I presented in the article can pass as commercial quality.

      More to the point, I’m not sure these “musicians with no money but have tracks ready for mastering” really exist. They either have gear and a room to record in, or they are paying someone else to record and mix their tracks.

      Whatever the situation, $99 to master a track (my rate, obviously prices vary) is far less than what it costs to record and produce said track.

      • Reply
        Robbie Belchamber
        March 9, 2015 at 3:53 am

        Just wanted to say that these “musicians with no money but tracks ready for mastering” exist absolutely everywhere.

        I’m a student, and myself and plenty of my friends produce music totally electronically on software which can be easily obtained online, albeit through slightly morally questionable methods (torrenting). This costs absolutely nothing, and with simply a laptop and a good ear it’s definitely possible to produce a track that could be sent for mastering.

        And gear and a room to record in costs so much less than it used to – at $99 a track, that’s the price one of my friends set up a cheap-as-fuck home studio to record a whole 10-track album in. Without landr, these get released quiet/distorted and clipped if we try and master ourselves as we have no knowledge regarding it at all, and the shittest gear available.

        Easy for people at the top to deny the benefits of technology that is only useful for those far beneath them.

        • Reply
          Brian Hazard
          March 10, 2015 at 3:29 pm

          Thanks for the comment Robbie!

          One could argue that software piracy is more than *slightly* morally questionable. 😉

          If you figure in the cost of software, your laptop, your interface, maybe even pressing CDs… even for a pure electronic musician, mastering isn’t a disproportionate part of the budget.

          I see your point though, that the cost of making music keeps going down, which is a great thing!

  • Reply
    Aaron
    August 13, 2014 at 5:40 pm

    To me, it sounds like the “thin” sound some are mentioning is probably a deep cut in the middle somewhere around 1500 to 2000 Hz. That can sometimes bring more clarity to a mix whilst taking away the warmth and leaving the bright only. It might work great for some tracks and kill others. Overall though, I really think landr could be quite helpful if you mix with it’s mastering approach in mind. Just my 2 cents. 🙂

    • Reply
      Brian Hazard
      August 15, 2014 at 4:22 pm

      I’m not sure if you can really compensate for the lack of warmth. My guess is that it will enforce its desired tonal balance, regardless of the starting point.

      Either way, if you’re a good enough engineer to second-guess LANDR into a good master, you’d probably be better off mastering it yourself.

  • Reply
    Conor
    August 14, 2014 at 10:13 pm

    Checked the article because I’ve met the guys from Landr and was interested to see the debate… Then just noticed your example of summer spliffs… Our BANKS remix is the #3 spot in that pic you have there…you can see it just cut off.

    Summer Spliff kept us out of #1… Must have been that human touch 😉

    • Reply
      Brian Hazard
      August 15, 2014 at 4:23 pm

      I’d like to think so! Congratulations on the chart placement!

  • Reply
    Jason
    September 9, 2014 at 12:46 am

    No. LANDR is useless.

  • Reply
    Jack Braglia
    September 22, 2014 at 3:37 pm

    wow…I’m astonished. I mean, I knew it would sound bad, but I didn’t think it would be THIS bad. It sounds phasey, overly filtered, like a much lower quality mp3 than it says it is. Its just straight-up garbage. My guess is landr will disappear once they’ve exhausted their marketing budget.

    • Reply
      Brian Hazard
      September 23, 2014 at 4:57 pm

      Well, you’re a mastering engineer too. To the average GarageBand musician, if it’s brighter and louder, it’s better.

  • Reply
    Andrea
    October 7, 2014 at 3:59 pm

    I agree with your closing conclusion. Even with the best programs it does not add up to having your sound professionally mastered. Maybe someday that will change with technology just not yet.

  • Reply
    Freeks
    November 1, 2014 at 2:15 pm

    It’s true that LANDR is not really an option yet.

    BUT i would never pay for ITB mastering. Anyone can be ITB mastering “engineer”, but every full time mastering engineer that i know or have heard about work OTB. Many with custom/self-built equipment.

    There is loads and loads of ITB mastering services these days. I do all my ITB mastering of my self, but when i’m doing anything that’s join to be released i will pass it on some mastering house.

    Last album was mastered here: http://virtalahde.com/virtalahde-mastering-en/ 60€/track and less for album.

    I also like to attend to the session if possible. Sadly not possible too often as budgets don’t allow plane tickets to New York anymore 🙂

    LANDR kind of things will be good someday. Sometime ago we still thought that it’s not possible to edit individual notes of chord. Now there is DNA and such. Gotta love technology 🙂

    • Reply
      Brian Hazard
      November 2, 2014 at 3:01 pm

      I have no problem with the all-digital aspect of LANDR. Some things, like linear phase EQ and brick wall limiting, can only be done in the digital domain. I also find that client attendance produces lesser outcomes. 🙂

      • Reply
        Freeks
        November 24, 2014 at 3:31 am

        But then again you are not a pro ME.
        You are one of the many, many hobbyist ME’s.

        None of the “real” ME’s i know make music, none. It’s a funny fact. Mastering is their main and only income. They make living with it as people trust their ears so much. Hobbyist’s are the ones that ALSO do mastering among other services and hobbies.

        There is nothing bad about as long as they don’t charge more than “real” ME’s. Many do. They can do it as they sell themselves as ME’s “that know the genre” while most pro ME’s make banging EDM masters. There is really no reason other than price to use non-pro ME.

        When mastering album the pauses between songs make world of difference in the outcome. It’s really hard to do that if client is not attending. When mastering demos to SoundCloudBandCamp it’s probably ok. Even on Spotify album release pauses make huge difference on how the album plays out.

        • Reply
          Kyle Williams
          December 11, 2014 at 8:36 am

          This guy is an idiot…

        • Reply
          Brian Hazard
          December 11, 2014 at 9:04 am

          Freeks, your comments are absurd on so many levels. I suspect you’re just trolling me, but I’ll give you the benefit of the doubt.

          Of course mastering engineers make music. Bob Ludwig, the best of the best IMHO, played principal trumpet in the Utica Symphony Orchestra. Pick a mastering engineer and Google their bio.

          I’ve been mastering full-time since 1999. 90% of my income comes from mastering, but because I find time in my mastering schedule every few weeks to work on my own music, I’m not a “real” ME? My pathetic release schedule speaks for itself. 😉

          As for the pauses, that only applies to physical releases, which are becoming an endangered species. The process is simple: I supply the client a single mp3 of the entire release, and if requested, adjust the gaps based on their feedback.

          What’s actually really hard to do is to make a good master with the mix engineer in the room! If you knew anything about the biases involved in human hearing, you’d understand what a terrible idea that is. The mix engineer should think the mix is perfect, and therefore should have nothing to say. Instead, their comments mislead and misdirect from the true deficiencies of the mix.

          Hearing is built on expectation. We hear what we expect to hear, or in the case of client attendance, what we are told to hear. And then we tell the client what to hear in return.

          Worse yet is the idea of mastering an entire album in a single session. I fine-tune over at least a half dozen short sessions. Sure, it makes client attendance impossible, but it results in a better final product.

  • Reply
    Colin
    November 10, 2014 at 6:21 am

    I grew up in the days of hi-fi and, poor student that I was, trying to get a great sounding system on a budget. That was then, when you had to go out and buy a physical thing, be it vinyl or cd or, god forbid, cassette. mp3 is a lossy format. Streaming sites, like Soundcloud, colour the music anyway, usually not in a pleasing fashion. However, this is where the great majority of humanity are now sourcing their music from. (Neil Young’s Pono is never going to be mass market surely?) Mastering is undoubtedly a black art, and at it’s best, the differences to a mix are staggeringly good! Not everyone is an audiophile though. Even back in the day, most people were content to listen to music on crappy AM radio stations, that often sounded like they were being broadcast from deep space. There will always be a market for craftsmanship and skilled artisans. And there will also be the mass market, which really doesnt care about loudness wars, poor encoding or any of that. These are strange times indeed. 🙂

    • Reply
      Brian Hazard
      November 11, 2014 at 6:35 pm

      Hey Colin!

      You are absolutely right that many listeners, perhaps the majority, don’t care one way or the other. SoundCloud and YouTube are good enough for them, and the difference between LANDR and professional mastering may not matter, at least on a conscious level.

      While SoundCloud and YouTube provide poor audio quality, Pono provides no benefit over CD. 44.1 KHz is able to capture any audio signal perfectly. At least he’s getting people to focus on sound quality though!

  • Reply
    shinmai
    March 13, 2015 at 3:55 pm

    “This is an apples-to-apples comparison”
    So okay, I tried LANDR out and it’s definitely not for me, but this is just an absurd claim to make, and one I’ve seen in almost every MEs review of LANDR I read while researching the service. Unless you actually charge ten bucks a pop to master tracks for clients, it’s in no way an “apples-to-apples comparison” just because the samplerate and bit depth match, any less than a race between a tricycle and an F1 would be, just because they both have wheels and a driver.

    • Reply
      Brian Hazard
      March 13, 2015 at 4:26 pm

      By “apples to apples,” I was only referring to the file format. I hope that’s clear in context. It wouldn’t be fair to compare a 16-bit .wav to a 192 kbps mp3. I see your point though!

  • Reply
    DigitalConvolution
    April 15, 2015 at 7:25 am

    Man oh man…

    I am underwhelmed by the understanding showed by the many comments here that music can be automatically processed and still contain the “feel” a human ME add to a mix… I am sick to my stomach when people think this is ok, in any way form or shape. Cookie cutters are for cookies, cars and houses…

    For you guys on a tight budget: master your own artwork yourself, it will nevertheless showcase you intentions and serves as a very powerfull guide also for a professional ME if that chance one day arrives. It will show how you hear and feel music.
    There are many many articles that shows some general rules for mastering, making it possible for the novice person to make something sounding ok enough to share.

    I love the original article, thank you!

    • Reply
      Brian Hazard
      April 17, 2015 at 3:43 pm

      Thanks for the comment! Totally agree with the sentiment.

      That said, I wouldn’t trust myself to do my own artwork!

  • Reply
    Nathan
    May 14, 2015 at 12:06 pm

    I dont’ find this fascinating at all. It sounds like the same EQ curve on every track. I don’t feel there is any kind of “intelligence” behind this. I think this is simple a preset and an adaptive limiter threshold. It sounds like it’s on the radio basically. People that don’t know any better will think that’s “professional”.

    • Reply
      Nathan
      May 14, 2015 at 12:13 pm

      As an additional comment, they could even go as far as using audacity for this. It has a command line function to allow batch processing.

    • Reply
      Brian Hazard
      May 15, 2015 at 5:30 pm

      Well, it certainly is a “preset” in the broadest sense of the word, just hopefully a little more flexible than most. It could very well be similar to the type of processing an FM radio signal goes through before transmission, to maximize volume and therefore reach. In fact, if it manages to become better than most radio processing, that could provide a new market for the engine. Hmm…

  • Reply
    Robert Hedin
    May 22, 2015 at 4:24 pm

    Yeah, LANDR is pretty awful. Thanks for doing the comparisons!

  • Reply
    Shane D
    June 2, 2015 at 5:38 pm

    Guys, for real. Seriously.
    History repeating itself, over and over.
    Remember when ’email’ first appeared. All those letter writers scoffing…
    “Oh it’ll never replace a good hand written letter.”
    Well, I don’t know anyone who has hand written a letter and used snail mail since. I certainly haven’t. The introduction of the motor car? All those Blacksmiths saying…
    “It’ll never catch on. Where do you put the stirrups?”
    Mobile phones? No more phone booths. Phone booth manufacturers at the time were just like you guys. It’ll never replace us. We’re good.
    Let’s stop kidding ourselves, please. Change is inevitable.
    This is WITHOUT QUESTION the beginning of the end.
    Right now, the team behind this LandR thing (and a legion of competitors in the making) are reading every single comment that Pro Mastering Engineers like yourselves are making about the tech’s limitations on web pages right across the net, they’re taking down notes, and they’re all crunching out some vastly new and improved code, like any other software developer over time. This is merely the public beta test stage, for god’s sake. And thanks to all your technical comments as such, you are unknowingly contributing to putting yourselves out of a career – without fail.
    A self-defeating exercise, if ever there was one. Congratulations.
    Frankly speaking, the number of times I’ve had troubled studio clients come back to me with sad faces because the so-called mastering engineer totally wrecked their album, well I lost count years ago. I’ve had similar experiences with my own music, and eventually turned to doing it myself.
    It turns out that real musicians LOVE dynamics. Not volume. Funny that.
    At least for now, I see this LandR service as an affordable alternative. It’s not perfect, far from it, but it’s already delivering better results than what those ME’s did for my clients’ music. I’ve experimented with some of their raw mixes and was terribly shocked at how good it was compared to those ME’s charging $1500 per album. Or maybe it’s a case that there are no REAL mastering engineers to be found anywhere here in Australia. Who really knows?
    Bottom line…
    Give LandR a few version revisions, with lots of unintentional input from people like yourselves, and you can kiss mastering studios goodbye. I’m not trying to stir people up here. This is sad news to the extreme. I just can’t stand seeing talented people like yourselves in a desperate state of denial.
    IT’S COMING.
    BRACE YOURSELVES.

    • Reply
      Brian Hazard
      June 2, 2015 at 6:06 pm

      Thanks for sharing your perspective Shane!

      I’m with you mastering engineers wrecking albums. I’ve done plenty of salvage jobs and complete redos. It’s a shame.

      And yes, you are absolutely correct that these tools will continue to improve.

      Still, there are some things computers just aren’t very good at, like transcribing an orchestral performance. They can’t tell an oboe from a bassoon playing underneath the string section. They can’t determine your emotional state from the tone of your voice.

      To my way of thinking, mastering is a similar task. It’s not just carving out an “ideal” tonal balance, because it depends on the instrumentation, which LANDR can’t deduce. It can’t tell that the singer is singing intimately in a breathy voice, and that that track should be softer than the rock anthem.

      I’m not sure those are surmountable problems, but maybe if the engine requests more input from the user, those types of hurdles can be overcome.

      Sure, it’s coming. Someday!

      • Reply
        Derek
        September 1, 2015 at 4:45 am

        i have a question. in reference to the algorithm not understanding the difference between instrumments. I use a lot of different instruments, acoustic and electronic and i tweak each one individually in the mix etc then i send it off for mastering when all the levels sound good and the instruments all stand out on their own in the right places.
        Is your comment referring to a flat mix?
        I’m still not as experienced as i would like to be but i thought that if i fix those sorts of issues in my mix, the master shouldn’t make them worse?
        I spend ages on my tracks making sure the quiet claps are audible and small wooden instruments are audible by compressing them individuallly or boosting them, so essentially each single sound in my mixdowns has some sort of setting attached to it and then i send a lot of my smaller instruments through a main compressor to boost their loudness and have it sitting properly in the mix.
        I’ve just never properly understood how mastering works on a mixdown. I assumed that if you have all your levels sounding good in a mixdown then you will run into fewer problems.
        The examples above are of recorded live sessions it seems where all the cymbals and drums are coming through microphones into the mixing console or whatever gear is being ised so i feel like landr isnt the best for live stuff yet but if you are composing electronic music and your mixdown is sounding good, then it shouldnt be as bad?

        Forgive my ignorance. i am new to the intricacies of mastering.
        My problem here is that i have really good monitors.

        Adam A8X and i just mastered a song on landr (WAV uncompressed) and the master sounds amazing.
        But reading these comments is making me doubt my ears because im getting the impression that it shouldn’t sound as good as it does to me right now.
        What are your thoughts on this guys?

        Regards,
        Music lover and novice

        • Reply
          Brian Hazard
          September 2, 2015 at 9:16 am

          When all is said and done, you have to trust your ears. If you’re happy with the LANDR master, there you go! If your budget allows for professional mastering, give it a shot and see how they compare.

          Remember that in mastering, we only have access to the 2-track stereo mixdown. We can’t turn up the guitars in the second verse, for example. LANDR doesn’t analyze individual instruments. Instead, it adjusts the tonal balance of the entire mix, letting the chips fall where they may in regard to specific tracks.

          As a practical example, a mastering engineer might be careful not to cut too much in a specific frequency range if it lowers the vocal level too much. LANDR can’t make that distinction.

          If the tonal balance of the mix is solid to begin with, that will be less of an issue, because it will require less EQ in mastering.

      • Reply
        Jude
        June 3, 2016 at 5:22 am

        “Still, there are some things computers just aren’t very good at…” Unfortunately this is more desperate denial speaking. There’s NOTHING that computer’s aren’t “good” at. Computers are merely processing power, an energy resource which is inherently good or bad at nothing. We can harvest that power to do whatever we can conceive using simple or complex instructions, code, algorithms. The limit is a human / developmental one; as long as the processing power is there, the only thing preventing computers from being able to make these intelligible human analyses and decisions you speak of, is the current sophistication of the author’s algorithms. AI is a very real thing threatening to take over and even surpass MANY current “human-required” jobs; we have auto-pilot planes and self-driving cars on the way, where computers are being trusted with peoples lives.

        If you seriously think your job of making the right mastering decision to suit someone’s single is somehow unique and protected, especially looking at the current state of the music production industry compared to as little as 15 years ago, and the fact that something like landr already exists and is only being refined… then you are either in a deep denial, or simply saying what you think needs to be said publicly in order to keep the boat afloat as long as possible before you have to hit the button on a plan B. Delaying the inevitable. The robot option already wins on convenience, speed and price, and it’s only so long before its quality level reaches a point where the consumer is inclined to go with it and all its benefits in 90+ percent of the cases, in spite of what the mastering engineers of old are yelling on their blogs and youtube channels.

        I give it 5 years. 10 would be surprising.

      • Reply
        Jude
        June 3, 2016 at 5:39 am

        *consistency. another robot benefit I forgot to mention. As you both mentioned its a risk just finding an “engineer” who wont take your money and fuck your work up; if a prospective client wants to gamble with a few hundred dollars there are casinos for that. If they want a predictable result the current price is 9.99 for a month of unlimited tracks.

        Along with intelligence and quality improvements, as the technology matures I forsee an expanded amount of options and settings over just “high/low/medium”, really giving the consumer freedom of personal taste to boot. Its bleak for you bro.

        • Reply
          Brian Hazard
          June 6, 2016 at 8:24 pm

          Algorithms will create the music too! Google is already working on it. Will it be preferable to what a human can do? Sure, at some point. We only disagree about how far out that point is.

          You say 5-10 years. I’d say 10-20.

          But today, do you think LANDR is consistent? Try giving it more than one track and see if they match each other.

          Hit me back in 5 years and I’ll give you a warm soft pretzel if LANDR can out-master me on an album job.

          • Jude
            June 28, 2016 at 5:38 am

            By consistent what I mean is, its responding to characteristics in what it hears based on predefined rules. If you put the same file into it twice on the same setting, you’re gonna get the same exact result back both times. The response is mechanical and thus yes over time one can get to know how she thinks, and combined with the immediate turnaround time and cheap/unlimited amount if redos, not only can you come to predict what to expect, but you can begin to manipulate the result if desired by tweaking the input.

            To the pretzel: The issue there is we’re not waiting for the day it can out master you; by that point you better be far off into a different field. The breaking point will come much sooner, and that’s when that subjective difference of how much “better” you can out master it, is no longer large or evident enough to enough consumers for them to select you — and your much slower, much more expensive service — often enough for you to remain in business.

            That point where even your longterm clients who already trust your work are inclined to submit to the robot behind your back.

            That’s when you owe me a pretzel. With the way technology moves no way that’ll take a full decade, let alone 2.

            Now, algorithms creating actual music? That’s just silly. Me and Justin Beiber are pretty safe dude.

  • Reply
    Bruno Duarte
    June 15, 2015 at 12:46 pm

    I think is unfortunate people are going for this. 80% of music mastering is in the ears and 20% in the tools (can be arguable)…..hehe…Non the less, from my perspective Its imperative having actually someone taking care of the music and doing it the right way.

    At the end of the day, automated stuff, algorithms and whatever…, you will only get what you paid for.

    The thing I do like, is the business model, Subscription service = recurring clients and payments and let it all run on autopilot. Smart biz move…

  • Reply
    The Eclipse Project
    July 13, 2015 at 11:46 pm

    Well played Brian. Good article, and really it’s not biased or overly opinionated. Personally I do have the cheapest LandR account. Would I ever use it for a final release? Not a chance, under any circumstances. Nothing replaces a good engineer using proper gear. I too am a well trained studio engineer after all, with a lot of years under my belt. But does LandR have it’s place for what it is? For me, yes.
    As a person who does all his own recording, composing, field recordings for nature samples, bringing in featured artists and plays multiple instruments……I have a lot on my plate for finishing every song. By the time I’m done with a mix, I really don’t want to hear it for a while. LandR at least offers me the option to do a quick and dirty job for SoundCloud so I can send out prereleases of what I’m up to & working on. It also gives me a general map of things to be fixed in the mix also, usually small things but necessary. And that’s about it. It serves a small, but certainly not insignificant purpose. For this purpose, the cheapest 192k MP3 export account option is fine. Your not selling them after all. It’s an interim until the final product is ready.

    Could I do a better job than LandR on my own using something like the Ozone plugin in my own studio for these prereleases? Yes, most certainly. Do I want to, or have the time by the time it’s ready for that? Probably not! And there you have it.

    In the end, music is emotional. Therefore, something as important as the final master of a song should not be trusted to a computer algorithm.

    • Reply
      Brian Hazard
      July 15, 2015 at 4:46 pm

      Thanks for the kind words and nuanced perspective! I agree completely with your conclusion.

      When I master my own stuff, I only compress and limit. If something else is lacking, I go back to the mix. So I’d just as soon do it myself regardless.

      As for Ozone, half of the modules do more harm than good, except in extremely rare cases. In my tests, I wasn’t able to get the results I wanted from the compressor, and there are better limiters out there. Fwiw.

  • Reply
    Derek
    September 1, 2015 at 4:33 am

    😕 i just mastered a track and from what i can hear, it sounds really good. it may just be the specific mix or this specific track sounding good but an unlimited pro plan is costing me less than getting just one of my tracks mastered by a pro. I’ll have to listen to it at a later stage when i can turn up the volume but currently it sounds pretty nice.
    Budget is the biggest problem. if i had it my way i would send everything i make to a mastering engineer because they can tweak the songs individually and also feedback on the track is invaluable.
    I’ll see how it goes and then go from there

    • Reply
      Brian Hazard
      September 2, 2015 at 9:25 am

      Obviously the biggest benefit of LANDR is cost. Like you said, it’ll never tell you how to improve your mix, but if you don’t have the budget for anything else, it’ll have to do.

      You could always use it for works-in-progress, demos, and even singles if money is tight. For an EP or full-length where every track needs to not only sound its best, but match each other, professional mastering is essential.

      At some point I imagine LANDR will allow you to submit a batch of songs to master as a set. When that happens, I’ll definitely revisit!

  • Reply
    momalle3
    September 6, 2015 at 7:01 am

    I tried this–I sent it ten different tracks, all in different styles. My personal tastes are extremely hostile to mastering for loudness in general. Not surprisingly, I found the default (medium) setting and the high intensity setting unbearable and useless. I found the low intensity setting really good. Surprisingly good. It scooped the mids some; it tightened the bass more than I would have and it hyped the treble some. These were not changes that suited my taste, but they weren’t terrible or offensive and lots of people might prefer them.

    No it’s not as good as good mastering engineer but it’s damn good to my ears on the low intensity setting.

    Thinking about this as a new work process, I can imagine sending it mixes, listening to LANDR’s master, adjusting the mix, listening again, till you get it exactly right. That’s on the one had a dumb way to work–why not send it to a good ME and get it done right the first time? On the other hand, it’s really cheap and you don’t have to deal with some snobby mastering engineer. If the target audience is people who tend to do it all themselves, LANDR is a godsend. I could master a track as well as LANDR< I think, but mastering your own stuff is really hard because you get wedded to your mix choices and cant hear your way past them. LANDR would then sort of function as an objective listener with different tastes.

    Interesting times

    • Reply
      Brian Hazard
      September 9, 2015 at 10:00 am

      Sure, that’s a good way to use it. You could keep running works-in-progress through LANDR and try to match the tonal balance in your original mix. I typically use major label reference tracks, but I suppose it would be easier to match the tone of the same mix you’re working on.

      I actually tested out LANDR recently and found even the high intensity setting to be conservative volume-wise. I imagine the low intensity would tweak the tonal balance without noticeably affecting dynamics in any way.

      Not all mastering engineers are snobby, y’know! 😉

  • Reply
    Peter
    October 5, 2015 at 4:48 am

    For the 2nd track, I like Landr’s master better.
    Most importantly, it’s pretty incredible that software like this exists so I think about where it will be in 5-10 yrs. time.
    I’m sorry – but yes mastering engineers will be out of a job then, at least in 20 yrs.
    In 20 yrs., we won’t need to mix our music we write, it will just do it for us.
    Anyone who can’t see that (along with these other trends) is just an angry taxi driver after Uber came along. Only the Uber drivers will be pissed because self-driving cars will take their cumuffins.

    • Reply
      Brian Hazard
      October 6, 2015 at 4:03 pm

      It’s impressive, for sure, but not that incredible considering we’ve had spectral matching EQs for a long time. Seems like a natural evolution to me.

      While the gap between professional mastering engineer and software algorithm will continue to narrow, I don’t see how mixing could be replaced. That’s like saying a robot arm will paint our paintings for us – it’s an art, not a science.

      That said, more and more of it could be automated. Like perhaps it could be reduced to a simple set of sliders, or even a selection of “moods.” In other words, it will become more and more accessible to amateurs.

  • Reply
    ace
    November 10, 2015 at 8:58 pm

    When I first started to do records with my Logic 2.5 routing audio through my Mac 1/8″ stereo jack, people laughed. Same thing when we recorded on ADAT machines. They were right, that was far from a Studer. When I tried recording guitar right into the board, that was also very funny. It was just before the Pod. And at that time, expensive and pro mastering saved all those albums.

    But now, everybody and his sister is making a full album with plugins, from drums to guitar. You have a simulator for every possible instruments or piece of gear existing on the planet.

    And upon comments I read, a majority of people don’t see why they would pay for music.

    In conclusion: if you want comfort in life, choose another profession. (To late for me though). And of course, we can’t stop “progress”…

    • Reply
      Brian Hazard
      November 12, 2015 at 3:38 pm

      I’m with you 100%. I discourage people from going into music on a daily basis!

      • Reply
        ace
        November 12, 2015 at 3:40 pm

        🙂

  • Reply
    Joeri Jungschlager
    February 27, 2016 at 12:18 am

    I have had tried landr to do some simple jobs.. I would teach to a 5 year old. And it really messed it up, it main focus is on louder. I wonder if it even uses dittering!? Now to be honest I be better off putting a limiter and dittering on a postfader inserts Cubase on my master output I am pretty sure that does about the same as landr does if I need something quick and unfinished. Only difference is that it probably sounds a lot better.

    • Reply
      Brian Hazard
      March 1, 2016 at 7:42 pm

      Well, to be fair, it’s getting better. It certainly depends on the source material though. There’s definitely EQ, compression, limiting, and dither. But you’re right that for maximum transparency, simply putting a limiter on your master bus will do in a pinch!

  • Reply
    Grant Gadomski
    March 20, 2016 at 9:36 am

    Not sure if it should (ever) be used for anything professional, but as a music hobbyist with a 40+hr job I really like this service. I can get my mixdowns from shit to good (not great) in minutes without spending years learning how to properly master and weeks working on one track.

    • Reply
      Brian Hazard
      March 20, 2016 at 2:21 pm

      Grant, you’re exactly who LANDR is geared towards! When the choice is LANDR, self-mastering, or no mastering at all, LANDR is generally your best option. They aren’t looking to put mastering engineers out of business – rather, they’re evangelizing the benefits of mastering, and providing a point of entry.

  • Reply
    Jack Morrison
    June 13, 2016 at 3:41 pm

    Yeah I had tried this out about a year ago, and wasn’t too impressed frankly. However, gave it another shot recently when a friend brought it up to me again, and I must say, for at least getting a test listen of your works and progress it’s pretty useful. I don’t think they use the same algorithm for everything anymore, if that’s what they were actually doing in the first place, and maybe I will still go with a “true-blue human” for the final product but as of now, placebo effect or not, I definitely feel they’ve upped their game.

    • Reply
      Brian Hazard
      June 14, 2016 at 8:14 pm

      The algorithms are constantly changing, so you’re right there. And some tracks just happen to suit their algorithms better than others.

  • Reply
    Michael
    June 21, 2016 at 11:50 pm

    Just tried this out for the first time…was expecting a piece of $%& master to be honest, but am actually really impressed. Would I use for a release? Maybe…need to listen more to get a better feel…but it seems if I could get the tracks to match volume on a collection, it would be a viable option. Especially for outtakes, etc.

    I spent $150/track last year for a pro and successful ME and received very poor quality. A lot of it was due to the mix I now know, but for that $$ I would hope an ME would tell me to fix the mix and not just take my hard earned cash. The human element can care and adjust the master for its subtleties, make sure every track is compatible, etc., but it can also just need to pay rent and do @#$% work itself.

    Nice to know this exists…as for now I’m still more comfortable with a trustworthy and affordable ME so long as they exist. But I’m keeping an eye on it for sure.

    • Reply
      Brian Hazard
      June 23, 2016 at 8:24 pm

      Unfortunately at this point, you can’t get LANDR to match volume or tonal balance. It’s coming though!

      Sorry to hear you had a bad experience with a pro ME. You just have to find one you click with! LANDR will never be able to suggest mix tweaks, and those often impact the final product more than mastering ever could.

    Share Your Thoughts