The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
My LANDR Test Multi-Ef­fects Plugins
Old 25th August 2018
  #1
Gear Maniac
 
Magnus Lindberg's Avatar
 

My LANDR Test

I recently did a test of L*NDR for the podcast I am doing with a couple of collegues, thought it could be of interest for the folks here.

What gave me the idea to do this test was a question that I heard somewhere which is: what happens if you feed an already L*NDR mastered track back into L*NDR?

I did my test on a very good sounding hiphop track that I mastered a while back. These tests where done in the High setting in L*NDR, since my master is also loud I wanted to hear L*NDR do the same.

So I submitted the track, got the file back and started by comparing to my master. The L*NDR master was not a disaster, but it wasnt good either. It had roughly the same loudness as my master but the kick drum sub was surpressed, the punch was not really there the vocals didnt really pop etc. Typical overly processed with probably lots of multiband stuff going on.

But, then the question above. What happens if I feed this back into L*NDR?
A good mastering engineer obviously know when to do stuff to the audio and when not to do things, so how will L*NDR react?
The result that came back from L*NDR was quite ugly. While the loudness "only" increased by 0,2dB the actual EQ/multiband/whatever it does to my ears sounded like it had done the exact same thing one more time.

Then I did yet another pass with the 2nd L*NDR master. The file came back, and this time it was quite atrocious. Again as if it had done the same thing one more time, and also this time the loudess increased by 0,7dB.

Then, I decided to give it one last chance. So I took the original mix and added an 8dB shelf from 5k so that the mix was clearly way too bright. (the mix was already really good and balanced)
L*NDR gave me the file back and indeed it had not correced my hi-shelf, at least not in any significant way and when comparing this to the very first L*NDR master it confirmed this.

My take away from this:
- L*NDR diesnt really care in any significant way what your mix sounds like, it does roughly the same thing anyway. With a partial exception for loudness.
- It is less intelligent than I thought it would be.
- It cant replace a good mastering engineer by any stretch, not yet at least.
- If you are lucky and input a mix which suits the processing it does you can get a quite ok to quite good result soundwise, but nevertheless it is basically only a slightly more intelligent version of a master preset.

I did not test any of the other robot master platforms so this only applies to L*NDR, but I wouldn't be surprised if something similar would happen there.

Last edited by Magnus Lindberg; 26th August 2018 at 05:55 PM.. Reason: Thomas nitpicking.
Old 25th August 2018
  #2
Lives for gear
 
the unik's Avatar
Sounds like a robot was forced to deep thr...

Quote:
Originally Posted by Magnus Lindberg View Post
I recently did a test of LANDR for the podcast I am doing with a couple of collegues, thought it could be of interest for the folks here.

What gave me the idea to do this test was a question that I heard somewhere which is: what happens if you feed an already LANDR mastered track back into
LANDR?

I did my test on a very good sounding hiphop track that I mastered a while back. These tests where done in the High setting in LANDR, since my master is also loud I wanted to hear LANDR do the same.

So I submitted the track, got the file back and started by comparing to my master. The LANDR master was not a disaster, but it wasnt good either. It had roughly the same loudness as my master but the kick drum sub was surpressed, the punch was not really there the vocals didnt really pop etc. Typical overly processed with probably lots of multiband stuff going on.

But, then the question above. What happens if I feed this back into LANDR?
A good mastering engineer obviously know when to do stuff to the audio and when not to do things, so how will LANDR react?
The result that came back from LANDR was quite ugly. While the loudness "only" increased by 0,2dB the actual EQ/multiband/whatever it does to my ears sounded like it had done the exact same thing one more time.

Then I did yet another pass with the 2nd LANDR master. The file came back, and this time it was quite atrocious. Again as if it had done the same thing one more time, and also this time the loudess increased by 0,7dB.

Then, I decided to give it one last chance. So I took the original mix and added an 8dB shelf from 5k so that the mix was clearly way too bright. (the mix was already really good and balanced)
LANDR gave me the file back and indeed it had not correced my hi-shelf, at least not in any significant way and when comparing this to the very first LANDR master it confirmed this.

My take away from this:
- LANDR diesnt really care in any significant way what your mix sounds like, it does roughly the same thing anyway. With a partial exception for loudness.
- It is less intelligent than I thought it would be.
- It cant replace a good mastering engineer by any stretch, not yet at least.
- If you are lucky and input a mix which suits the processing it does you can get a quite ok to quite good result soundwise, but nevertheless it is basically only a slightly more intelligent version of a master preset.

I did not test any of the other robot master platforms so this only applies to LANDR, but I wouldn't be surprised if something similar would happen there.
Old 26th August 2018
  #3
Gear Head
 
moostapha's Avatar
 

Are you surprised? I'm surprised.

</sarcasm>
Old 26th August 2018
  #4
Quote:
Originally Posted by Magnus Lindberg View Post
I recently did a test of LANDR for the podcast I am doing with a couple of collegues, thought it could be of interest for the folks here.

What gave me the idea to do this test was a question that I heard somewhere which is: what happens if you feed an already LANDR mastered track back into
LANDR?

I did my test on a very good sounding hiphop track that I mastered a while back. These tests where done in the High setting in LANDR, since my master is also loud I wanted to hear LANDR do the same.

So I submitted the track, got the file back and started by comparing to my master. The LANDR master was not a disaster, but it wasnt good either. It had roughly the same loudness as my master but the kick drum sub was surpressed, the punch was not really there the vocals didnt really pop etc. Typical overly processed with probably lots of multiband stuff going on.

But, then the question above. What happens if I feed this back into LANDR?
A good mastering engineer obviously know when to do stuff to the audio and when not to do things, so how will LANDR react?
The result that came back from LANDR was quite ugly. While the loudness "only" increased by 0,2dB the actual EQ/multiband/whatever it does to my ears sounded like it had done the exact same thing one more time.

Then I did yet another pass with the 2nd LANDR master. The file came back, and this time it was quite atrocious. Again as if it had done the same thing one more time, and also this time the loudess increased by 0,7dB.

Then, I decided to give it one last chance. So I took the original mix and added an 8dB shelf from 5k so that the mix was clearly way too bright. (the mix was already really good and balanced)
LANDR gave me the file back and indeed it had not correced my hi-shelf, at least not in any significant way and when comparing this to the very first LANDR master it confirmed this.

My take away from this:
- LANDR diesnt really care in any significant way what your mix sounds like, it does roughly the same thing anyway. With a partial exception for loudness.
- It is less intelligent than I thought it would be.
- It cant replace a good mastering engineer by any stretch, not yet at least.
- If you are lucky and input a mix which suits the processing it does you can get a quite ok to quite good result soundwise, but nevertheless it is basically only a slightly more intelligent version of a master preset.

I did not test any of the other robot master platforms so this only applies to LANDR, but I wouldn't be surprised if something similar would happen there.
thanks for this Magnus. really interesting. not surprised at all actually. this further proves that real mastering is in fact custom work where these kind of services, like presets, are just that - presets that may or may not work. a crap shoot basically.
so if you are an artist, do you want the custom job on your mix on which you worked hard for months, or you just want a template?
Old 26th August 2018
  #5
Gear Addict
 

Thank for the effort to test Magnus.

I have learned that LANDR seems to be a popular option for people doing podcasts & indie radio shows.
Old 26th August 2018
  #6
Gear Maniac
 
Magnus Lindberg's Avatar
 

Quote:
Originally Posted by melopie View Post
Thank for the effort to test Magnus.

I have learned that LANDR seems to be a popular option for people doing podcasts & indie radio shows.
Ahh, yes probably quite useful for that sort of stuff.
Old 26th August 2018
  #7
Airwindows
 
chrisj's Avatar
I want to hear this result.

I could do a plugin called Flanders, which does the same 'preset thing' but is free
Old 26th August 2018
  #8
I also recently did a LANDR test. I've got three different mastered versions of the same track. The first one up was mastered by the engineer I've worked with for my recent releases. The second two masters are done with LANDR, one with the "high intensity" setting, the other with "medium intensity." (I've put those in quotes because GS otherwise thinks I'm linking to the Znaptiq ting!)

There's not much more to say - see what your ears think...

LW - LANDR vs Myles by Anton Barbeau | Free Listening on SoundCloud

Last edited by Grumblepig; 26th August 2018 at 12:13 PM.. Reason: Removed signature/extra links
Old 26th August 2018
  #9
Quote:
Originally Posted by Grumblepig View Post
I also recently did a LANDR test. I've got three different mastered versions of the same track. The first one up was mastered by the engineer I've worked with for my recent releases. The second two masters are done with LANDR, one with the "high intensity" setting, the other with "medium intensity." (I've put those in quotes because GS otherwise thinks I'm linking to the Znaptiq ting!)

There's not much more to say - see what your ears think...

LW - LANDR vs Myles by Anton Barbeau | Free Listening on SoundCloud
I'm not a fan of any of them, but the first file sounds the worst, wooly and muddy in the low mids with vocals separated from the music, I like the LANDR in high the most
Old 26th August 2018
  #10
My experience with LANDR is quite good, or I could say experience with human mastering engineers quite bad, it's subtle in low setting and because of that it usually wins with mastering engineers who like to "fix" things, or have worse day or whatever. I did around 5 shoot-outs with the bands I mixed (LANDR vs good mastering engineers), and 3 times out of 5 LANDR won. It won't fix any mixing mistakes, but to me it's not the role of the mastering so maybe that's why I like it.
Old 26th August 2018
  #11
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
i refer to it as BLUNDER or BLUNDR.

probably best that we quit mentioning it, giving it free PR advertising.

hopefully then it will die in obscurity, penniless and insane.

jt
Old 26th August 2018
  #12
Lives for gear
 
Ben F's Avatar
 

Verified Member
Quote:
Originally Posted by Jerry Tubb View Post
i refer to it as BLUNDER or BLUNDR.

probably best that we quit mentioning it, giving it free PR advertising.

hopefully then it will die in obscurity, penniless and insane.

jt
It can only really improve. I personally think the Ozone 8 mastering assistant is a better tool.
Old 26th August 2018
  #13
Quote:
Originally Posted by Jerry Tubb View Post
i refer to it as BLUNDER or BLUNDR.

probably best that we quit mentioning it, giving it free PR advertising.

hopefully then it will die in obscurity, penniless and insane.

jt
Well, who knows how long it'll be around? I've been curious to try it for a couple years. I used it on a full album - a synthpop collection released "secretly" to my Patreon supporters. The whole thing was an experiment. For that collection of tunes, the robot mastering service was fine - or fine enough. It simply squashed everything together in a cohesive way, one track felt connected to the next and to the previous. That was all I needed. I'd not use it again for anything "proper," though. I've been fortunate to work with wonderful mastering engineers and I'll continue to do so.
Old 26th August 2018
  #14
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by Magnus Lindberg View Post
I recently did a test of L for the podcast I am doing with a couple of collegues, thought it could be of interest for the folks here.

What gave me the idea to do this test was a question that I heard somewhere which is: what happens if you feed an already L mastered track back into
L?

I did my test on a very good sounding hip-hop track that I mastered a while back. These tests where done in the High setting in L, since my master is also loud I wanted to hear L do the same.

So I submitted the track, got the file back and started by comparing to my master. The master was not a disaster, but it wasnt good either. It had roughly the same loudness as my master but the kick drum sub was surpressed, the punch was not really there the vocals didnt really pop etc. Typical overly processed with probably lots of multiband stuff going on.

But, then the question above. What happens if I feed this back into L?
A good mastering engineer obviously know when to do stuff to the audio and when not to do things, so how will L react?
The result that came back from L was quite ugly. While the loudness "only" increased by 0,2dB the actual EQ/multiband/whatever it does to my ears sounded like it had done the exact same thing one more time.

Then I did yet another pass with the 2nd L master. The file came back, and this time it was quite atrocious. Again as if it had done the same thing one more time, and also this time the loudess increased by 0,7dB.

Then, I decided to give it one last chance. So I took the original mix and added an 8dB shelf from 5k so that the mix was clearly way too bright. (the mix was already really good and balanced)

L gave me the file back and indeed it had not correced my hi-shelf, at least not in any significant way and when comparing this to the very first L master it confirmed this.

My take away from this:
- L diesnt really care in any significant way what your mix sounds like, it does roughly the same thing anyway. With a partial exception for loudness.
- It is less intelligent than I thought it would be.
- It cant replace a good mastering engineer by any stretch, not yet at least.
- If you are lucky and input a mix which suits the processing it does you can get a quite ok to quite good result soundwise, but nevertheless it is basically only a slightly more intelligent version of a master preset.

I did not test any of the other robot master platforms so this only applies to L, but I wouldn't be surprised if something similar would happen there.
Note "L" = L#NDR

I notice that every so many weeks/months some one puts up a post about L#ndr on this list and mentions the name a lot of times in their post. (14 in this one alone), I am never sure if this person is a shill for L#ndr or someone just wants to test it. It seems strange that it happens so frequently. No offense meant but why would anyone mention the same name 14 times in one post? it is almost like they are trying to get more people to go to the L#NDR site. FWIW

Last edited by Thomas W. Bethe; 26th August 2018 at 05:19 PM.. Reason: Got rid of product ID
Old 26th August 2018
  #15
Gear Maniac
 
Magnus Lindberg's Avatar
 

Quote:
Originally Posted by Thomas W. Bethe View Post
I notice that every so many weeks/months some one puts up a post about L#ndr on this list and mentions the name a lot of times in their post. (14 in this one alone), I am never sure if this person is a shill for L#ndr or someone just wants to test it. It seems strange that it happens so frequently. No offense meant but why would anyone mention the same name 14 times in one post? it is almost like they are trying to get more people to go to the L#ndr site. FWIW
You seriously think that I am secretly doing an ad for "the site" by sharing a test in which I draw the conclusion that it is not so much more than a mastering preset?

On topic, I think it can have some good uses while at the same not being a replacement for a good ME.
Old 26th August 2018
  #16
Lives for gear
Quote:
Originally Posted by Thomas W. Bethe View Post
why would anyone mention the same name 14 times in one post?
it's 28 now that you quoted the entire post.
Old 26th August 2018
  #17
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by Magnus Lindberg View Post
You seriously think that I am secretly doing an ad for "the site" by sharing a test in which I draw the conclusion that it is not so much more than a mastering preset?

On topic, I think it can have some good uses while at the same not being a replacement for a good ME.
No, but why mention the same name 14 times??? It almost looks like you are shilling for them. You can use their name just use L#NDR or something else that the BOTS won't pick up on. FWIW.

You do realize that every time you mention a name that the internet BOTS pick up that name and it increases the companies internet presence? I would rather see YOUR studio name repeated 14 times than L#NDR's name.
Old 26th August 2018
  #18
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by scraggs View Post
it's 28 now that you quoted the entire post.
Taken care of. Thanks for your keen observation. Back to 14
Old 26th August 2018
  #19
Gear Maniac
 
Magnus Lindberg's Avatar
 

Quote:
Originally Posted by Thomas W. Bethe View Post
Taken care of. Thanks for your keen observation. Back to 14
Happy now?=)
Old 26th August 2018
  #20
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
Quote:
Originally Posted by Ben F View Post
It can only really improve. I personally think the Ozone 8 mastering assistant is a better tool.
I don't use or endorse any Oz*ne products either!

Best, JT
Attached Thumbnails
My LANDR Test-automstr.jpg   My LANDR Test-magnus_robot.jpg  
Old 26th August 2018
  #21
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by Magnus Lindberg View Post
Happy now?=)
Not happy or unhappy but why would anyone who is a professional mastering engineer like you "advertise" a website who maybe taking a lot of business away from people like you??? To me it just does not make a lot of sense.

People today do NOT understand what a real mastering engineer can do for their mixes BUT they know they have to get them "mastered" so why NOT send them to a place were they are charged very little or nothing. After all "ALL MASTERING" is the same, at least to their way of thinking. It is not until they go to someone like you when they can experience what a real mastering engineer can do.

FWIW
Old 26th August 2018
  #22
Lives for gear
 
Ben F's Avatar
 

Verified Member
Quote:
Originally Posted by Jerry Tubb View Post
I don't use or endorse any Oz*ne products either!

Best, JT
You’re missing out on one of the best limiters in the business Jerry!
Old 27th August 2018
  #23
Lives for gear
 
SmoothTone's Avatar
 

LANDRLANDRLANDRLANDRLANDRLANDRLANDRLANDR

Sorry, I don't know what came over me

Last edited by SmoothTone; 27th August 2018 at 12:38 AM.. Reason: I think LANDR is controlling my mind.
Old 27th August 2018
  #24
Lives for gear
 
robert82's Avatar
Thomas,
I get that business is tough. It's just that you're starting to sound paranoid.
Old 27th August 2018
  #25
Lives for gear
 
Jerry Tubb's Avatar
 

Verified Member
Quote:
Originally Posted by Ben F View Post
You’re missing out on one of the best limiters in the business Jerry!
Ben, i’ve been pretty happy with the PSP Xenon or Fab Filter Pro-L!

Best, JT
Old 27th August 2018
  #26
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by robert82 View Post
Thomas,
I get that business is tough. It's just that you're starting to sound paranoid.
Sorry! lots of things going on and none of them very good but they seem to be getting better all the time.

No doom and gloom but I get tired of people unwittingly promoting an online AI mastering website which could be taking a lot of work from people on this web board. I had a client call up and needed something done right away. Good client and I was happy to help EXCEPT that he said "if you cannot get this done I may use L#NDR instead" I guess that was an implied threat. Anyway got it done for him and he was pleased. FWIW
Old 27th August 2018
  #27
Lives for gear
 
Thomas W. Bethe's Avatar
 

Verified Member
Quote:
Originally Posted by Jerry Tubb View Post
Ben, i’ve been pretty happy with the PSP Xenon or Fab Filter Pro-L!

Best, JT
Both GREAT but the limiter in Ozone is also GREAT. You can down load a demo if you want from the Izotope site. FWIW
Old 27th August 2018
  #28
Gear Head
 
moostapha's Avatar
 

Quote:
Originally Posted by Thomas W. Bethe View Post
Sorry! lots of things going on and none of them very good but they seem to be getting better all the time.

No doom and gloom but I get tired of people unwittingly promoting an online AI mastering website which could be taking a lot of work from people on this web board. I had a client call up and needed something done right away. Good client and I was happy to help EXCEPT that he said "if you cannot get this done I may use L#NDR instead" I guess that was an implied threat. Anyway got it done for him and he was pleased. FWIW
If they can't hear the difference between L&R and an actual human, are they really going to be repeat customers?

There's a part of me that thinks the way L&R does things is actually kind of genius. It can't do a good job, but it does make a difference. And the mid-scooped sound I've heard from it would improve a lot of the bedroom mixes I've worked on...just not as much as pretty much any human who had any idea what they were doing.
Old 27th August 2018
  #29
Gear Guru
 
Brent Hahn's Avatar
 

Quote:
Originally Posted by melopie View Post
Thank for the effort to test Magnus.

I have learned that LANDR seems to be a popular option for people doing podcasts & indie radio shows.
If you're doing podcasts on the Blue-Yeti-in-the-Kitchen level, LANDR makes perfect sense.
Old 27th August 2018
  #30
Quote:
Originally Posted by Thomas W. Bethe View Post
No doom and gloom but I get tired of people unwittingly promoting an online AI mastering website which could be taking a lot of work from people on this web board.

Maybe not from this board, but there are far more humans involved than you seem to expect. The demand for humans is clearly growing.

The first offer here for example looks like an excellent position for an ME looking to reduce income volatility: Jobs - landrcom

The fourth also look pretty cool IMHO. If you can't beat the enemy, join him.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump