My new Post-Rock song Mastered, Suggestion needed!

Svemir

Member
Hi there!

I need an opinion from someone who's quite a pro in mastering.

I have recorded and mixed this song with Cakewalk and used LANDR (AI Automatic Mastering tool) to master it.

For who doesn't know LANDR, you upload your song and AI does the mastering instantly for you, you can upload a reference track also it will apply the same mastering sound.

I have two different types of masters here, the first is the Landr standard one, the second I used a reference track from Mogwai album to make it sound similar.

What's the best in your opinion?

Any other suggestions are welcomes especially about dynamics. Thanks!



https://drive.google.com/file/d/1-Td2uUkeB5Ag3Q2C-IAFkoF0AdbaEl3I/view?usp=sharing

https://drive.google.com/file/d/1lpJwg_Nd4z3ZWopLYt5ZyiRW8BAPtGY4/view?usp=sharing
 

Svemir

Member
PS: In my opinion the first one is wider, in the sense that the instruments are more panned left and right, you can really hear the difference with headphones when you switch to the second song everything is more centered.

I have heard a lot of times that 100% panning LCR is the most used and suggested by mixing engineers, to make it sound wide,

at the same time I'm not sure, it seems to me that the second one sounds much better, the first one is more clear and you can better define each instruments because they sound more separate of course, everything is more panned, but still the second one is more powerful and mashed together, doesn't sound like is all separate.

Also consider this is an instrumental song, with drums, bass, and sometimes from 2 to 6 guitar layers.
I'm not convinced about using a 100% panned guitar left and one right if I have only 2 guitars playing cause it seems that something in the center is missing, and sounds wider but less powerful.

I'm not sure which is the best way to go, the instinct and my ears tell me the 2nd track (referenced one) sound better. Your opinion?
 
Last edited:

Massive Master

www.massivemastering.com
[1] It's really difficult to judge this sort of thing without hearing the original track.
[2] They both have really, really severe phase issues in the low end -- This is why it's difficult to judge without hearing the original track - It's hard to get past that without knowing if it was in the original mix or if they're just running it through Ozone or something)
[3] Okay, I'm in the louder part near the end and both are ridiculously distorted. Again, it'd be handy to know if the mix was like that or not.

The only thing I can tell you off the bat is that "Reference-High" is much more "damaged" than "Balanced-High" but the phase issues are much worse in "Balanced" --

I'm assuming this isn't what you wanted to hear. Still, I'd love to hear the mix to see where everything occurred.

[EDIT] As far as panning goes, it's pretty rare that anything in a mix is panned 100% off other than a stereo pair of room mics - maybe the drum overheads (although even that can induce a sense of the entire recording being "inside the drums"), possibly two guitars playing the same or similar parts (but again, all the way can put them in an unnatural sounding field). I'd suggest [1] Starting the mix in mono. Get things to play well with each other spectrally while they're in the same place spatially. [2] Pan them out as you would see them standing in front of them on stage. If the guitars on the stage in your head aren't all the way off to your side and pointed at one ear, don't pan them like that - at least at first. Stereo instruments - meh, maybe all the way, depending on the instrument. Synth pads and the like. But even that may sound better with one channel all the way over to the left (for instance) and the other channel only partially to the right.

Piano is another one -- It's only 6' wide in real life -- and *on its own* panning two mics out for a solo piece might sound perfectly fine. But as an instrument in a stereo field with other instruments, it'll sound like the entire band is "inside" the piano - especially if the piano is up front in the mix. THAT SAID -- You'd probably be using some distance in there also - solo instrument or not - so you can adjust the {we're finally getting to the word of the day here) DEPTH of the instrument in the field - which is just as important when trying to fit it in with the other instruments.

Blah, blah, no rules and what not. And yes, I'm aware that Eddie's guitar was typically panned all the way to one side. But the giganto-verb that was filling in the other side is what made it sound like he was in the room with you. Solo that left channel (Driver's side - I used to call it "Pan-Halen when we'd pan guitars all the way to the side) and you have a guitar on the side of your head pointed directly at your ear.
 
Last edited:

Svemir

Member
[1] It's really difficult to judge this sort of thing without hearing the original track.
[2] They both have really, really severe phase issues in the low end -- This is why it's difficult to judge without hearing the original track - It's hard to get past that without knowing if it was in the original mix or if they're just running it through Ozone or something)
[3] Okay, I'm in the louder part near the end and both are ridiculously distorted. Again, it'd be handy to know if the mix was like that or not.

The only thing I can tell you off the bat is that "Reference-High" is much more "damaged" than "Balanced-High" but the phase issues are much worse in "Balanced" --

I'm assuming this isn't what you wanted to hear. Still, I'd love to hear the mix to see where everything occurred.
Ok, I get it you mean when the heavy drums part starts?I’m struggling to understand whether you are speaking about an audio / mixing problem or a simple matter of sound taste.
Notice that the distortion I applied in that part on the rithm heavy guitar is a big muff pedal, which makes a very fuzzy distortion. Used mainly in grunge, shoegaze bands. So that is intentional, if you mean that.

https://drive.google.com/file/d/1UZML-VlzERq9juPUAkf__sxQ-A3Rx646/view?usp=sharing
 

Massive Master

www.massivemastering.com
Okay - The distortion on on the AI guys. The Haas filter is on you -- You have your drums (all of them, from what I can tell) going through some sort of Haas / delay that's wrecking the phase, especially on the low end. It's something you can get away with on overheads / room mics to some extent (was big in the 80's metal stuff, where you'd delay one of the room mics by 5 or 10ms, or as an auxiliary send that was high-passed in an otherwise unnatural fashion, so as not to mess with the phase in the low end).

The reason I was asking about the original mix was that if the AI's are using something like Ozone, it's one of the few things I know of that actually makes it very easy to throw the phase and time (via the Haas effect) of the low end out of whack. This is one of those "It's your mix and you can do what you like" things - but I'd highly recommend panning the kit however seems natural and if you want to add a slight delay - likely in the form of a short, dense reverb or slap that isn't actually identical to the source audio - you'll wind up with a much more balanced and natural (for lack of a better term) result.

Instead of having "width" you're winding up with *all width* - anchored to nothing. There's *nothing* in the center. Everything is just passing through on the way to the other side.
 

Svemir

Member
Okay - The distortion on on the AI guys. The Haas filter is on you -- You have your drums (all of them, from what I can tell) going through some sort of Haas / delay that's wrecking the phase, especially on the low end. It's something you can get away with on overheads / room mics to some extent (was big in the 80's metal stuff, where you'd delay one of the room mics by 5 or 10ms, or as an auxiliary send that was high-passed in an otherwise unnatural fashion, so as not to mess with the phase in the low end).

The reason I was asking about the original mix was that if the AI's are using something like Ozone, it's one of the few things I know of that actually makes it very easy to throw the phase and time (via the Haas effect) of the low end out of whack. This is one of those "It's your mix and you can do what you like" things - but I'd highly recommend panning the kit however seems natural and if you want to add a slight delay - likely in the form of a short, dense reverb or slap that isn't actually identical to the source audio - you'll wind up with a much more balanced and natural (for lack of a better term) result.

Instead of having "width" you're winding up with *all width* - anchored to nothing. There's *nothing* in the center. Everything is just passing through on the way to the other side.
Bloody hell,

so if I understand correctly, pardon my english, is that there is a HAAS (no idea what that is) filter on the drums, so it's a delay basically.
Honestly I'm not a drummer and not a mixing engeneer, I just made these MIDI programmed drums with a plugin.
I have sent the track also to a friend of mine who runs a recording studio and he's a drummer, to ask for an opinion, he said it sounds great, but I understand friends might not be so critc.
So would you please confirm what you just said by listening only to the drums track here attached? so you can hear it by itself and tell me about this delay, which honestly I don't hear, I increased the room and distant room mics to make it less dry thats true, also cymbals volume seems fine to me, but again, I'm not a drummer.
Here's the track

 

Massive Master

www.massivemastering.com
The Haas effect is sort of a time-based stereo exaggeration. Where you can take a mono source and make it sound like a stereo source by delaying (and usually messing with spectrally) one side. You can use it on a stereo source (I mentioned earlier delaying one room mic to mimic a much wider room) to exaggerate the apparent width of that source (you will be messing with the phase relationship of course, there's no free lunch with "creating" width unless that width was actually there to begin with).

Those drums are fine. In the mix, it sounds like there's a copy on the left and then a copy on the right, split apart by a few milliseconds. If you're not putting a delay on them, something is. It sounds like it's on nearly everything in the whole mix.

This (drums only) has a nice, centered core and a relatively wide and sparse stereo image. The mix has no core - just a hard left and a hard right.

[A MOMENT PASSES] Loaded it in - There's basically no difference (in the mix) at all between the sum and the difference. I'm not sure I could adjust a delay on one side and completely cancel the other out, but listening to the mid and side info, it sounds like it would be possible.

This drum track though -- Solo the sum and it's all there as you would expect (just with no stereo spread). Solo the difference and there's all the stereo information, but with basically no kick, snare, etc. Again, just what you'd expect.

And (at the risk of sounding like a broken record) is why I was asking about *when* (where) this was happening. The stereo drum track on its own is fine. Something is happening after that. And the reason I asked about Ozone is that it actually has presets that do this sort of (I still call it "damage") on purpose. It sounds like the entire mix is going through a Haas filter turned up to 100% (all delay, no center). If you're using a time-based stereo widener (well, I don't think there's an "if" in there - There *IS* a time-based stereo widener in here somewhere), it's one of those things you can sort of get away with as an auxiliary send feeding a dense reverb, or maybe on certain elements (again, drum overheads or room mics), but this mix sounds like literally everything is in there. Including the bass. The amount of bass in the sum is equal to the amount of bass in the difference. It's 90 degrees out of phase with itself. 99 times out of 100, it should be *0* degrees out (dead center) and if the bass is heard in the difference, it's usually a stereo chorus effect return or something along those lines that's added to the signal path (as opposed to inserted *into* the signal path).

And again -- If this is what you're shooting for, it's your mix and there you go. But for me, it's twisting my ears rather uncomfortably.

Ooh - Looks like I can add images in line - Visually speaking, here's the drums. Nice core, nice width, collapses to mono very well, very natural sounding. If these are MIDI drums, I'm actually impressed with the realism of the stereo image. And this is pretty much representative of a typical mix. Okay, maybe a little more energy off to the sides with a wall of guitars, and certainly more information along the stereo spread, but there you go.

POLAR - Drums.jpg

THIS is the MIX ---

POLAR - Mix.jpg

So that scares me right off the bat. Essentially three nearly equal lobes. And I'd almost bet money that the info in the center is transient (as in, "moving" not transient like the whack of a snare drum) -- The signals colliding with each other from left to right. Print this to vinyl and you may have a stylus literally just bouncing up and down out of the groove.
 
Last edited:

Svemir

Member
The Haas effect is sort of a time-based stereo exaggeration. Where you can take a mono source and make it sound like a stereo source by delaying (and usually messing with spectrally) one side. You can use it on a stereo source (I mentioned earlier delaying one room mic to mimic a much wider room) to exaggerate the apparent width of that source (you will be messing with the phase relationship of course, there's no free lunch with "creating" width unless that width was actually there to begin with).

Those drums are fine. In the mix, it sounds like there's a copy on the left and then a copy on the right, split apart by a few milliseconds. If you're not putting a delay on them, something is. It sounds like it's on nearly everything in the whole mix.

This (drums only) has a nice, centered core and a relatively wide and sparse stereo image. The mix has no core - just a hard left and a hard right.

[A MOMENT PASSES] Loaded it in - There's basically no difference (in the mix) at all between the sum and the difference. I'm not sure I could adjust a delay on one side and completely cancel the other out, but listening to the mid and side info, it sounds like it would be possible.

This drum track though -- Solo the sum and it's all there as you would expect (just with no stereo spread). Solo the difference and there's all the stereo information, but with basically no kick, snare, etc. Again, just what you'd expect.

And (at the risk of sounding like a broken record) is why I was asking about *when* (where) this was happening. The stereo drum track on its own is fine. Something is happening after that. And the reason I asked about Ozone is that it actually has presets that do this sort of (I still call it "damage") on purpose. It sounds like the entire mix is going through a Haas filter turned up to 100% (all delay, no center). If you're using a time-based stereo widener (well, I don't think there's an "if" in there - There *IS* a time-based stereo widener in here somewhere), it's one of those things you can sort of get away with as an auxiliary send feeding a dense reverb, or maybe on certain elements (again, drum overheads or room mics), but this mix sounds like literally everything is in there. Including the bass. The amount of bass in the sum is equal to the amount of bass in the difference. It's 90 degrees out of phase with itself. 99 times out of 100, it should be *0* degrees out (dead center) and if the bass is heard in the difference, it's usually a stereo chorus effect return or something along those lines that's added to the signal path (as opposed to inserted *into* the signal path).

And again -- If this is what you're shooting for, it's your mix and there you go. But for me, it's twisting my ears rather uncomfortably.

Ooh - Looks like I can add images in line - Visually speaking, here's the drums. Nice core, nice width, collapses to mono very well, very natural sounding. If these are MIDI drums, I'm actually impressed with the realism of the stereo image. And this is pretty much representative of a typical mix. Okay, maybe a little more energy off to the sides with a wall of guitars, and certainly more information along the stereo spread, but there you go.

View attachment 108855

THIS is the MIX ---

View attachment 108858

So that scares me right off the bat. Essentially three nearly equal lobes. And I'd almost bet money that the info in the center is transient (as in, "moving" not transient like the whack of a snare drum) -- The signals colliding with each other from left to right. Print this to vinyl and you may have a stylus literally just bouncing up and down out of the groove.
OMG, Thanks for your detailed explanation, very kind.
These are the drums, indeed doesn't seem out of phase to me, on the track there are no effects just freeG meter, also on the drums bus and masterbus there are no effects or EQ or any limiter, nothing.
I don't understand how the drum track can sound out of phase with the whole mix, and it's not just by itself. Especially when I didn't put any effect or reverb on the bus and master bus? How can it be? Maybe a bug? Some settings like latency or something like that?

For the mastered version I didn't use OZONE but I used LANDR (AI mastering platform), but in any case you said the phase issue is on the mixed version as well?
First step I suppose is to exclude the master caused this phase issue just by listening to the non-mastered version, I cannot just hear it by myself honestly.
NON MASTERED VERSION: https://drive.google.com/file/d/1UZML-VlzERq9juPUAkf__sxQ-A3Rx646/view?usp=sharing

Regarding the bass, please consider is a temporary bass, I have to record the real bass still, so it might not be very aligned with the drum kicks, can it be that causing the phase issue?

Regarding the second image you posted about the mix, it's not very clear to me what should I do to fix this, signals colliding? We are speaking about a different thing here right? not about the phase issue.

1616058080998.png
 
Last edited:

Svemir

Member
Regarding the phase issue, I frankly think that if the master is excluded, and the drum truck by itself is not out of phase, then must probably be the bass which is out of phase with the drums? Because when I recorded the bass I didn't pay that much attention to the timing since it's just temporary, needs to be recorded again.
 

Massive Master

www.massivemastering.com
I'm saying that basically the entire mix, everything in it, is around 90 degrees out of phase with itself in relation to the other side. The sum literally sounds the same as the difference. The drums, the bass, the guitars, everything. There may be typical stereo information in there -- but if there is, it's being drowned out by the (whatever is causing the time delay from one side to the other).

The solo'd stereo drum track sounds absolutely and completely different than the drum track sitting in the mix. The track on it's own has a (surprisingly realistic for a MIDI based source) "typical" stereo spread with a typical anchor in the center. In the mix, there is NO center, there is NO anchor. Only a collision in the center during the short periods of time where a *part* of the signal exists on both sides simultaneously.

Understood on the LANDR / Ozone stuff -- I only asked originally as I've heard some rather "Ozone" sounding stuff from a few of these "AI" places - and if someone presented my with a normal sounding mix and said "Can you make it sound like [this mix we're talking about]?" I'd probably grab for something like Ozone. But again, whatever is going on is in the unmastered mix.

Side-noting - this is one of the beefs I have with a lot of these "AI" things. The first thing I'd do if a mix like that came in would be to call you and ask if it's like that on purpose or what not. It might be - I've had it happen. But the decisions that I'm going to make on a mix like that will be wholly and completely different than on a mix with a typical stereo image.

The long and the short is, somewhere, there is something affecting the stereo spread of (what sounds like) the entire mix, in a way that's affecting the time relationship from one side to the other (Haas Effect) as opposed to the relationship in volume between the sum and difference. Either is going to cause changes to the phase relationship to some extent, but one turns down the sum and turns up the difference while the other essentially removes the middle information entirely.
 

Svemir

Member
Ok thanks a lot I got it, so basically the issue is in the mix, the whole instruments together are out of phase of 90 degrees in the sense that everything coming out of the right headphone for instance is coming some milliseconds later than the sound on the left headphone.
Is that correct?
It's very strange though, I'm afraid must be some of my DAW settings, I'm using Cakewalk, since I have no delays on the busses or master, must be some DAW option or export track option, I will post this in the Cakewalk forum.
Thanks a lot again.
 

Svemir

Member
what about this?
I’m not sure I understood but this is worrying me since I would like to make some vinyls.
Whats the name of this plugin you used?
7109CB72-5173-4420-9AF4-620FA4A5162E.png
 
Last edited:

Svemir

Member
That's basically it. Might be off by degrees or milliseconds, but that's what it sounds like to me.
Aboot the previous post, I just followed what this guy is saying, not sure whether you agree but he’s a pro mixing engineer.
So to make it wide you just use RCL rule panning, is what I did. 2 guitars, one 100% right other 100% left, with 3 I add one in the center. The rest bass and drums are all in the center

 

Tadpui

Well-known member
The drums are distractingly wide in the non-mastered version as well, so it's either baked into the samples or there has been some sort of stereo widening somewhere along the way. It sounds like the drums are panned mostly right, and just a slapback delay of the drums is on the left. The phase does sound a little weird, but I'm not as sensitive to it as MM's trained ears are.

By the way, the cymbals on those drum samples sound terrific. The smooth bloom to them is just really pleasant. I'm not sure if that's baked into the samples, or if it's the choice of drum bus compression but they sound really good.
 

Svemir

Member
The drums are distractingly wide in the non-mastered version as well, so it's either baked into the samples or there has been some sort of stereo widening somewhere along the way. It sounds like the drums are panned mostly right, and just a slapback delay of the drums is on the left. The phase does sound a little weird, but I'm not as sensitive to it as MM's trained ears are.

By the way, the cymbals on those drum samples sound terrific. The smooth bloom to them is just really pleasant. I'm not sure if that's baked into the samples, or if it's the choice of drum bus compression but they sound really good.
Thanks for the suggestion but Have you seen the video!? He says 2 guitars must go 100% left and right respectively.
about the drums, how can that sound too much panned if all kick, snare, cymbals are pan dead center? Drums are MIDI programmed drums! I can screenshot it for you and show you the hobs, The drums are stereo, just toms and hi hat are panned a bit L/R.

1616142236215.png

 
Last edited:

Svemir

Member
ANd I finally confirm what the mixing engineer says about panning just by listening to one of my favourite albums.
If you play with the headphones removing one on the right hear then on the left, you can hear the guitars are 100% panned L/R indeed.
And I love the sound, this proves that this is probably due to personal taste and not a mixing mistake, as this way of recording is recommended by lots of mixing engineers out there.

This song: https://envy.bandcamp.com/track/light-and-solitude
 

Svemir

Member

DM60

Well-known member
Thanks for the suggestion but Have you seen the video!? He says 2 guitars must go 100% left and right respectively.
about the drums, how can that sound too much panned if all kick, snare, cymbals are pan dead center? Drums are MIDI programmed drums! I can screenshot it for you and show you the hobs, The drums are stereo, just toms and hi hat are panned a bit L/R.

View attachment 108897

I think if you look at the OH and room Width, you will see that the knobs are full right. In the close mics, you have you panning. I wonder in your software if the panning on the close mics are fighting the Kit mics? Just looking at the close mic panning, everything looks pretty good, nothing hard L/R. Just for fun, try bring your room and OH width closer to the middle and see if that doesn't sort it out.
 
Top