Advertisement

feedback/advise on final mix and master of a game soundtrack

Started by April 25, 2015 04:52 PM
4 comments, last by Breakdown Epiphanies 9 years, 6 months ago

Hi everybody,

I have been working on the soundtrack for a game called Battle Brothers and the release on Steam EA is drawing near. We already got more than an hour of music recorded but I feel that the material still needs some polish.

You can find many of the tracks in this soundcloud play list:

https://soundcloud.com/breakdownepiphanies/sets/battle-brothers-ost

The game is a turnbased tactical/rpg hybrid taking its inspiration from older titles like the original X-Com, Jagged Alliance or Warhammer - Shadow of the Horned Rat. The setting is low fantasy/medieval and you captain a mercenary company.

I would be grateful if some of you took the time and listen to some of the pieces to give some feedback and criticism. I would be mostly interested in suggestions regarding the overall sound/mixing/mastering. This not only is my first soundtrack release but the first time I worked with orchestral arrangements so there must be tons of stuff that I don't understand yet. Of course we can talk about the compositions and arrangements too but I am not sure how much I will be able to change in that regard as the project is close to its release.

What needs to be done:

Remastering all the tracks and try to achieve even levels and a cohesive sound. Revisiting some of the mixes, especially the ones that were created early in the production process.

What has been done:

- Regarding libraries and instruments the orchestral stuff is mostly Albion 1-3 from Spitfire, then some instruments (mostly percussion) from the Kontakt library and Komplete 9. There are a few "booms" and transitions from Hybridtwo's Project Alpha in there, but not that much as we did not want to go down the "hybrid" route but aimed for a warmer orchestra sound. Some of the tracks feature acoustic guitar parts that we recorded in the living room "studio" ;) (2 channels into a Focusrite Liquid Saffire 56 interface, one signal directly from the pickup and one recorded with a Shure KSM42 Mic).

Mixing:

- Up until now the tracks are mostly organized with an orchestra seating in mind. We did not use a lot of processing on the instruments. Strings are just eq-ed, brass have some compression on top of that and drums/percussion are often treated with with compression too + they. We did not use compression on the strings because I felt they sounded nice and had a lot of "air" to them as it stands, but that might be debatable.

Generally I felt that the instruments did not call for a as much treatment as for example a rockband setup. Levels and panning/three-dimensional space seem to be very important...

- Everything has a send that is routed through a channel with some glue compression and a reverb to simulate a room for the instruments (a big space from the NI RC 24). This signal is fed into the master out on a low volume.

- On the master out in the DAW we had again a tiny bit of glue compression and eq-wise a low-cut and a little bit of a boost in the higher frequencies starting at about 3.5khz

Mastering:

- We mastered everything using software (T-RackS CS) as we do not own an analogue mastering setup. The chain was EQ->Compressor (working around 1-2dB reduction)->EQ (for some boosting)->Limiter. I am still unsure on the limiting part, especially on the ceiling. At them moment some of the tracks are at a ceeling of -0,8db and few go up to -0,2. I am thinking to go with the lower levels. But on the other side that reduces the dynamic range.

So that's roughly it. I am still debating whether I should try out a different reverb for the Room-Channel, whether the strings need more processing, whether the drums should sound less bumpy and be more in the background etc. etc. smile.png

Again thanks for listening and thanks for reading. I will be working on the remastering during the next few days so any comments would be greatly appreciated!

Cheers,

Patrick

Breakdown Epiphanes, Musicians from Hamburg, Germany.

https://soundcloud.com/breakdownepiphanies

Hello Patrick,

I'm not going into detail on each track, but generally speaking your mix is a bit of a mess. I think you already know that, it's why you're asking for help. :)

Mixing:

Step 1: The first step towards a great mix is a great orchestration. Going into that would require a deeper analysis of each track, but for future reference I recommend you study these Principles of Orchestration by Rimsky-Korsakov. There are a lot of things that could be improved for better balance, but for a first time working with an orchestra, this is pretty normal, and you seem to have a relatively decent intuitive understanding already. Pretty good.

Step 2: The second step—and this is basically an element of orchestration—is pan and volume. As you noticed correctly, pan and volume are very important. In this step, you wanna achieve a perfect stereo balance already. So you wanna pan your instruments, and set levels, and then if any parts in your songs call for it, automate volume faders on specific tracks to create a better balance throughout the song. If something doesn't take enough or too much space in the stereo field, this is also where you use effects like chorus or stereo enhancers and so on on individual tracks.

Step 3: The third step is where you listen closely for any frequencies that clash or take away clarity from another instrument, and then use EQ to cut all of that away until everything sounds crystal clear. This is also the step where you use EQ to fix instruments that sound a bit too thin, or something that sounds a bit too nasaly and so on.

Step 4: The fourth step is where you use compression to tighten instruments up that could use a bit of tightening (most often drums), or use reverb for effect (like on drums).

Step 5: The fifth step where you address the third spatial dimension in your songs: depth. Any instruments that sounds too far away, you bring closer to the listener. Anything that sounds too close, you take farther away. You can achieve this with reverb, compression, and EQ. Don't do this excessively, just do it where you need it.

Step 6: The sixth step is creating coherence or a certain ambiance. This is where you choose an ambiance reverb that has the feel (ambiance) you are looking for (wide open, close and cornered, etc.). Send just enough of each track to this reverb to create this feeling and coherence, but no more. You don't need to hear the reverb, you only need this feeling of coherence and ambiance. You can usually cut the low end on this reverb to retain the clarity in your bass frequencies. This is also the step where you can use “glue” compression on the master buss for coherence (I generally don't think it's needed, but some people like the sound).

At this point, if everything went right, your mix should sound perfect to you. You should feel that if you sent this to a mastering engineer, you'd have no idea what he would actually want to fix because you feel everything that can be fixed has been fixed. If you don't have this feeling, then look through your settings and fix things where they need fixing. Take breaks. Listen to reference tracks that have good mixing. Then come back to your mix and fix everything that you didn't pick up last time. Repeat ad nauseam or until you feel your mix is great.

Step 7: Send your absolutely awesome crystal-clear mixdown to your mastering engineer, wondering whatsoever he could even find to fix that you couldn't find.

Mastering:

Step 1: Do you hear all these absolutely awful mistakes in the mix with your crystal-clear high quality mastering monitors* that your mixing engineer couldn't even hope to hear? Is it bad? Call your mixing engineer and tell him how to fix it in the mix. What, your mixing engineer is unavailable or says there is no time to go back to the mix? Go to Step 2. Does the mix only need minor fixing? Skip to Step 3.

(optional) Step 2: Use all your high-end multiband compressors and so on in an attempt to save the music in whatever way it can even be saved.

Step 3: Use your (preferably linear phase to avoid phase issues) EQ with very narrow bands to apply surgical procedures to the two or three very narrow frequency ranges that require a less than 1dB cut or boost (because anything more would fall into step 2).

Step 4: Set your limiter ceiling to whatever max dB you want for your project (typically between -3 and 0, but usually at most a tiny bit below 0 to avoid clipping when rendering to certain formats (like low-quality MP3)), then for the rest of your project forget you even have that knob. Seriously. Don't touch it. Nope.

Step 5: Pretend the other knobs on your limiter don't exist; or pretend you have a clue how to use them effectively and fiddle around until you get bored, hoping you don't break anything in the process. If you do somehow break anything and everything sounds worse, set these knobs back to default, then pretend this never happened and these knobs don't exist.

Step 6: Crank up your gain until your levels are just right, or you have achieved the desired level of distortion from clipping your limiter if you like how it colors the sound.

Step 7: Automate volume faders (before your limiter in the processing chain) as you deem fit.

Step 8: Dither if required, and render to target-format.

*In the event that you should not have crystal-clear high quality mastering monitors (like most of us mere mortals), listen on different devices or through different monitors and headphones to find any problem areas.

Generally speaking, the farther back in this chain you can fix anything, the better the quality of your mix will be. Now I know not everyone will 100% agree with the order of these steps, but I would think the steps themselves are solid. Any processing you do on the master buss can technically be considered mastering. In that sense, your mastering chain, as you told us, is literally Comp -> EQ -> EQ -> Comp -> EQ -> Limiter. I think you can see that looks a bit silly when Comp -> EQ -> Limiter should be just fine. :) That's generally what I noticed most in your songs: There seemed to be too much processing for the sake of processing, and everything seemed unclear. Too much ambiance reverb that took away more clarity than it added coherence and ambiance. The cellos were sitting right next to me, but the violins were at the far end of the room. The stereo balance was a bit off in some songs, and in some the center was rather crowded.

I suggest you ditch whatever mastering you have done, and clean up your processing chains and fix everything that you can fix in the mix, where you have much more control over every individual track, until you are truly satisfied with what you hear. Then, and only then, get to mastering your project. For an Early Access release the quality seems fine, so I would take the route of really going into your mixes and fixing things now rather than trying to save things in mastering for the EA, because that will save you a lot of time long-term for the final release. Try reaching out to some mastering studios and see what deal you can get. They couldn't make a living if they weren't affordable, so that might be a good option for your game.

Perhaps someone else can chime in with more specific advice on this or that (or any additions or corrections if I messed something up).

Good luck with your game,

Chris

Advertisement

Hi Chris,

thx for all those helpful general guidelines and the good wishes. We will definitely keep your suggestions in mind when revisiting the mixes which is on our to do list, especially having a closer listen to whether there is too much room used on the tracks and where we can cut some processing in the chain. Unfortunately giving the stems away for mastering is way out of our budget calculations for this project. So we will work on the mastering ourselves using pretty decent ADAM monitors and a room we are accustomed to.

Cheers,

Patrick

Breakdown Epiphanes, Musicians from Hamburg, Germany.

https://soundcloud.com/breakdownepiphanies

Just a quick note, when you're mastering the last step is to go to your near-field monitors and/or 'flat-response' headphones. Because if your mix sounds awesome on your good mixing monitors that doesn't necessarily mean it will sound good coming out of the end-user's laptop speakers or cheap-o headset. You'll want some good headphones that accurately reproduce a consumer-level sound like the Sony MDR-7506 or Yamaha NS-10M monitors.

Also from what I'm hearing everything sounds very muddy, Nyaanyaa already covered what you'll need to do to fix the mix. But, one additional thing I might try is to throw an aural exciter on there after your multi-band compressor and before the mastering limiter. It should make your mix 'sparkle' a bit more. The Wave's Aphex plugin is pretty good for that, it's a psychoacousic effect so it can be difficult to get just right. The trick is to get your settings to the point where you think you hear it, not where you hear it working for sure =p

Hi CCH! Listening to the tracks on different systems and with headphones is on our list. I'll have a closer look at the Aphex plugin though I will rather use it in the DAW for the mix as I don't think that the T-Racks standalone mastering suite supports vsts. I'll be careful though as you guys already pointed out that there might be too much processing going on overall. Anyways, thanks for the good suggestion, the plugin looks very useful in general.

Breakdown Epiphanes, Musicians from Hamburg, Germany.

https://soundcloud.com/breakdownepiphanies

This topic is closed to new replies.

Advertisement