if you are going to self master

is there any advantage about rendering before mastering? except faster processing cpu i guess

thanks

 

Btw, when I export my mixes I usually do it in 32-bit FP, and at the same sample rate as the song is recorded in ofc (to avoid any unnecessary sample rate conversion). You don't have to export to 32-bit FP, but I rather keep everything in as high of a quality as possible until the final exporting of the song, to the various bit and sample rates needed for different formats.

I have Reaper set to render "apply FX, glue, freeze etc" to 32-bit FP for this reason as well. You can change this in Project Settings under the Media tab. If you have plenty of drive space this shouldn't be an issue, but if you don't have much space to spare you can leave it at "Automatic: 24-bit WAV".

 

Your first answer sentence is perfect. The rest of the workflow stuff is dependent upon the user and their choice of DAW. Or even the combination of different ones. Because he is not sending it to anyone.

In this scenario, of a user "self mastering", they are probably also "self-producing" and "self-mixing" as well. So sample rate conversions are really an immaterial discussion. They would all enter at the DAW's bit and session depth of the project at some point already. Unless the producer is making lo-fi, there is no reason to not upsample the material during it's initial import or internal recording. It's not even an internal bounce of difference. The first set of stems cements this to whatever your DAW's internal bitrate/session depth is at that point in time as well. All you accomplish by not doing it earlier, as impressive it is that REAPER can; it's just procrastination to upsampling time later. And a good way to overly complicate something within your explanation.

It also lets you check how much you like the sound of any dithering up/down that will happen with each converted piece of material individually. Rather than the sum of it across all your channels at one time.

 

Last edited: Nov 6, 2022

All the above + consider that at times mastering is done to not one song but an entire album/ep. You can't do that (comfortably) unless you either render the mixes first or have one big ass mixing session which could be very resource heavy even for high end machines.

Also, mastering can get fairly complicated, some people would go as far as using different compressors or EQs for differnt sections of a song, manually turn knobs as the song goes, etc. That's just extra headache that you don't want to add on top of an already busy session

 

You can't master your own stuff. So don't worry about it. Just make a good mix.

 

The answer is as follows:

If self mastering a single song, there is no benefit (besides CPU load potentially being lower). If CPU consumption is not a concern then full session mastering actually provides advantages that the 'old way' of mastering does not, such as being able to tweak and make adjustments on the fly as you hear what the mastering chain is doing to your mix. For example, I always keep at minimum a limiter on my master bus, and mix through it as it helps me get an idea of what the finish product will sound like once maximized. To me this is an invaluable workflow technique even if you aren't going to master your own songs. It helps you avoid any unexpected surprises during mastering.

Anyone who says you can't master your own songs are behind the times and don't understand how things have progressed. Not to devalue any skillful mastering engineers talent, but modern engineers are expected to be able to serve as mixing and mastering engineers, especially on smaller releases. It's rare that I see independent release that is sent to mastering houses nowadays. Even major label releases really aren't getting much treatment from mastering engineers today. Mixes and albums are many times sent out for mastering just as an old school protocol, even though the mastering engineer may send it back with minimal or no changes. That's because best mix engineers are no longer relying on mastering to fix or polish anything, because all of the tools for mastering are available to everyone.

Both Serban Ghenea, and his partner John Hanes, as well as Jaycen Joshua are examples a few engineers that send their mixes off pretty much finished, and will tell the mastering engineer "if you can enhance it at all, go ahead" But many times now a days, a little maximizing and formatting is all the mastering engineer will do. Here is an interview with John Hanes about him and Serban's process, search for the section where he talks about their 'mastering bus': https://gearspace.com/board/interviews/1362525-interview-john-hanes.html

I've heard both mixes from Serban, and Jaycen before mastering and can tell you the difference for some mixes after mastering were zero.

In short there is very little need delegate mastering into a separate stage, especially if doing singles. For an album, worked on by different mix engineers, a more traditional mastering approach makes sense.

 

Last edited: Nov 8, 2022

What @starkid84 said - I agree. The word 'master' gets thrown around loosely nowadays. Many times, a label will say, "Don't master your tracks. We'll master it for you. Submit them between -20dB and -3dB."

What the label is actually going to do is make sure all the tracks peak at -1dB or -0.5dB so there is uniformity and consistency across the entire disk. They don't master in the sense that they tune the kick to the bass, level the vocals, reduce resonance in string patches, bring the background vocals forward, and so on. They can't because they'd need the individual stems.

 

That's literally what mastering used to be. It was to make songs conform in a project, that or be within tech limits to survive cutting to vinyl. It was NOT some magical final stage of the mix. George Martin and Alan Parsons didn't sit around going, wow, these mixes are good, I can't wait until they come back from mastering. That was it. If anything, they worked hard to ensure that what got cut to the album reflected what they had done.

 

Sure you can. Just make a bad mix and a good (stem) mastering, problem solved.

 

It was preparation for Manufacturing of a Product, along with everything we refer to it now as. This includes the preparation of Acetate Glass Masters (aka dub plates), which the artist would most certainly get to listen to prior to agreeing to make a million vinyl/cassete/ or whatever other media real copies, with their names all over them. At minimum their engineers checking them. My point is, yes they were truly waiting on every little stage of this stuff. Just like now, likely a lot worse turnaround times on every one of them. (delays).

 

Last edited: Nov 8, 2022

But the point was to preserve the sound for the limitations of the respective medium, not expect some kind of improvement. For essentially transparent mediums like open reel tape and later cd there would be no mastering beyond making it not clip. Adding eq and compression would be considered a sacrilege against the producer's mix. It was vinyl where compromises had to be made. The idea of "the mastering guy will make it sound better" came in with home and project studios.

 

Yes I completely get it from the perspective of your point, because it is 100% true or as close as necessary. There was no "make it quietly and bring it up to full volume inside the computer" stage that people now call mastering. This has only really become standard fare since our machines have all become so much more resources rich. higher bitrates, more headroom, and so on. But whatever you call Post, the idea of Paul McCartney sitting around with his thumb up his ass waiting to listen to it, is most likely not what happened in reality either. Those two things can co-exist.

 

Last edited: Nov 8, 2022

No, I agree. When they heard the mix in the studio, it was done.