Mix Challenge

So well put.
I also think it's a lot to do about creating credible illusions.

With respect to listening environments, absolutely, that is such a massive factor.
Perhaps you wouldn't listen to certain classical music on earbuds, or a cellphone speaker, on a bus/tube, right?
I don't think I would.

Perhaps it's safe to say that some forms of music don't translate very well into those kinds of environment.
I'm a headphone junkie and I guess that works quite well across the board, if you're motivated to listen and cooperate a little (like @scottfitz mentioned, about the emotional impact).
After all the sound, the image, they're all formed in your head. Consumer Vs listener again?

I'm trying to be aware of the inherent danger of this during the mixing process. You can tune in to your mix so much, that obvious flaws can turn out really hard to spot. Everybody makes this mistake. Beginners like me more than experienced engineers, that's obvious. Taking breaks is essential (or so I have heard).

trackerjack wrote: ↑

Sat Aug 26, 2023 08:49 CEST

I could have spent another few days tweaking, too, but I was really worried that I would end up making it worse and at some point drew a line (pun unintended) and thought, now is the time to wait for client feedback. To see which way to go from here, if I get a chance. But I also knew there would be a hot weekend coming up, and no way my PC could have coped with that. It shut itself down several times during rendering when it wasn't that hot. It's a very old machine, I'm saving up for something new. Until then - anyone have tips how to adjust rendering speeds in Reaper? I would be fine with real-time, but not a fan of bouncing.

Hi!

When Rendering to file, there's an option in a scrolldown menu to the right to chose Online render ( default is Fullspeed offline)
I pretty much always use online since I every once in a blue moon like to incorporate my dustcollecting hardware units, plus it's the golden opportunity to listen for mistakes one more time still....And boy do I find them! It takes a few runs before i don,t stop it half way.

Also in Options/preferences/Rendering you can tick the box saying "Limit Apply FX/render to realtime" or something like that.

I assume you have already set your buffersize to 1024 or higher whemn mixing, otherwise try that aswell.

Hope this could help!
Cheers! / Hoek

Ah, very cool, thanks @Hoek!

Yeah buffer size is way up there.
I was surprised how far I got on this old mule.

It certainly didn't help that first try, I left the Youlean GUI up. Holy cow.

Been there, for the longest time I made do with my hardworking champ hairdryer of a desktop build from 2011 and it usually let me get away with it, surprisingly few projects that let me choke it! If it did I was reasonably happy with freezing tracks. Upgraded in 2019 though to a middle of the road Ryzen 3600 and now I think I'm set for life! Room is much cooler and dead quiet too! Managed to run some 600 instances of pro q 3 if I'm not mistaken when trying to max it out. I usually never use more than 560!

Maybe check this video out aswell, got my system running like a dream!

https://youtu.be/8z-07hWqBmc?si=cjXv5Mww7SXOJ43C

Reaper is just awesome, I've been on all kinds of Cubase since 1998 but switched to Reaper 10 years ago and never looked back! It's also the most CPU efficient Daw there is.

Hoek wrote: ↑

Sat Aug 26, 2023 15:49 CEST

hardworking champ hairdryer

Beautiful.
I'm considering to go M2(Pro) Mini some time soon. That should probably cover my needs for a while.

scottfitz wrote: ↑

Sat Aug 26, 2023 01:05 CEST

Anyway the point of all this is that throughout the mixing process I was wondering how can we do a thing like set a global vocal compression setting and similarly on other elements when this is such a blunt and almost disrespectful way to treat the music? Perhaps the constraints of time simply don't allow us to do more, but I did consider splitting the music into separate sections which each had their own compression settings, sadly I had nowhere near enough time to attempt it. Did anyone else start to think like that or have a different way to understand what needs to be done?

I did it in Gaz's voice. I divided it into three channels, vox soft, vox strong and vox fxs. For the first two channels, I used the same compressor with slight differences in the settings. I could also have kept a single channel and automated some parameters, but I find it much easier to create new channels for respective processing.

That is a great tip, @rvalle !!!

I hope I am not going off on too many tangents here (sorry if so, I will shut up now). But I think this one is such a sweet example of great "pop" music that gives the impression of being very dynamic, albeit more on the soft side overall (by one of the greatest bands ever, sorry if you can't follow the lyrics, or don't like this kind of music - maybe check out their first album which was performed entirely in the lingo of our Anglo-Saxon friends). Very different style obviously, but they were also ridiculously progressive.

https://youtu.be/Ov_bbFCZ62k

'Any World' and 'Doctor Wu' (Steely Dan) are also very nice examples, imo.

Edit: With Doctor Wu, I think it's interesting how the side stick (which seems way too loud if you focus on the rest of the drums) feels like anchoring the volume of the first verse... I hadn't noticed before. And that some of the cymbals sound shockingly bad, too, haha.

Just to quickly respond to certain brought up topics.


A) Username changes can not be done by users themselves. Only the admin can do that.



B) Queen's "A Night at the Opera" was recorded/mixed in 7 different studios actually. But the mix console that their most famous track on that album was mixed on (and actually the first single, which also broke the "3:30min radio limit"), was a Trident Audio Developments "Trident B", which did indeed use "computerized VCA" busses that happened to crap out due to all the controlled layers having no more headroom, and then they had to mix half of the song by "hand automation".

https://vintageking.com/blog/2018/12/queen-night-opera
https://www.soundonsound.com/people/queen-night-opera



C) I've had this conversation on various Discord servers as well, with people struggling with their audio rigs' CPU usage and the multi-track amount.

While I mix, I usually never touch the DAC buffer size, even though I could (and in rare occasions, I do need to). I also don't really remember the last time I utilized channel freezing. And I am using Cubase, and I still work on a 1st generation Intel Core i7 on occasion (equivalent to a fully fleshed out iMac 11.3 -- but no PSU limit and better cooling to an oversized heatsink -- though granted, my CPU is overclocked, and my OS is highly tuned). In fact, all multi-tracks / integrity checks in recent years have been checked on this very rig with a SSD as "playback drive".

Maybe I'm just working different, with still porting "concepts of old" from analog equipment days. As in... only have 3-4 modules per channel strip (VU, console emulation, EQ, Compressor), and utilizing both mix groups and AUX sends.

Some food for thought maybe




I did get no more response (via email and PM, actually) regarding me reaching out to a possible accidentally withdrawn entry.
The Statistic Sheets will arrive soon.

Thank you for your patience.

Thanks for the tips, my current machine is some Thinkpad Core i5 (ca. 2010), with 4GB and 256 SSD.
Which was bought second hand, lol. Fan has been replaced twice. They're good machines but it's on its last leg. It's also rather hot in my room, this time of year. After this challenge I will probably make the switch, it's long overdue. Also trying to be rather conservative about plugins.

Just kidding about the username.

rvalle wrote: ↑

Sat Aug 26, 2023 16:40 CEST

scottfitz wrote: ↑

Sat Aug 26, 2023 01:05 CEST

Anyway the point of all this is that throughout the mixing process I was wondering how can we do a thing like set a global vocal compression setting and similarly on other elements when this is such a blunt and almost disrespectful way to treat the music? Perhaps the constraints of time simply don't allow us to do more, but I did consider splitting the music into separate sections which each had their own compression settings, sadly I had nowhere near enough time to attempt it. Did anyone else start to think like that or have a different way to understand what needs to be done?

I did it in Gaz's voice. I divided it into three channels, vox soft, vox strong and vox fxs. For the first two channels, I used the same compressor with slight differences in the settings. I could also have kept a single channel and automated some parameters, but I find it much easier to create new channels for respective processing.

That's referred to as "multing" and it's used a lot.

In regards to the dynamics discussion earlier, I came across a quote from Jacquire King a while back where he said that for "bigger" sections he likes to add something that adds "the appearance of loudness" i.e. it sounds like the instrument is being played louder, or it's become louder using some distortion or whatever. In terms of absolute level it might not actually be louder though. It's "smoke and mirrors"!

Cheers!