Before you watch this video, make sure you've watched this one: • Why MANY ANALOG EMULAT... MERCHANDISE: teespring.com/... EXTRA CONTENT: / whiteseastudio STUDIO: whiteseastudio...
you're absolutely 100 percent right about the details. its the same with any art form, attention to detail (without losing overall perspective, of course) is the thing that really shines through with its cumulative effect.
In the Bounce to Disk settings of Pro Tools you can select an option called 'convert after bounce'. This makes sure you bounce in the session's sample rate before converting the result of this bounce to a lower sample rate.
My first finger-up for you, - hey man, you‘re able to describe important things, incredible : go on with this- your nervous style gives this sweet appearance of reliability :-)
this is the topic where ify pondering and guessing youre alright will set you back years. youll get all invested in gear and work all day, ruining every mix by bouncing wrong. youll even stop checking the mixdowns, the repeated truama of having your vision destroyed will be too much to risk. you start rendering at -6db and hire a mastering engineer eventually, assuming you have no ears and never did.
Thank you from the bottom of my heart! This is something that has bugged me forever. I'm just a dude in his bedroom playing around but I want the best quality that all my amazing PC can put out :-)
I am so happy you are doing videos that take on this subject. Solving these kinds of issues have been the keys to me finding a clean sound on a DAW. You have certainly aided in my understanding. Thank you.
Ok people here is what I recommend as a professional studio engineer. As you record each of your tracks, add your effects, automation, etc. Then, right click the track if in Reaper and select Apply Track/Take FX As New Items. What this does is that it lowers your CPU use so that before you start the project you can pre set the project as 192 instead of 44.1 This turns your track into a .WAV. Remember though that if you do this and you want to edit your tract you can revert by turning the track to the following state called Active again. When you have done all your mixing and you want to render/bounce then just bounce your project as usual that will each be a .WAV and you'll have a great sound rendering it as 192. My favorite mastering plugin at a reasonable price recommendation would be the Abbey Roads Mastering Chain and select the preset called Balanced. Here is one instrumental where I played all the instruments and did what I recommended to you guys. P.S. the guy throughout the video is not me and you can see what my real name is. Cheers. Frank (New York City) ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-D6hZtFMx12o.html
It's like this I always say people never even finished quotes because they don't even understand too take everything with a grain of salt means if you like the taste season your food with it not just judge everything and be skeptical that's why I appreciate your videos man thank you
Only reasonable thing to do now is to make your next video a blind test with different audio samples at different sample rates to prove that it matters and actually sounds better. Perhaps you would not be able to tell which is which in some cases. Make examples for us as well. Play A/B parts bounced out at different sample rates, then play them without showing which is A and B, then we can hear for ourselves. I have had so many discussions with audio engineers that claim this and that, then we do blind tests and they make a fool out of themselves. Would make a fun video either way :)
Yes, I agree and I was all was always taught if you can't hear it don't do it. Makes no difference to me anyway as I run on a core duo with 4gb ram, it only just copes with 48000.
J W while i agree 44.1 is fine, there are benefits to going higher at least in the studio and they have to do with pushing aliasing artifacts out of the range of human hearing increasing the quality and authenticity of analog emulations and synth, saturation and distortion effects, also reducing latency,
"foldback distortion" refers to a kind of waveshaping, not aliasing. check out the free vst Driveshaper for an example of how foldback distortion sounds.
This kind of thing is what makes people record a bounce inside the DAW in real time. Sending Mix Bus to an Audio Track then press Record. Your recorded file is your bounce
After reading one book about how Digital processing and how things are converted, it's become a primary concern . Right next to Acoustic Treatment and Ear training.
Wow, thanks! Bringing up the project sample rate to 192 before bounce made it sound a lot better and/but showed in the sound overuse of the soothe plugin, because im taking away harsh sounds that only appear in lower sample rate that im mixing in...
Cool, another video :) I love your Snake oil vids but these really set a balance and show that you really know your stuff. Very informative and interesting . I use reaper so this is useful for me, thanks.
Man, this was the best tip ever! So I have to condition myself to think of setting 192 kHz in the 'project settings' before rendering. While producing/mixing I would leave it at 44.1 because otherwise my RME (or let's say my CPU) can't handle this amount of processing well.
Hey there, actually heard about that only a handful of producers tempt to organize their projects like that. I was never certain if I should do it either, but this video gave me a clear answer. Thank you for that.
Hey thanks , I know this is an old video but it really helped me . I have been struggling to render a finished track and getting it to sound just the way I hear it in the mix . This works ! -cheers from Canada
100%... Actually you're correctly thought out on everything I've watched of your channel. This video educated me on some questions I've had, great work!
This is the message I've been trying to get across to all my friends who are kind of on the fence about how the world is going complacency is the devil whether or not you believe in one or not it is the very thing that destroys Humanity diversity and detail is everything because we are all the same yet so unique! Wake up people!!! Be different, search for YOUR sound👊👊👊👊and keep doing what you doing man I appreciate your videos and what you're doing
Kclip Zero bounces at 8x even though it has no switch. But its weird because it sounds different then what you were working on. Which can be confusing for beginners. I believe Sytrus also does this in FL. Which drove me crazy when I was younger because i really like the aliasing sound for background dubstep effects. But for anyone who is thinking of buying flatline or knock or standard clip, I highly recommend FreeClip. It has 32x oversampling switch and beats out most clippers in sounding analog because of it. It also has 5 different softclip modes. The quintic setting sounds very proffesional when bounced at 32x. Other plugins might seem better because they have a percentage of how hard to clip, but nothing comes close to FreeClip. It just sounds so clean and analog. I stopped using Kclip because of it. Kclip is so good on paper but it sounds digital and almost exactly like Fruity Clipper but with more options. Some guy did a comparison and showed that FreeClip has the least digital artifacts out of all of them, even though it's free. I've never tested flatline myself, but it seems to be snakeoil. Maybe I'm wrong.
In Pro Tools, I have always bounced to disk at 24/96 stereo, then brought the resulting file into a 24/44.1 pre-mastering session, letting it down-convert upon import using the best settings. I then pre-master my final to 16/44.1 with medium dithering.
your best to keep all files at the higher recorded rate until after mastering and then convert them for specific mediums like cd via good dithering software .Your approach is wrong totally as downsampling via import is both a substandard quality and also its pointless prior to mastering where the higher quality will allow better subjective evaluation of the music for mastering.You should keep the main master at working rate and secondary masters at lower rates off the back of the finished master.
@@cornwallradiophonic6250 You are absolutely correct. Nathan's approach is so weird and unprofessional that I was wondering whether he was a troll? By the way: use Voxengo's free R8brain for SRC - it's better than most DAWs SRC algos!
@@cornwallradiophonic6250 Those are good suggestions except for the understanding of the sample rate conversion on import, which uses exactly the same set of algorithms (for the various speed vs. quality settings) as if you converted when bouncing, at least in Pro Tools, which is what I have used for years.
It would be cool if you made a series of process checks that helps illustrate different concepts like this video, with example files and different scenarios to help drive the concepts home.
I've seen a few of your videos, this one made me subscribe. I was not even aware that rendering at the project sample rate wasn't automatic. Keep it up!
HQ offline rendering is supported by some plugins, and several of the clipping plugins i use (standard clip, v-clip) offer internal oversampling to counter aliasing. Setting my interface to 192kHz uses a lot of cpu, so I’d rather stay at 44.1kHz and use oversampling where needed (in the plugins themself).
you just made my day with video..we are learning about this rn in my audio course and this rly sums why i love to put high sample rate for projects, and we render at 44.1! :D thanks!
Thank you so much for your info on aliasing. Being aware of this issue has had me second-guessing my use of saturation plugins to ensure I'm not muddying up my busses. Plus, this episode may help me to figure out why my mix downs, in Reaper, always come out sounding more muddy than the project. I assumed it was the down sampling, but more specifically, it could be the aliasing caused by these settings. Much appreciated. Keep up the good work.
Very true. Great video! I completely agree with - I want a bounce that is exactly the same as what I hear. The only way to get that, that works for me is to use my virtual channels on UAD Apollo (and I'm guessing you can do this with other I/Os that have more advanced mixers, RME comes to mind) and record back to DAW ( I use S1). After that I use Ozon standalone to covert it to whatever format I need...
@@dodo.danciu Hi bud! So, the concept is simple: You need to go from your DAW to UA console's virtual channels and then go back to the DAW by making a new track that has those channels as input (and mute them during recording). Most of the DAWs will automatically assign the available inputs and outputs (if not then add them yourself) and then you only need to link either 1 and 2 or 3 and 4 Virtual channels in UA console for them to work in stereo. I hope I was clear enough :)
In my opinion and from my experience there is no need to use plugins only in mixing ( it is all about balancing the tracking) .. once you hear everything in a perfect way , your job is done which leads to less is more.. in mastering yeah you need plugins that adds taste to the sound
Wow that is incredible. I had no idea that Reaper was doing that in the background. I'm going to bounce at my native resolution from now on and then do the conversion separately.
Some Voxengo Plugins can be set to use maximum oversampling in bounce mode. I don't know how well this works, but since I got plenty of CPU I usually set these to permanent 4x oversampling anyways. Which results in 196khz processing inside these plugins since I usually work in 48khz in Mixbus.
INSECURITY is a HUGE problem worldwide and it's the main reason people do not listen or learn. It's not that these people don't care, they care about the wrong things, usually their own ego and attention seeking, the sort of person that wants to be an influencer with no actual ability to do anything other than seek attention. We ALL know at least one of these idiots, if you don't know any, then it's likely you're one yourself as these people lack any real self awareness too, that's why so many of them end up as DJs and not musicians! Taking the time to learn and develop the skills needed just gets in the way for these people, you don't have an audience when you're practising by yourself and they struggle with that, as they are doing this purely to get attention. This is all you need to know when dealing with people like this. Interesting video too!
Wow. This is actually really interesting. I am working in 48k for Processing power... I think i will try this next time and Render at 192 k. Should work If i understood correct.
Awesome thought provoking video. If only my CPU could keep up with realtime 32x oversampling all all plugs at once. Is there was way to pull this trick off in Ableton?
Ableton 10 automatically renders in higher sample rate and then exports at whatever you choose. If you set it to high sample rate in project and then go to export it tells you that it will render at higher sample rate
Great info!, I feel that maybe is a lot to ask, but many of us that follow you could really benefit if you could do the same experiment of exporting and checking distortion in different DAWS. I feel many of us want to improve our sound, but in regards to sample rate conversion we are a bit lost; plus, the options for exporting are not the same in every DAW. Maybe if you can help the community you have built many of us would be forced (in a good way) to jump to another DAW to mix the project, like e.g producing in Ableton or FL and then exporting the session to Reaper or Logic or whatever has the exporting option that has the less artifacts. Thanks for all the information you are constantly delivering! Greets from Argentina
ummm... As a mixer with an older computer, I'd rather be able to use more plugins and more tracks. I have to do tons of bouncing, splitting projects into separate pieces, disable all kinds of stuff as it is just so I can keep working and I'm thrilled when I can even barely finish a project on 44. It feels like I performed a slight miracle every time and I deserve a medal or something. Given the choice, I'd rather run a high quality convolution reverb (which I can't do) then run at a higher sample rate.
Great video, I love the technical science side of audio and this is very useful information. Sometimes I see people post in forums about when they render a file it sounds different than when they were mixing it and I've always wondered what could be the cause of that and this could be one of the explanations in certain cases. Also what frequency analyzer are you using with all the pretty colors?
I think that the reaper test tone generator it is not band-limited. With "normal" VST instruments, you shouldn't see that enormous amount of aliasing. I tested it by initializing a Retrologue patch, setting OSC1 to triangle and generating a test tone of ~ 16.4Khz @ 44100hz sample rate: no aliasing. At the same time, I agree with you that with higher SR some virtual effects and instruments sound a lot better! Cheers :-)
Curious to know people's thoughts on Bouncing to disk, or just printing to an audio track within the session, then exporting that newly made audio clip. I can say with 100% confidence, that bouncing to disk will give you a slightly different sound to sending your aux mix bus to an audio track and recording/printing that in session. For those who are skeptical, try both of these methods, then open a new session with both of the audio files on seperate tracks, and then phase invert. In theory, the result should be silence, however it is not. As for which method is better, that is subjective, however I have found to take a liking to the printing method as opposed to 'bouncing to disk'.
I use 48kHz as the project sample rate and also bounce the tracks in 48kHz. This is a really interesting topic. I will definitely try out sampling down from 192...
This is really interesting... I never knew about this trick, but I've been doing this ever since I got an interface that can do 196K...good to know that I was already doing things "better" :)Seems to be a reoccurring theme with me. I watched a "pro-guide" and I was already using all the plugins they were using for the same thing....and pretty much the same way, but I'm just guessing my way through all of this...self-taught on everything. Great videos!!
Holy Sht! I thought about this all last week! But I didn't have much time to check it! And YES! Of course it works! OfC it will be sounds better! But i don't have much system power to work in 192, and I think I can change sample rate when I bounce master track. Its not a great idea, but its better then bounce in 44.1 or 48kHz. Its crazy, to work in 48, then, when you bounce, you just change sample rate to 192 and BOOM, you have two different tracks: one what you hear, when do the mix, and one bounced. Thank you, man! Great youtube channel!
Agree, small things became a huge thing if we don’t care about it. But the problem with that way of thinking, is that most of the people wants to have results without effort. But, no pain, no gain my friends.
It is great to know that it actually makes sense working at higher sample rates when you are processing and harmonics are newly produced (is this right?), and it is a bit weird to see that you need to ask Reaper to convert after exporting, I thought that was a given, Ardour and Mixbus are not as flexible as Reaper but this is not a conversation needed there. - Quick question: assuming you were working at 44.1Khz, how would you avoid these fold back harmonics, does it make sense adding a hard LPF on the master? Buses? Tracks? Nowhere at all? Sorry, this is not the most simple topic 😅
You'll find a lot of plug ins have built in oversampling, but, the issue is you then rely on the accuracy of their own SRC, so yep, best to work at 96k or above if you can. What he doesn't mention is not all SRC is equal, Izotopes is top notch in tests, ie, RX7. There's a website somewhere that tests all DAW similar to how he has here, can't recall the name but I'm sure you could find it. Btw, most of the plug ins have intelligent auto sampling, so if you run at 96k it will know and either disable or maybe reduce to x2 depending on the plug in.
I’ve found myself getting ridiculously obsessed with getting rid of aliasing recently (doing crazy things like running at 192 at 16 or 32x oversampling) but then I think.. often it’s only really at a problematic level when you’re really smashing it to destruction, and if you’re just using mild saturation or something, perhaps all that is really achieving is adding needless processor load. I dunno.
It depends, if your doing any higher frequency saturation then you should oversample but really 8x oversampling is enough in almost all cases, a couple instances of aliasing distortion in the odd track or two isn't really an issue it's when you start using tonnes of saturation on busses or your master that it's a problem because it's introducing inharmonic frequency content along with the innevitable intermodulation distortion that occurs with buss processing, if your subtle with it, it's barely an issue but it's certainly something to be aware of along with things like the nyquist shannon theroem and bit depth.
