Last night I recorded a guy at 44.1 even though my intention was to record him at 48. I could have sworn I set the converter to 48, but apparently I didn't. Anyway, when I went to mixdown the song, I set the sample rate to 48 on the converter and the song sounded great on playback in Cubase. The pitch was higher and it felt livelier. When I mixed it down, both at 44.1 and 48, it didn't sound as good. The pitch was lower again.
So, not knowing enough about sample rate theory, was the converter upsampling the files? If so, why didn't it bounce to the same pitch?
Thanks to anybody with experience with this who weighs in on the subject.
As an aside, I mixed something that was in 96 last night and it really sounded great. I have been avoiding going 96 because my Hearback is connected via adat and it screws with the channel count and routing. Might just have to del with it.
As an aside, I mixed something that was in 96 last night and it really sounded great. I have been avoiding going 96 because my Hearback is connected via adat and it screws with the channel count and routing. Might just have to del with it.
Any idea about the pitch sounding different?
I've done things at 96 as well. I stopped due to the hassles of downsampling and going back and forth changing the SR at the converter/interface. However, I just got something from that was recorded at FAME at 48 and thought I'd try it.
Yes, you're not "upsampling". You're playing back a 44.1 sampled audio at the rate of 48khz. If that's how you want it to sound, you need to capture it with a second system via analog connection.
Yes, you're not "upsampling". You're playing back a 44.1 sampled audio at the rate of 48khz. If that's how you want it to sound, you need to capture it with a second system via analog connection.
Thanks popmann.
Here's a video of what it sounds like when the files are played back at 48 and 44.1. The first shows my RME interface showing 48K, with Cubase at 44.1. The second shows RME at 44.1 and Cubase at 44.1.
I really don't understand the relationship between Sample Rate and tempo and pitch. The weird thing is, I swear I recorded at 48 and that the first audio in the video is what it sounded like live. Is there something that Cubase could have done to downsample the file?
Last Edit: Feb 16, 2017 12:07:50 GMT -6 by swurveman
Go into the pool. Cmd+P....the details of the file rate is listed in it's attributes.
Cubase can and will resample the file--sure. If you change the rate of the project, it will ask you a series of (IMO) unnecessary and confusing quesitons. That should be replaced with "are you SURE you want to convert this project and all it's media to the target rate of X?" Y/N.
First video won't play for me, but honestly--There's nothing obscure about the idea that if you play back a 44.1 recording at 48 the pitch goes up.....and visa versa....it's literally the way PCM audio works. AND it's something I've had many an experienced engineer not grasp at all--they can't "get" why the bounce doesn't retain it. I literally had someone email me last week about it--they had some power outage during a live show and ended up in a crazy in between sample rate (somehow?) and need to mix the tune to it's original pitch. He'd adjusted the variSpeed in his app to achieve it and couldn't understand why the bounce never sounded the same.
Sampel rate is functionally not much different than tape speed. In more ways than one. But, why that wasn't as apparent, is that no one sped up the tape (which we ALL did at some point) and then printed the mix back TO the sped up multitrack tape....it went to another standard speed 2 track tape. Had they printed it back to the same multitrack tape, they'd be in this position.
If the pool in Cubase says the files are 44.1, they are. Whether they were tracked at 44.1 or resampled by mistake to 44.1, they're NOW 44.1....and should be played back at 44.1 if accurate playback is the goal. If you just LIKE how it sounds to playback the 44 files at 48, you need to take the ANALOG output of that DAC you're listening to....into the ANALOG input of a second recorder running WTFever sample rate....and record it while the "tape" is sped up.
popmann
RGO Senior Member Longwinded Waxer of Gear Poetic/He Who Is Obtuse/Resident Underachiever
Also note that there IS a way in Cubase to play the 44 files back at the "48 speed" and digitally bounce them. It involves enabling musical mode, Elastic "tape mode" on ALL the tracks and changing the project tempo until you get the speed/pitch you desire. All without changing the rate of the session....that should actually digitally bounce to 44.1 (or whatever destination rate you choose) sped up and pitched up....but, honestly--if you are going to try THAT, "Save AS" first....because there are 900 ways to ruin the whole project by enabling musical mode with different combinations of track timebases....and given your current understanding of the tech, I wouldn't recommend in unless you literally have no second system to use at all....
If the pool in Cubase says the files are 44.1, they are. Whether they were tracked at 44.1 or resampled by mistake to 44.1, they're NOW 44.1....and should be played back at 44.1 if accurate playback is the goal. If you just LIKE how it sounds to playback the 44 files at 48, you need to take the ANALOG output of that DAC you're listening to....into the ANALOG input of a second recorder running WTFever sample rate....and record it while the "tape" is sped up.
The sped up tape is a good analogy. I simply must have recorded it at 44.1, since I didn't resample.
The tracks were us just trying to get a guide vocal and get a tempo that worked for the song for the recording session to come. I emailed the guy and asked him to listen to the sped up version. It's his call.
I've been down the the elastic audio rabbit hole before. Thankfully this is just a setup session, though it's been fruitful to hear the speeded up audio.
Thanks for explaining this. It's always something new...
Frank, which version is "right?" I kinda like the one that sounds like James Taylor better. (Higher pitch and tempo.)
Me too Don. I hope he chooses to capo up and sing higher. It was one of those "happy accidents", but it's up to the client. Always kind of weird when you have strong feeling of what is better, but have to wait and see what they think.
Has happened to me before in PT dragging a 44.1khz file into a 48 session or maybe it was the other way around. As far as why the pitch and tempo change IDK. Sorry I'm not more help, just wanted to confirm your DAW is working like mine ðŸ¤
That you guys like the sped up pitched version better i guess might be somehow similar to symphonic orchestras "tuning war", where some orchestra leaders let their orchestras tune higher because they like the way the strings sound better (more brilliant, less dark). For a while there have been quite high tunings and some kind of competition... Also, sped up mixes can sound more precise rhythm-wise. The rhythmic "errors" minimize in absolute values, obviously....
