Samplerates. Is bigger better?
- Thread starter joeritson
- Start date
AudioGeekZine
arsehole know-it-all
Alright, my take on this.
It sounds better so it's worth doing.
Why limit yourself to CD quality? Downloads are overtaking cd sales so you can release higher quality formats such as FLAC and keep the mastered tracks at 24/96KHz. The people that care about audio quality will appreciate that. Everyone else can get the mp3 and suck a dick.
It sounds better so it's worth doing.
Why limit yourself to CD quality? Downloads are overtaking cd sales so you can release higher quality formats such as FLAC and keep the mastered tracks at 24/96KHz. The people that care about audio quality will appreciate that. Everyone else can get the mp3 and suck a dick.
Hmm, interesting, I hadn't thought of it that way, but I see you're point - it's like when looking at a video of a wheel on a car traveling at high speed, it seems like the spokes of the wheel are going backwards, because really the frame rate of the camera recording it can't keep up. I wonder at what frequencies this becomes noticeable with audio though! (cuz if it's ~16k or above, I really don't care, and will simply low-pass to eliminate!)
Ermz
¯\(°_o)/¯
Actually I think that may be a better analogy to describe aliasing! Think of this as 'approaching aliasing'. Not quite bad enough to misrepresent waveforms, but bad enough to diminish their detail significantly.
In this case it's more like filming a passing car where the camera only runs at 30fps, so as the car speeds by at 30mph it looks defined. But when the car does a pass at 100mph it looks like a blur because there aren't enough frames in the shot to render its detail adequately.
Ah right, thank you (and it actually was during a description of aliasing that my Audio Arts Production teacher used that analogy - whoops
)
wishtheend
clip the apex
CD is really becoming a dead format - so I do think it's time for digital downloads to be provided in higher bit/samplerates.
Kazrog
Kazrog, Inc.
I did some tests recently of some in-the-box softsynth stuff at 44.1 vs. 88.2 vs. 96k, and I feel that the difference is noticeable but very subtle, and very much material dependent. If I was doing quieter music with more spacey delays and reverb tails, then the difference might matter to me a little bit, but it's not a big difference.
The major pain for me is that going up to 88.2 made me feel as though I had stepped back about 10 years in terms of processing power. I'd rather be able to run more plugins and synths, it's worth the almost negligible difference in quality to have things run way more efficiently.
The major pain for me is that going up to 88.2 made me feel as though I had stepped back about 10 years in terms of processing power. I'd rather be able to run more plugins and synths, it's worth the almost negligible difference in quality to have things run way more efficiently.
joeymusicguy
Member
the biggest thing i can offer here is what i've noticed now that i've switched to 44 k
when i recorded in 48 k, my daw / playback headroom seemed invincible. i could add hundreds of layers and hear all of them
but when you bounce to 44 k, its all gonna mush and mesh
if you work in 44 k, you're gonna hear that mush and mesh before you bounce, and you can react according before hand
this might sound like garbage who actually knows this shit by the books, but i dont and im self taught and this is what i've experienced first hand.
when i recorded in 48 k, my daw / playback headroom seemed invincible. i could add hundreds of layers and hear all of them
but when you bounce to 44 k, its all gonna mush and mesh
if you work in 44 k, you're gonna hear that mush and mesh before you bounce, and you can react according before hand
this might sound like garbage who actually knows this shit by the books, but i dont and im self taught and this is what i've experienced first hand.
Gomez
Member
the answer is no. For metal is totally unnecessary. I just finished mixing and mastering an album that was recoded at 88.2 and it was tedious, a waste of cpu and hd space. I often find that mediocre engineers record at high resolution in hope to make up for their lame efforts (or at least, that has been my experiece).
And this is from our resident pure analog obsessor - case closed IMO!
(but as I said, if you've got the HDD and processor power to cope with 88.2 sessions, there's really no reason not to)
Ermz
¯\(°_o)/¯
But if it's mainly analogue workflow, what difference does 88.2kHz make? It's ultimately the client's choice, and to call them mediocre for it feels a bit strange to see in a professional scenario. Those sorts of decisions are for the producer to make - not the mastering engineer.
Incorrect. The reason it looks like a blur is because the shutter of the camera is slow, so in the time it is open the car moves, perhaps by a meter or so. 30fps is more than enough to accurately render a car at 100mph
Ermz
¯\(°_o)/¯
Saculus
Member
- Jan 15, 2008
- 451
- 0
- 16
One would probably want to use a higher sample rate when running a hybrid system of great analog and digital. Like recording to 2" and bouncing to digital for editing. Then mixing the stems on a large format console or a analog summing bus with real analog compression chain. Beyond that more than 44.1k is probably a waste due to loss from reducing sample rate for master and bandwidth loss for processing.
I know, I just meant that since he's the analog guy, he's probably incredibly snooty about sound quality, so if 44.1 is good enough for him, that's pretty telling! (nothing but love Gomez
)Your headroam has nothing to do with your sample rate, only way to get more headroom is to increase your bitrate
oscillat0r
beyond silent black
+1
In theory, yes 96k (or 88.2k) has advantages but there are so many things that yield much better returns for your time/energy/hard disk space IMO.
I haven't thought this statement through 100% yet but it goes something like this...Even though recording/production/playback quality has gotten better every single year, the percentage of new releases that I count as 'favorites' has not.
-j
MetalMiller
¯\(°_o)/¯
I wonder that none talks about SACD and DTS-(HD). Both work with 24bit and 192khz or 96khz on DTS 24/96. Imho the option to record at higher samplerates are just for this kind of formats. In this case it makes sense that loudspeakers (amp and cd-player as well) can reproduce higher frequencies than 22khz, cause there are no LPF at 22k in the converter. Otherwise i think i now understand why some engineers cut at 18-19k because a good/exact eq certainly sounds better (maybe less harsh or something) than a build in LPF in few D/A converters in standard 16bit cd-players. Just a thought...
I stand corrected!
Work that into the analogy whichever you see fit.
Haha, yeah dude, it makes your analogy even more accurate, no?
Harry Hughes
âââ¬â¬
- Apr 25, 2009
- 4,353
- 0
- 36
I agree with those that say it's not worth the hard disk space to use big sample rates. I just use 44.1KHz, but I make sure it's in 24 bit instead of 16. The bit rate makes more difference to my ears anyway.
