|
Post by geoff738 on Dec 4, 2019 23:32:31 GMT -6
Why in the world would you only do 16 bit? Yeah, that’s a lot of headroom not being used. Cheers, Geoff
|
|
|
Post by popmann on Dec 5, 2019 12:25:47 GMT -6
And especially given that he's sampling 24bit...and only writing 16 of them--to do that you'd need to set up an input channel chain with a dither...and ideally a limiter with 16bit dither to pull the 24bit off the ADC and do that DSP before it hits disc--but, all that seems like a seriously weird thing to do when you could simply flip the project to 24bit and only incur a TINY disc penalty.
|
|
|
Post by EmRR on Dec 5, 2019 14:28:03 GMT -6
yeah, come to think of it, we all went 20 bit, what, 21 years ago?
|
|
|
Post by chessparov on Dec 5, 2019 18:33:01 GMT -6
IIRC my six track Vestax HDR6 IS 20 bit. Chris
|
|
|
Post by MorEQsThanAnswers on Nov 26, 2020 12:23:10 GMT -6
Love this philosophy! Record 88.2/Mix 44.1.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 27, 2020 0:20:14 GMT -6
Everything is 44.1khz. I record in 24-bit 44.1 khz, mix it 64-bit float 44.1 khz with a 24-bit dither on the master track that gets sent to the driver, mix down to 24-bit 44.1khz, master to 16-bit 44.1 khz.
|
|
|
Post by drumsound on Nov 27, 2020 2:58:26 GMT -6
24/88.2
|
|
|
Post by Ward on Nov 27, 2020 7:00:53 GMT -6
Everything is 44.1khz. I record in 24-bit 44.1 khz, mix it 64-bit float 44.1 khz with a 24-bit dither on the master track that gets sent to the driver, mix down to 24-bit 44.1khz, master to 16-bit 44.1 khz. Remember back in 1981-1982 when radio started playing CDs? And it was the new super ultra high fidelity 16 bit 44.1Khz digitl format. And we were warned not to turn it up too loud or it might fry our car stereos and blow our speakers! Did I listen? No, of course not. Clearly there's danger in 44.1KHz. Or not. See... how much of what we're told with bit rates and sampling rates is real and how much is hooey? 24 bit has, what, 3 times the dynamic range of 16 bit? But we're still running on pulse code modulation and there's a fair bit or filling in the bits. And other buzzwords . . . that get used by super gear salespeople on their way to explaining the science behind it all. is 32 bit float better than 24 bit, is 64bit that far ahead? Will we use the Sony "least significant bit" method of culling the weak bits out, like Norm professed beer would kill off the weakest brain cells in the episode of Cheers back in the early 80s? Not long after, There was an ancient platform that first brought out 20bit, was it Sonic something? supposedly 3x the dynamic range of 16 bit, but we knew that was false and every mastering house that used it still down-sampled and down-bitted to 16/44.1 What's the best sampling rate? That one's been raging for 20 years or more. What works best for accurate reproduction? What works best for plugins? Anyone have an issue with 48K? How big a pipe do you need going into mixing and mastering before you/they squish everything down to 192bit per second at 44.1 for an MP3 that will probably get further smooshed by some streaming service? It's early and I only have one coffee in me. Bless me father, for I have sinned.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 27, 2020 12:17:39 GMT -6
Everything is 44.1khz. I record in 24-bit 44.1 khz, mix it 64-bit float 44.1 khz with a 24-bit dither on the master track that gets sent to the driver, mix down to 24-bit 44.1khz, master to 16-bit 44.1 khz. Remember back in 1981-1982 when radio started playing CDs? And it was the new super ultra high fidelity 16 bit 44.1Khz digitl format. And we were warned not to turn it up too loud or it might fry our car stereos and blow our speakers! Did I listen? No, of course not. Clearly there's danger in 44.1KHz. Or not. See... how much of what we're told with bit rates and sampling rates is real and how much is hooey? 24 bit has, what, 3 times the dynamic range of 16 bit? But we're still running on pulse code modulation and there's a fair bit or filling in the bits. And other buzzwords . . . that get used by super gear salespeople on their way to explaining the science behind it all. is 32 bit float better than 24 bit, is 64bit that far ahead? Will we use the Sony "least significant bit" method of culling the weak bits out, like Norm professed beer would kill off the weakest brain cells in the episode of Cheers back in the early 80s? Not long after, There was an ancient platform that first brought out 20bit, was it Sonic something? supposedly 3x the dynamic range of 16 bit, but we knew that was false and every mastering house that used it still down-sampled and down-bitted to 16/44.1 What's the best sampling rate? That one's been raging for 20 years or more. What works best for accurate reproduction? What works best for plugins? Anyone have an issue with 48K? How big a pipe do you need going into mixing and mastering before you/they squish everything down to 192bit per second at 44.1 for an MP3 that will probably get further smooshed by some streaming service? It's early and I only have one coffee in me. Bless me father, for I have sinned. Sample rate doesn’t matter if you use those non-linear processing plugins with good anti aliasing and oversampling algorithms. If you don’t, there’s no free lunch. 44.1/48 kHz will alias and with higher sample rate the inaudible frequencies will cause imd in the audible band from non-linear processing. bit-depth, all of the major daws (Pro Tools HDX/current Native, Cubase 11 Pro, Reaper, Logic) use floating point processing now but don’t dither automatically when rendering to fixed point or monitoring. You need a dither. I just use the Goodhertz Good Dither on high with no noise shaping. Cool developer. PSP X-dither is cool too but more expensive.
|
|
|
Post by mrholmes on Nov 27, 2020 22:19:57 GMT -6
I know everyone has an opinion on this and some are adamant that their preference is the best. From reading countless threads over the last couple decades, it becomes almost a religion. With good converters, I don't think many of us could tell the difference between 48/24 and 96/24. You can disagree, but I think moving the snare mic half an inch is going to make more difference to a mix. Converters are so good at this point that is is ridiculous to get hung up on it. Here's a couple thoughts: 1) If I was doing classical, I would probably opt for 96/24. Why? Because it's often sparse with limited track count. 2) If I was mixing OTB, I "might" think about 96/24, but in the stuff I produce, I wouldn't think anyone could hear the difference. 3) I mix ITB and 44.1/24 or 48/24 are what I use and my mixes do not suffer from the sample rate. If the mix sucks, I know where to look. It's not the gear. 4) Try mixing a 150-200 track project ITB with 96/24. Impossible without a ton of bouncing and rendering in place. That would be a deal breaker for me right there. 5) No one, I mean no one, cares about this stuff except for us geeks. Is the song good? That's all anyone really cares about. No one cares how it was recorded... except us. Lots of great songs were recorded in less than optimal conditions and still made a huge impact. Not to forget to mention tracks were people invested big money in big studios but the mix sucks - the song is a world hit nonetheless. Just today I listend to a song by young famous British artist. I thought OMG... even I could do a better mix on my laptop.... That leaves the question open if some producers don't worry too much as long as the song is transporting the vibe/feel?
|
|
|
Post by Bat Lanyard on Nov 27, 2020 23:25:54 GMT -6
That leaves the question open if some producers don't worry too much as long as the song is transporting the vibe/feel? IMHO, that's 99.8% of "it". Edit: to stay on topic, I record at 44.1, 32-float. It does what it needs to do.
|
|
|
Post by Ward on Nov 28, 2020 10:08:56 GMT -6
Ok, back to To Mega... @tomegatherion: Sample rate doesn’t matter if you use those non-linear processing plugins with good anti aliasing and oversampling algorithms. If you don’t, there’s no free lunch. 44.1/48 kHz will alias and with higher sample rate the inaudible frequencies will cause imd in the audible band from non-linear processing. bit-depth, all of the major daws (Pro Tools HDX/current Native, Cubase 11 Pro, Reaper, Logic) use floating point processing now but don’t dither automatically when rendering to fixed point or monitoring. You need a dither. I just use the Goodhertz Good Dither on high with no noise shaping. Cool developer. PSP X-dither is cool too but more expensive. Great reply. Thanks so much.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 28, 2020 12:42:28 GMT -6
Ok, back to To Mega... @tomegatherion : Sample rate doesn’t matter if you use those non-linear processing plugins with good anti aliasing and oversampling algorithms. If you don’t, there’s no free lunch. 44.1/48 kHz will alias and with higher sample rate the inaudible frequencies will cause imd in the audible band from non-linear processing. bit-depth, all of the major daws (Pro Tools HDX/current Native, Cubase 11 Pro, Reaper, Logic) use floating point processing now but don’t dither automatically when rendering to fixed point or monitoring. You need a dither. I just use the Goodhertz Good Dither on high with no noise shaping. Cool developer. PSP X-dither is cool too but more expensive. Great reply. Thanks so much. I’m not sure linear phase oversampling and anti-aliasing cuts are 100% transparent but High sampling rates won’t save you from non-linear plugins without good anti alias filters anyway even when we have awesome 300 ARM core Apple computers 10 years in the future: www.gearslutz.com/board/mastering-forum/968641-some-thoughts-quot-high-resolution-quot-audio-processinHigher sample rates are better for warping and stretching too. It keeps the highs more natural. Think dialogue in movies but so many movies and tv now have awful mixes (Thx bass is annoying and I’m sure few of them use working compression On the overall Audio mix) and the modern imd/aliased plastic “sheen” distortion. For bad plugs vs good plugs, here’s a quick mix of a dry kick and guitar bus I tried to match the tone of in the 50% defective PSP infinistrip at 44.1khz. Keep in mind that only the 12-bit sampler ADC, 1970s Nevey pre (nothing special and 2d but it works), FET compressor, and the EQs (not the cramped high end) work well (if it all), in the plugin: we.tl/t-pmBpmBETV6we.tl/t-PKpDTij2faI need to change the guitar mud cut to a dynamic eq and the kick in the real deal plugins mix is overcompressed but the difference is real. I’ll attach the kick chains later in an edit when I get on my computer.
|
|
|
Post by MorEQsThanAnswers on Nov 28, 2020 13:20:45 GMT -6
Higher sample rates are better for warping and stretching too. It keeps the highs more natural. In a huge way too. I have a grid-obsessed pop client that likes recording live drums and this was a game changing realization
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Nov 28, 2020 14:22:19 GMT -6
Here are the kick chains. Infinistrip feels like I'm using the dope but in no way an 1176 Fetpressor with a bunch of Waves plugs. Gross. They should've made it with just the 12-bit ADC -> filters -> EQ -> Fetpressor for Golden Age hip hop drums. Think a PSP version of Voxengo drumformer. The limiters and opto compressor straight up don't work. The recent Fuse VCL-4 proves opto can be done without oversampling yet they failed. Attachments:
|
|
|
Post by Bob Olhsson on Nov 28, 2020 14:30:15 GMT -6
FWIW, back in the early Pro Tools daze, most major label work was recorded at 48k and mastered analog. Today, most major label music is done at 96k while sound for picture is still 48k.
|
|
|
Post by chessparov on Nov 28, 2020 18:05:47 GMT -6
Honestly, I had forgotten who started this Thread. (It was me!)
Wow, the psychological difference of Pre-Pandemic compared to current. Anything from January on, seems like "Dog Years" to me!
Anyway the occasional vocals I expect to put on, from received backing tracks-Mostly Post-Pandemic...
Will probably be 48 KHz, unless there's 44.1 Sampling going on. Chris
|
|
|
Post by MorEQsThanAnswers on Dec 1, 2020 23:35:21 GMT -6
Came across this, thought about this thread, figured why not pass it along
“Finally, I recorded the whole session in 24-bit at 44.1kHz, because most of my sample library sounds best at 44.1. If I try to convert the samples they don’t sound as punchy. But I do mix to 24-bit/88.2kHz, to an Alesis Masterlink ML9600. If someone wants to master digitally, 88.2 converts much better to 44.1. The record company creates all its different formats from my Masterlink master.”
-Mark Endert about “Makes Me Wonder”
|
|
|
Post by johneppstein on Dec 4, 2020 19:45:40 GMT -6
24/96 but other studios do bitch that I am wasting space. I have always figured that like 4k I can always down sample to lower but keep a copy of higher resolution just in case. Who are these "other studios"? If they're not known professional studios I'd ignore their "input".
96 sounds better. Downsampling incurs a hit to your sound quality.
|
|
|
Post by johneppstein on Dec 4, 2020 19:50:28 GMT -6
Physical CD's. Believe it or not, they still get made. But yeah, that's a good point. I'd rather work at 48 and downsample for CD than upsample for video. I'd rather work at 96 and downsample if necessary. Really shouldn't be necessary these days. I someone complains about "wasting space" these days they're probably not professional.
Upsampling sucks, BTW. It doesn't get you back what you lost by initially recoreding at too low a rate.
|
|
|
Post by johneppstein on Dec 4, 2020 20:13:03 GMT -6
Honestly, I had forgotten who started this Thread. (It was me!) Wow, the psychological difference of Pre-Pandemic compared to current. Anything from January on, seems like "Dog Years" to me! Anyway the occasional vocals I expect to put on, from received backing tracks-Mostly Post-Pandemic... Will probably be 48 KHz, unless there's 44.1 Sampling going on. Chris I've been at 96K long before the pandemic even started.
|
|
|
Post by chessparov on Dec 4, 2020 20:33:08 GMT -6
Interesting John E. Really not enough experience/judgment here, to fully evaluate (meaning me!) all this. Shoot even 16/44.1 sounds good to me, IF it was recorded well. Otherwise a jagged pill, to swallow! Chris
|
|
|
Post by jeremygillespie on Dec 4, 2020 21:17:55 GMT -6
Higher sample rates are better for warping and stretching too. It keeps the highs more natural. In a huge way too. I have a grid-obsessed pop client that likes recording live drums and this was a game changing realization I’d encourage everybody to not use the stretch function for editing. Unless something is so far off (I mean really off) time - wise, it’s always better sounding to just move stuff around and drag out the edit points as then fade. You can hear the top end on stretches and it’s pretty awful, especially with cymbals, pianos, and the top end of acoustic guitars.
|
|
|
Post by drbill on Dec 5, 2020 10:25:23 GMT -6
For me, it's incredibly simple to get past all the issues with aliasing, SRC, etc. I just try to use hardware as much as possible. Makes a palpable difference for me....
|
|
|
Post by Bob Olhsson on Dec 5, 2020 15:34:19 GMT -6
All my testing included comparing a final 44.1 down-sample. Even MP3s sounded better when the original recording and all the processing was done at 96k. It meant buying a new computer, so my prejudice was against it.
|
|