same happened with my CTC-1 story while using 48k

 

Only reason to stay in 44.1K that is:
- Use absolutely no saturation, or saturate only lowpassed sounds in the DAW.
- Don't compress with fast settings, ITB.
- Don't EQ around Nyquist.
Seeams like too much "don't" for me. But i deffinetly can live without fake ITB saturation.

Please, let me tell you my little story about sample rates.

I was defender of 44.1k before i make a legit (i mean real song not simplified stuff) project test vs 48k.
I notice how became my highs in the 48k mix.
I switch to 48k after 10 blind tests (few days delay between them) + small focus group (two colleagues sound engineer + 2 musicians + 2 "regular humans").
When i can afford 96k (i work hybrid, my AD\DA converters that i have than, works in daisy-chain at 48K max) i make another real-work project for testing how the sample rate affect the end results.
Same blind tests (off course gain matched with no even 0.1 dB difference bias) - i notice how my highs change from clean to pristine.
Really.
Now i can afford 8 channels of DA conversion to pre-mix on Reaper than send tracks and buses to my analog console at 192 kHz sample rate.
I think few days how to organize my real-work project for this test. I need to avoid any smallest bias or bug.
Most (i mean 95%) of my Acustica plugins and Nebula libs are "leave the chat" and dosen't work correctly at all in 192k.

(i need no advice on this. No Nebula XML tweaks will fix this behavior. Just because people from Acustica dosen't want make support of 192k)

Spoiler

I decide to stay away from AA stuff for this project and use only few really stable plugins. If i use something that can go behave odd on freeze\render i will draw wrong conclusions about results. So i freeze all my pre-mix plugins than real-time print mix on hardware back to DAW. All steps are 96k for 96k project, and all 192k for 192k. All synt's, samplers, fx processing - everything.

I'v open both prints - 96k file + 192k file in 192k Reaper session to avoid downsampling for last one.
96k file will just upsample to 192k session rate.
Plus i SRC both to 44.1k 24 bit with Weiss Saracon (THE best SRC i even heard. Believe me - i try and blind test them all!)

And now it's going down guys.
I can't believe what i'v hear in blind tests!
I'm finally got that sound that i search whole this 17 years.

Transients are clean and fast.
Stereo field, a lot more detailed and wide.
Low mids are round, bass are deep and mids are detailed, not harsh.
I don't hear any "fluffy dusty blurry pillow effect" like before in 44/48k.
Mix have that clarity and separation that i search.

I re-listen over and over, a few weeks this 96k vs 192k and can't believe how easy to spot the 192k blindly.
If i can try 384K i will try. But i can't and i doubt that most computers can handle that load.
So 192k it's a pretty decent "ceiling" to be closer to original analog input.

I will not post anything (files) here guys, by copyright reasons (i compare on my unreleased song)

I urge each of you not to believe a single word I say.
Please follow my path. Do everything meticulously and taking into account all the small details, avoid even minor errors. And with good monitoring, you can hear that no matter what sample rate you work in, you should go higher.

 

Last edited by a moderator: Sep 27, 2023

with reaper you could run it at 96khz for nebula then oversample everything else up to 768k, right? although I wonder if you can do the reverse, not that I know of, where you selectively run some tracks at lower than the project. I'm no expert though, I run at 768k 8 bit.

 

does he make decent songs?

 

Most decently/properly coded plugins used on a track will act better, re: foldback of undesired harmonics, at higher sample rates. I hate engaging oversampling in plugins because it's too unpredictable. Hence 96/192 khz is where I keep my gear clocked. IMO, final mixes once dithered down to general distribution specs always sound sharper and more defined as a result.

this thread seems like click-bait as there are at least 1 BILLION other threads dealing with the same topic...

 

I only use one note so I find 56,320 Hz works perfectly..

 

Less aliasing depends a lot more on higher bitrate.

 

This guy isn't presenting any scientific reason, it's just the opinion of his "ear." The reason why he might prefer 44.1kHz is because, as he says, his equipment handles it better (and some commercial audio players won't even play 48k files).

 

44.1 khz or 88.2 khz ,,,,,,, 48khz or 96 khz ,,,,,,,,, i wont lose my time explain it , go learn some digital audio theory instead of opening threads writing or posting bullshit .

 

Most sample rate I'm using is 48kHz (live performance).
44.1kHz for final mastering (CD-Audio / MP3).
Recording at 96kHz from OBS, mixing at 96kHz from Bitwig Studio 5, or REAPER 6, downsampling to 48kHz for final audio.

 

Last edited: Sep 27, 2023

Yeah, but WHICH note do you use?
It makes a difference.

I swear I won't steal it.

 

I use 44.1 but it identifies as 96.

 

Until this is settled, I will stick with 33 1/3.

 

Reaper use very bad approach for oversampling.
Make a two samples.
One oversampled.
One don't.
Normalize both to exact same value (-23 LUFS-I)
Loop half of the bar and fast flip between them.

You will hear if you got trained ears and decent monitoring.

768K oversampling is bad idea!
A lot of plugins will crash or bug out.

Oversampling is a highly overrated thing. Because somebody spread that, it makes something better by default.
It don't!
OS makes your sound dull or unnatural, and I a million times said why.
Better work in 192K than use 96K with 2x/8x/4x/16x etc oversampling.
Try to compare like i said before on your material, and you will shocked.

 

all this palaver when in the end most music is played back on shit formats on shit devices with shit earpieces and speakers

 

Last edited: Sep 28, 2023

Sample rates, audio formats, cables, it's all astrology for audio nerds.

There's a reason you don't see the dudes people want to sound like arguing endlessly over things no one should really care about, they're busy with their craft. It either sounds good, or it doesn't.

YouTube once recommended me a video of Andrew Scheps talking about mixing on headphones and what he said is the truest truth of them all: "it turns out if you don't know that [you can't mix on headphones], you can just do it"

Edit: phone auto correct lol

 

I look at it like video. If you shoot your footage in a higher resolution and bitrate, say 8K 16bit raw/log, but your final delivery is youtube (4k 8-bit rec709), you have more flexibility in post to push and pull your image before "printing" it in a lower format. It helps fine tune the image and gives a better overall look to everything. Filming in 4K 8 bit would be much harder to get the same results, if you could at all. It also leaves way to re-release your creation in a higher quality one day when there's a platform that supports it (future-proofing). Audio isn't as clear cut, since it's harder to hear than to see sometimes. I've always believed in overkill resolutions and formats. You never know what the future holds, and you might thank yourself one day for being so extra.

Plus I noticed my 96k recording was like removing the rag off the mic. Downsampled to 44.1 and the clarity disappeared. This was a raw project, no plugins. It was subtle but definitely there.

 

I remember dealing with cd audio and mp3 when I was young, like in 2000, using Win98 and Napster
What is the purpose of burning CD's and encoding mp3's in 2023?