Dubstepforum.com

at the moment i have my daw set to 24bit/44.1khz, as most of my samples are at 44.1khz so it'd avoid converting them.

what i was wondering, was if there would be any notieceable benefit to switching to higher settings? the thing is, i'm processing samples over and over, with long effects chains. surely each time i bounce it must lose a bit of quality? so maybe if i did the producing at 32 bit/192khz, then bounced the master down to 24/44.1, it may sound better than if i'd done it all at 24/44.1?

i know it'd eat up my hard drive and system resources faster, which is why i haven't tried it. just was wondering if any of you had used higher settings and heard any improvement to your sound?

Default sample rate. I tried to switch it once, got all fucked up. Things off beat etc.~
Bit depth: 32bit float for samples. 24 for the mastering house.

zaberman whussup with the montreal airport stuff!?!?!

on dat pierre elliot trudeau swagger?

op, ive been reading extensively on that subject and while there may be some differences between samplerates.. they are so unsignificant to the average ear that u shouldn't worry about them until it's the last thing ud have to worry for (like when ur in a treated room with the perfectly tuned gear and playback system uve been dreaming of, and that u know inside out)...

basically 44,1 is more then twice the young, healthy, undamaged ear of a human being.. (20k)... it thus respects the nyquist criteria which stipulates that for the highest frequency to be represented in a system, that system should have a rate that is at least twice that frequency (20k x 2 = 40k; then 40k<44,1k !)

i was still wondering about the other rates... and bob katz wrote about it... there seem to be differences and benefits are higher rates but these are the slightest, unnoticeable things ever... ud have to produce a really really really good mix to actually justify such small gains imo... there was a pretty cool thread about that a while ago.. try to find it...

to double that statement id add that most mixing engineers for pop releases seem to work in 44,1 ..

now for the bitdepth, theres numerous threads about that. id aim for the highest my system allows me while still keeping in mind that its probably goint to have to be lowered on the way out (then comes dithering and such things...)

good post, these questions, while they seem very abstract, are touching the most crucial concepts that support digital production imo

btw im still learning so if anything in that post was unaccurate or plain wrong, lemme know.

24
44.1

Ldizzy - thanks for the info. i've spent the past few hours researching the subject too and from what i can see you're totally right. one thing i picked up on is that it could actually sound worse if you'd done everything in 96khz, then changed the sample rate yourself. apparently its best to leave any sample rate conversion to a mastering engineer! so i'll stick with 44.1khz as that seems to be the standard for music. film and studio stuff is obviously higher but i just want to make dubstep tunes!

and as for bit depth, i've found out that cubase does internal processing at 32bit float anyhow. one thing i have learnt is that if you set your project to 32bit float, then it won't clip at 0db. i didn't realise that. and apparently its not any heavier on the cpu, it can actually be easier than 24bit if you're using a 32bit os! (source: steinberg mixing video) so it may be an idea to send in 32bit wav files to a mastering engineer, in fact theres a mixing engineer who's posted on this forum earlier, i'll ask!

Ldizzy wrote:zaberman whussup with the montreal airport stuff!?!?!

on dat pierre elliot trudeau swagger?

What are you on about?

Here's an interesting article on a study on sample rates http://mixonline.com/recording/mixing/a ... _sampling/

As for what I use: 44.1 and 24 bit.

I render everything in 32-bit float and 320 kbps. And I think 44100.

Zaberman : sorry for some shady reason my browser now permanently displays adds from the montreal airport, where i was 5 days ago... its really weird i took some screenshots .. i thought u were advertizing for them and was like wtf

OP : 32 bit float is available for most daws.. thing is things are going to have to be brought back to a lower rate at some point... thats exactly what i was referring to in my first post. internally, if u have a good db hygiene u shouldn't avoid systems with a floating point, they're tits.

yeah that's fine what you're using. 32 bit isn't good as you'd probably have to dither it when going down to 24 bit when mastering which can cause some muddiness in the mix

Writing at 24/44.1. Mastering, it depends- some plugs, especially eq's, respond better at 88.2 or 96, depending on the source audio.

32 bit is great as an internal working thing for daw's, but the benefits of it are nonexistent at a bounce level.

Recording:

88.2/24

Post:
48/24 or 15 depending on client

Music
44.1/32float

Sharmaji wrote:Writing at 24/44.1. Mastering, it depends- some plugs, especially eq's, respond better at 88.2 or 96, depending on the source audio.

interesting

what type of symptoms should i lean my ear towards??? what are the typical gains, even so slight they might be, of the 88.2 s.rate?!?!

Sharmaji wrote: 32 bit is great as an internal working thing for daw's, but the benefits of it are nonexistent at a bounce level.

so well put. i think the novel i posted above could be summarized to that

one thing to consider is the file sizes. a 1gb project file could quite easily turn into a 4b project file at higher bit depth and sample rate settings.

thor beatz - why do you use all those different settings? just seems theres no consistency and you might be converting files a lot.

Given how cheap and easy it is to get large amounts of quality data storage these days, file size is a non issue IMO.

VirtualMark wrote:one thing to consider is the file sizes. a 1gb project file could quite easily turn into a 4b project file at higher bit depth and sample rate settings.

thor beatz - why do you use all those different settings? just seems theres no consistency and you might be converting files a lot.

Depends on your DAW if you need to convert (I dont need to convert it in FL or AL- only in PT you need to mind the project settings). In any case; recording at 88.2 because I'll be editing them extensively (plus some of the recordings might end up in a library.) Also arguable its better to have 88.2 than 96 due to conversion reasons

48 because it's standard for post (just the way it is) [edit] post is short for post-production(videos)

44.1 because thats the bounce rate most likely to play on all device (consumer)s. When sending to mastering 96 btw (depends on the engineer tho)

With 96KHz, a signal needs to get to -120db by the time it reaches 48KHz instead of only 22.05KHz, meaning the filter/slope can be much gentler. This, although it’s not been able to clearly be determined in various listening tests, could be the reason why 96KHz can sometimes sound slightly better than 44.1KHz; a gentler slope has the potential to cause less “ripple” in the audible band.

I have the space - and some clients are keen on high rates. Who cares, i'll just buy another HD a little sooner I'm a bit of a freak with these things. Over-sampling etc - arguable the difference is not audible after aliasing elimination. Same with dither. Maybe in the end it adds up. IF so i rather be on the safe side.

Oddly enough, my soundcard only supports 16 bit 48Khz, not 44.1khz


Not that I even have much going on past 18khz in my tunes.... I suppose I need a new soundcard.