Mastering at 24bit or 16bit

[evon]

Converting them down to 16 bit you have to put three 24bit parts together and cut them down to 16 bit(That's the way a converter do it for you). So you get it ''glued'' at two places and cut on three others.
Too many artifacts will always be made by that.

Seems to mt that that should be "two 24 bit parts" then divide by three.

[TRS80]

Give the mastering guy the highest quality file possible- so yeah 24 bits.

He will need those extra bits for headroom and clean processing. Taking it to 16bit 44.1kHz should be the very last step.

[Tone Deft]

32bit at 44.1 Hz, that's my way to go...!

It's also recommended in the Ableton-Manual...!

Since I do it this way, all my music sounds the same after exporting.

that's because you wouldn't hear the difference anyway.

Why, because Live internally works always in 32 bit floating.

that makes no sense whatsoever. 32 bit floating does not mean 32 bit fixed point. it's apples and pears. there's still a conversion to be done from 32 bit floating to 32 bit fixed point.

I use 44.1 because it gives me enough quality...!

agreed.

Need more Quality? Then double it up, to 88.2 , but never go for 48Hz.
This is for DVD only, and you can double this up to for more Quality.

nope. 44.1k for CDs is the oddball out. 48 kHz is the standard for pro products. 88.2k is for stuff going to CDs, same for 176.4kHz.

Never use this for CD, converting from 48 down to 44.1 is a pain and you loose at the end.

agreed.

I don't understand why all the producers still go for 24 bit...!

because AES/EBU aka SPDIF is 24 bit, for professional equipment.

Makes no sense at all. Converting them down to 16 bit you have to put three 24bit parts together and cut them down to 16 bit(That's the way a converter do it for you). So you get it ''glued'' at two places and cut on three others.
Too many artifacts will always be made by that.

that's totally completely incredibly wrong. you just pad it with zeroes.

32 bit just needs one cut and you have two 16bit parts, and less artifacts then you can get at this high quality.

that's also very wrong. you deciminate (opposite of interpolate.)

Everyone still using 24 bit is wrong...!!!

yeah, and the guy that doesn't understand basic math is right?? you need to read up on the topic.

[Tone Deft]

Converting them down to 16 bit you have to put three 24bit parts together and cut them down to 16 bit(That's the way a converter do it for you). So you get it ''glued'' at two places and cut on three others.
Too many artifacts will always be made by that.

Seems to mt that that should be "two 24 bit parts" then divide by three.

no. it's basic math. forget it's binary what would you do with decimal? it would be the same process for binary numbers as it is for decimal numbers.

[davepermen]

tonedepth, just a quick think, i'll reply to the rest afterwards

converting a 32bit float (24e8) to a 32bit fixed point is completely lossless as long as the float is in the range of the fixed point (else clipping would occur). it's not lossless if the float represents a tiny value, there it will be adjusted to the resolution the fixed point delivers, obviously.

so using 32bit floats and then go back to 32bit fixed point is no loss. but it allows for peaks to not get clipped, which is a huge gain.

[Tone Deft]

shit, you're right. it's just moving the decimal point, the numbers scale themselves. ultimately it's all relative. I've been doing stuff like that at work recently, floating point twos compliment binary math, it can be a head trip how the numbers map.

my bad, just waking up here, I saw some errors and went too far.

[davepermen]

easybeasy. that stuff can be hell. i've done only basic work with it (but tons of theory). it can be very messy at times. i'm coming from a graphics programmer background (thus often references to 3d graphics), as they have in essence the same issues, resulting then in wrong pictures instead of wrong sounds. for me, visual stuff is much easier to understand.


but in the end one lesson is learned: highest quality during processing, minimal needed to not hear any errors to store is way to go. so once a track is done, it might be saved as (i go extreme now, bare with me) 320kbits mp3. but you don't want to master it then based on that file

[synnack]

I am not convinced no matter how much theoretical math can be quoted that if you write a song using all 16bit samples, it will somehow sound better if you export it above that.

If my mix sounds great. I export at 16bit and save the disk space. I have never seen or heard of a mastering engineer go "sorry, even though your song was recorded at 16bit, your release would have sounded a lot better if you would have just exported larger files".

This is a lot like a group of guys comparing dick size to a girl who has already orgasmed.

[Tarekith]

It depends on the song, the levels and so on (blah blah blah), but there's definitely times you can clearly hear a difference between 16 and 24bit when mastering. Much more so than say 44.1 to 48 or 96kHz for instance, most people would be surprised at the difference it can make.

I never understand the storage concerns in this day and age. A 1.5TB external drive is less than $100 these days, are people REALLY using that much space they can't used higher bit rate files?

[synnack]

It depends on the song, the levels and so on (blah blah blah), but there's definitely times you can clearly hear a difference between 16 and 24bit when mastering. Much more so than say 44.1 to 48 or 96kHz for instance, most people would be surprised at the difference it can make.

I never understand the storage concerns in this day and age. A 1.5TB external drive is less than $100 these days, are people REALLY using that much space they can't used higher bit rate files?

I, like many people, switched to using a laptop for everything. Using a bunch of external drives defeats the portability argument for doing so. My laptop fills up quick.

Also I've had a long week and I am grumpy in general. (wow that's a pun)

[Tarekith]

I use a laptop for everything too, evne then, still only about $80 to put a 500GB HD in there. I know what you're saying, but sooner or later you gotta plug i tin to recharge, not like you can't have a HD there to quickly swap any files you need. I have another 1.5TB HD hooked up to my wifi netowrk, so I don't even need to plug things in to grab files from around the house. Very handy.

[bridgealantee]

Need more Quality? Then double it up, to 88.2 , but never go for 48Hz.
This is for DVD only, and you can double this up to for more Quality.

I agree you should always double up, but I think you should double 48 to 96.

[evon]

But what does converting the sampling rate have to do with any artifacts?
IMO artifacts are created in converting the resolution. I cant see much artifacts being produced in resampling. I always thought dithering corrected the problems in bit depth.
I could be wrong.

[Tone Deft]

But what does converting the sampling rate have to do with any artifacts?
IMO artifacts are created in converting the resolution. I cant see much artifacts being produced in resampling. I always thought dithering corrected the problems in bit depth.
I could be wrong.

go read up on sample rate conversion. it's a fundamental concept, you've got it all twisted.