Ableton Forum

levimoniz wrote:

H20nly wrote:so Ableton should let you know that when you change a sample to a different sample rate that the sample rate is changing.

Sample rate in Live is converted automatically

so why do you need a popup window or a letter personally signed by Gerhard stating that?

jebus chris...

stop using samples.

then you can make all your music at the same rate without the Ableton staff having to show up with milk and cookies or beer and blow to let you know you changed something.

levimoniz wrote:]As I said, they could have something that lights up or have something in the info box.

This would definitely go a long way in preventing people who import tracks with a diff sample rate saying that Live sounds fucked up in interviews.

a warning box or something might help beginners.

it does bother me though that beginners are getting caught up in this kind of thing instead of concentrating on whats really important. i've seen people get hung up on what type of dither they are using when they've yet to understand the basics of rhythm, tone and basic mixing.

lunabass wrote:

levimoniz wrote:]As I said, they could have something that lights up or have something in the info box.

This would definitely go a long way in preventing people who import tracks with a diff sample rate saying that Live sounds fucked up in interviews.

a warning box or something might help beginners.

it does bother me though that beginners are getting caught up in this kind of thing instead of concentrating on whats really important. i've seen people get hung up on what type of dither they are using when they've yet to understand the basics of rhythm, tone and basic mixing.

^this.

getting a good gain stage going is all HUGE in getting a nice fat sound, for example.

as people learn this stuff they learn of flaws in audio processes, like SRCs. then they freak out over it and then move onto the next new topic. over time you learn the BIG ways to screw up.

it's the 21st century, people have long ago figured out how to record audio. we're pretty spoiled with all this digital gadgetry and recording studio in a box stuff.



I had an idea on a Live test. make a huge 20+ track set, render it out. then disable all but one track and render each one out one at a time. does the result of the big huge track cancel when played with the smaller stems inverted and summed?

levimoniz wrote:

Palmer Eldritch wrote:Summing and SRC are technically totally different things. (and SRC is much more complex than pure summing).
But regardless of that please welcome to the more rational side

He said that he suspects the summing and src algorithms to be based off the same coding; is there any validity to this or what's the deal?

No, this is complete nonsense. But if he meant "of the same coders" then he might be right

levimoniz wrote:

Palmer Eldritch wrote:Summing and SRC are technically totally different things. (and SRC is much more complex than pure summing).
But regardless of that please welcome to the more rational side

He said that he suspects the summing and src algorithms to be based off the same coding; is there any validity to this or what's the deal?

I doubt it but this is all speculation.

what, you're a programmer now too? what does "based off the same coding even mean?"

levimoniz wrote:Does anybody read all the way through posts anymore? Or do you just skim until you find something you feel like picking out and waving like a You Are Wrong flag?

i just skim, i really dont have the desire or time to read 20 odd pages on sound quality... 900 posts in 10 years seems about right to me.

anyway i'll leave you guys to it. i need to get this track mastered.

levimoniz wrote:This is exactly what I mean, like WHY are you asking me to explain that? WHY are you talking in this thread if you don't know what is even being discussed?

AGENDA?

Really 24 pages
Unbelievable, but in fact I have read them all

Good night

levimoniz wrote:Read what NF wrote. It's like in the first two god damn sentences

hey man, now you're trolling with other peoples' posts that you don't understand. this is pretty nuts. these topics are hard enough to keep sane as it is.

I need to put Live on a computer without an internet connection.

Levi - you should hit up Tarekith's site and read up on his tutorials. over the years I've also found amazing essays on the Sound On Sound web site.

always use Hi-Q mode, always use the same sample rate, learn to make good gain stages and IMO good sounding songs start with good sounding recordings.

edit -
I'll vouch that lunabass knows his shit, he's not affiliated with Ableton nor does he have an agenda. he's a smart guy that knows his shit, if you're lucky he can teach you some stuff.


this is ALL speculation by all of us, but if you want an answer, any answer...
SRCs are an important bit of code to write, same as a gain stage. you want them to be the best bits of code you can write. IMO putting them into one operation would hurt both. they're also quite different operations. SRCs involve interpolation, decimation and an FIR filter with a coefficient generator (or a lookup table.) a gain stage would handle the 32 bit floating point adds. now that I think about it out loud maybe you could do it with the coefficient generator but that might be tricky to do in real time.

the post you're referring to holds water but like mine it's speculation. we'll really never know.

stay hungry, you'll get this stuff down. do go research this stuff on your own and don't get it all from here.

hth

NF wrote:...so I assume that the summing engine is based on nearly the same way of calculation....

and

NF wrote:... but I could imagine that the DSP summing core is based on nearly the same algorithm/part of code....

lol, to clear that I quote myself

lives audio processing core is coded mostly in C/assembler, remaining of some classes/libraries, consisting of some algorithms, called methods.
my assumption was that the engine, calculating the main summing is nearly the same or has parts from the SRC algorithm.

As Palmer Eldritch said, it's not. I accept that and will learn more in the near future. At the moment I'm only a working as a student in image processing so I was not able to work with audio in code. I know IIR filters and the idea of SRC- but not in realtime. And a summing bus in a DAW...I don't want to make u asleep.

but I am, its 2am here.

Gute Nacht Palmer X-D

Actually guys, the SRC and summing are based of the same bits of code...

1's and 0's.

Mixing audio is just basic arithmetic. Addition, multiplication and logarithms.

SRC is a complex DSP algorithm based on a mathematical process called convolution - it won't use any of the same code as the mixing bus.

The best SRC simulates reconstructing the original waveform by applying an oversampled reconstruction filter and then re-samples the output from that at the new bit rate.the sample rate of the reconstruction filter needs to be the lowest common multiple of the original and new sample rates.

Another approach is to use some kind of interpolation, but this is only ever a guess and will inevitably add noise.

No idea how Ableton do it.

All SRC has to be done before the audio is summed.

my 2p worth

crumhorn wrote: All SRC has to be done before the audio is summed.

yep.