Samplitude user disses Live as 'shit for mixing in' at GS

[kb420]

im done with the thread. i've doubled my total posts just in this thread

OK! Don't let the doorknob hit you on the ass on your way out!!!!!!!

i wont...

have fun at the football game

Always do!




SEASON'S TICKETS BEEEEEEEEEEEEEEEEEEEAAAAAAAAAAAAAAAAAAAAATTTTTTTTTTTTTTTTTTTTCHHHHHHHHHHHHHHHHH!!!!!

[Tone Deft]

just read the midi fact sheet from ableton and tell me that its no joke

ok, just read it...

page 1 - intro, definitions of recording and playback.
page 2 - definition of playthrough, jitter and latency
page 3 - how Live handles those issues with timestamps. they favor low jitter over low latency.
page 4 - a guide on jitter and latency issues that a DAW cannot compensate for or deal with.
page 5 - a guide on how they test their jitter at Ableton headquarters.
page 6 - the results of their tests on other bits of hardware
page 7 - tips on low latency, mostly really n00b stuff.
page 8 - a short summary.


so what's your problem with this??

[kb420]

All that big man talk diskowipe did, but as soon as one of the Abes shows up he disappears.

What a wimp!!!!!

[ethios4]

This is the part from the audio fact sheet he diskowipe was talking about.

Please note that, while 64-bit summing is applied to each single mix point, Live’s internal processing is still done at 32-bit. Thus, signals that are mixed across multiple summing points may still result in an extremely small amount of signal degradation. This combination of 64-bit summing within a 32-bit architecture strikes an ideal balance between audio quality and CPU/memory consumption.

He's making a big deal about it, but there is no mention of how much distortion we are talking about, or of how other DAWs handle this issue. I'm guessing the distortion levels are on the -160 dB scale, which would not account for 'radical' differences in DAW sounds.

Oh well.

[3phase]

i know summing is not neutral in any application. thats why programs sound different.

False. Using the exact same settings results in identical audio files. Digital audio is pure math as in 1+1. Sorry to disappoint you.

I really belive that.. its logical.. And except theese Harrison Mixbus thing mixing in the digital domain was allways seen as a plain summing of signals...

However there are differences when mixing in differt daw´s.. or digital mixing desks..
Some very drastical. I really belive in that.. cant say if its worse than logic.. but mixing in prottols hd defenetly sounds better.. ok..its 48 bit fixed.. does this make a difference?

why is that when 2+2 is allways 4?


And i recently had again the problem of listening to a mix via the apple quicktime player..
importing the mix in live to hear some sound fx over it n sync...and..
Sounded not the same as played from the quicktime player anymore..
Projekt speed on the file..warp in repitch..

shouldnt this be an acoustc transparent setting? or is only beat warp on song tempo transparent?


And other question..Are all 32bit floating audio engines the same? because its just math they should be...

but why you can overload one 32 bit floating point system easy while on others like ableton you can record with +30 db and turn it down on the master and everything is allwright?

I wonder why some companys choose to have a clipping risk while others choose to dont have one because this luxury comes without an extra price.
Does that come really without a price? Or is the price ignored because its payed down in the -100db area?


So i really would like to know why some companys are so stupid to do the "easy to clip" design when there is no benefit..

Is it saving cpu cycles and therefore a remain from old computertimes and you only find this still with the old digital audio companys?

[henke]

Land of Confusion

That old Monolake track was created using digital synthesizers with 12 bit converters. Mixed with an analog console. Recorded to a 16 bit DAT tape. 'Mastered' with a horrible TC Finalizer 'mastering tool'.

It is one of the tracks on my list that cry for a remastering.

The new Monolake tracks were created using digitals synthesizers with 32bit fp. resolution and using field recordings, 24 bit 96k. Plus Altiverb IRs. No compression.
Most tracks were recorded 96k 24bit. Album was mastered using a lot of analogue EQs.

I had the chance to listen to the new album over 50k$ PMC monitors. It sounded AMZING to my ears. There was no lack at all in the low end....

So, what is the conclusion here?

The old track is the result of a different working process, but it does not sound as good as it could from the artists perspective.

Both albums are the result of analoge and digital processing, but there are good rational reasons, why the new ones are sounding much better from a technical perspective.

And, the tracks on the Cinemascope CD were mastered using Live 2 as my DAW!!!!!! How does this make sense now???????

It doesn't !

Live does not sound. Music does.

Robert

[kb420]

This is the part from the audio fact sheet he diskowipe was talking about.

Please note that, while 64-bit summing is applied to each single mix point, Live’s internal processing is still done at 32-bit. Thus, signals that are mixed across multiple summing points may still result in an extremely small amount of signal degradation. This combination of 64-bit summing within a 32-bit architecture strikes an ideal balance between audio quality and CPU/memory consumption.

He's making a big deal about it, but there is no mention of how much distortion we are talking about, or of how other DAWs handle this issue. I'm guessing the distortion levels are on the -160 dB scale, which would not account for 'radical' differences in DAW sounds.

Oh well.

That may well be what he was referring to, but he totally misunderstood what that means. He said that because summing is non neutral, all programs sound different. He didn't understand or want to admit that if all of the settings are the same, you will get the same results in different programs.

But we already went through all of that, and yet he failed to prove anything.

[henke]

oops, logged in from wrong account. but yes, it is me!!
Robert

[Tone Deft]

3phase you're well intended but you don't know what you're talking about. that's fine but don't be so paranoid, you just need to learn some things. nobody is born knowing this stuff.


"why is that when 2+2 is allways 4?"
bad example, you need to add numbers that take more digits to store, like
"why is that when 5+5 is allways 10?"
when you don't have two digits to store the answer, that's when.


"And i recently had again the problem of listening to a mix via the apple quicktime player.."
you never found out why it sounded different so you can't blame anyone but yourself.


but why you can overload one 32 bit floating point system easy while on others like ableton you can record with +30 db and turn it down on the master and everything is allwright?
that's already been covered in this thread. because there are design designs that have to be made. the math can be the same but how you design a function in a DAW would be different. go back in the thread and look for a quote by Henke, posted by ethios4.


"And other question..Are all 32bit floating audio engines the same? because its just math they should be..."
I wonder why some companys choose to have a clipping risk while others choose to dont have one because this luxury comes without an extra price.
Does that come really without a price? Or is the price ignored because its payed down in the -100db area?
that depends on the headroom designed into those points of the program. some places you design in more headroom than others. where you do that is a design decision.


"Is it saving cpu cycles and therefore a remain from old computertimes and you only find this still with the old digital audio companys?"
maybe, some DAWs have been around a long time and I'm sure parts of their codebase haven't changed for a long time. a similar example is recording automation into session view, that's a core part of the software that's now a limitation. sometimes DAWs get to the point where they rewrite the codebase altogether. I have no clue where Live is in this, they did overhaul the summing for Live 7.




the point is, show us a problem don't start ranting that there are problems because there are things that don't make sense to you.

[ethios4]

That may well be what he was referring to, but he totally misunderstood what that means. He said that because summing is non neutral, all programs sound different. He didn't understand or want to admit that if all of the settings are the same, you will get the same results in different programs.

Exactly. And his implication is that summing at multiple points in Samplitude (or whatever) is either neutral or somehow 'better'. Annoying.

[3phase]

just read the midi fact sheet from ableton and tell me that its no joke

ok, just read it...

page 1 - intro, definitions of recording and playback.
page 2 - definition of playthrough, jitter and latency
page 3 - how Live handles those issues with timestamps. they favor low jitter over low latency.
page 4 - a guide on jitter and latency issues that a DAW cannot compensate for or deal with.
page 5 - a guide on how they test their jitter at Ableton headquarters.
page 6 - the results of their tests on other bits of hardware
page 7 - tips on low latency, mostly really n00b stuff.
page 8 - a short summary.


so what's your problem with this??

The problem is that all this big text sells really bad data as normal...

What i measured lately shows that even the own data of ableton are much better now ... actually can match with other midi daws now... but the handling :-/
so midi still sucks with life...

However.. 8 pages to sell bad data... that are better now without having the text updated...
So in case something goes wrong again.. 5ms is all we got promised...? or just forgot to update that?
However its just midi.. nobody cares much..

But in practise regarding midi jitter in the output of events or more important the clock.. that is not timestamped as far as i know.

my personal experiance.. everything over 1 ms midi jitter is too much ..at least when its random..

"grooved" jitter can be tollearable maybe up to 2 ms...

good values are within the 0,1 ms domain.. thats what an akai mpc or old atari computer spits out

[Tone Deft]

just read the midi fact sheet from ableton and tell me that its no joke

ok, just read it...

page 1 - intro, definitions of recording and playback.
page 2 - definition of playthrough, jitter and latency
page 3 - how Live handles those issues with timestamps. they favor low jitter over low latency.
page 4 - a guide on jitter and latency issues that a DAW cannot compensate for or deal with.
page 5 - a guide on how they test their jitter at Ableton headquarters.
page 6 - the results of their tests on other bits of hardware
page 7 - tips on low latency, mostly really n00b stuff.
page 8 - a short summary.


so what's your problem with this??

The problem is that all this big text sells really bad data as normal...

What i measured lately shows that even the own data of ableton are much better now ... actually can match with other midi daws now... but the handling :-/
so midi still sucks with life...

However.. 8 pages to sell bad data... that are better now without having the text updated...
So in case something goes wrong again.. 5ms is all we got promised...? or just forgot to update that?
However its just midi.. nobody cares much..

But in practise regarding midi jitter in the output of events or more important the clock.. that is not timestamped as far as i know.

my personal experiance.. everything over 1 ms midi jitter is too much ..at least when its random..

"grooved" jitter can be tollearable maybe up to 2 ms...

good values are within the 0,1 ms domain.. thats what an akai mpc or old atari computer spits out

a guy with such high standards and so little understanding would have problems, I can dig that.

"big text to sell bad data"??? what???


evidence, give us evidence!!

[3phase]

[

b]"why is that when 2+2 is allways 4?"[/b]
bad example, you need to add numbers that take more digits to store, like
"why is that when 5+5 is allways 10?"
when you don't have two digits to store the answer, that's when.

But i thaught we have unlimited digits with 32 bit floating point...
or is it not that unlimited? maybe as unlimited as digital pure 16 bit sound back in the 80´s?

"And i recently had again the problem of listening to a mix via the apple quicktime player.."
you never found out why it sounded different so you can't blame anyone but yourself.

oh i found out... playing the file in logic sounded better than ... it was life...
An old problem..thaught that was solved.. i reported it as a probable bug... but maybe the fact sheet is just wrong or not precise enough about how to get transparent audio thru life...

i thaught warping is automatical of when the file is set on song speed.. tru or false?





[

b]"And other question..Are all 32bit floating audio engines the same? because its just math they should be..."
I wonder why some companys choose to have a clipping risk while others choose to dont have one because this luxury comes without an extra price.
Does that come really without a price? Or is the price ignored because its payed down in the -100db area?[/b]
that depends on the headroom designed into those points of the program. some places you design in more headroom than others. where you do that is a design decision.

Isnt it possible that other daws have choosen more wisely than? at least the debate regarding the ableton sound is rather old and dies hard

[

b]"Is it saving cpu cycles and therefore a remain from old computertimes and you only find this still with the old digital audio companys?"[/b]
maybe, some DAWs have been around a long time and I'm sure parts of their codebase haven't changed for a long time. a similar example is recording automation into session view, that's a core part of the software that's now a limitation. sometimes DAWs get to the point where they rewrite the codebase altogether. I have no clue where Live is in this, they did overhaul the summing for Live 7.

I remember..and the sound qualiy got better since.
only thing that irritates me is that the 32float =32 bit float argument was braought by ableton allready back in live 3 or 4 times in the same discussion.. claiming back theire that ableton sounds as good as any other 32 bit float DAW...


just logical..when this was false back than.. it can be false now as well..or?

the point is, show us a problem don't start ranting that there are problems because there are things that don't make sense to you.

I am not ranting.. i am talking about the problem and so showing it...
MAny people do. many people think that there is an abketon sound...
And quite a few think its not the best one out there.. some like it.. its to a certain degree a matter of taste.. but it should be possible to play back transparent if needed

[shatzer]

Well shit.. look at that! Theres OTHER people that can hear differences tooo!!! REALLY?!?!? I thought i was the only one!

And there you are! ------------------> http://www.gearslutz.com/board/4790490-post132.html

The shootout I did was done in a professional studio and monitored through $25k Monitors --> http://www.transaudiogroup.com/atc/scm3 ... ount.shtml - They were all internal laybacks rendered in each program. You could EASILY hear the differences between EVERYTHING we tested. Did the test about a year and half ago, and only tested the real major players at that time.

Heres what we tested and heres my first impressions, if i can remember correctly...

Ableton - Nice transient attack but kinda plastique sounding compared to others.. Not horrible, but noticable
PT 7.4 - Actually had "no comment" for it on my notes.. didnt like it at all - Boring and lifeless(I laughed when i saw the results)
Reaper - Was my second choice - The low end was super fat in Reaper
Samplitude - My final choice - Smoked everything.. Very clear, Very wide.. "smooth" sounding, very plesant to the ears. Best Balance
Cubase - Decent - Lacking depth and lacking the clarity that Reaper and Samplitude had.
Nuendo - Almost the same as Cubase - Slighty more space and mid/bass response.
Sonar 8 - Very clear, Very Balanced - Low end wasnt quite as fat as the Reaper and Samplitude. Was my third choice.
Logic 8 - Messy Low end - Harsh Mids - Decent transient attack - Not that great compared to the others.

Take it for whatever its worth.

But again, i could care less what any of you use, what you hear/dont hear.. I'm basically saying the same thing as drBill in that gearslutz thread. Ive heard it with my own ears. It becomes VERY apparent when you get into large tracks counts, which programs sound better then others. Do i need to post a screenshot with 100 tracks, and 60 plugins so that you can see that people DO mix large tracks? Hell, read around gearslutz and youll see a guy named Ken Lewis talking about 225 track songs with some of those Puff Daddy artists.. Danity Kane and that other boy band he has. 100 tracks just of vocals alone.. THATS REAL WORLD!

When i refer to strain on the audio engine. I'm saying.. It doesnt take a lot of processing to playback two tracks of pinknoise and calculate "the sound" for your ears. Try it with 100 tracks in two different programs and tell me if they null out. IT WONT! And whats that say?? Each engine is different.. And whats thats say? They all sound different because they arent nulling at 100 tracks.. end of discussion.

But go ahead and believe what you believe and I'll believe what i believe and laugh all the way to the bank.

Wow, great proof there! (enter sarcasm)
You can obviously tell which daw he uses most.
I'm sorry but this thread has became complete crap. Post some damn files and some REAL specs. Everything you use from the mic, the pre, compressor, eq, THE ROOM, AD converters, will all affect the audio. Blaming it on the software is just a poor excuse to shitty engineering. Telling us major players helped with the test doesn't tell us anything at all.

[Tone Deft]

"But i thaught we have unlimited digits with 32 bit floating point...
no. you have a finite number of values to work with but you can apply them to scale any range of numbers.

or is it not that unlimited? maybe as unlimited as digital pure 16 bit sound back in the 80´s?"
that makes no sense. 16 bit is limited. man, you really need to read up on this stuff.

maybe the fact sheet is just wrong or not precise enough about how to get transparent audio thru life...
why blame Live when it's the user that can't figure out how to play a file?

i thaught warping is automatical of when the file is set on song speed.. tru or false?
depends on the warp mode. again, you really need to go learn about this stuff.

Isnt it possible that other daws have choosen more wisely than? at least the debate regarding the ableton sound is rather old and dies hard
yes, and this has already been discussed. give us evidence on Live's sound quality.

I remember..and the sound qualiy got better since.
you can hear the difference with the 64 bit summing engine? the difference was on the order of -130dB, something like that. I don't believe you.



you need to go read up on these topics before you go making claims. you'll pick it up in time, clearly you want to know.