Rendering to disk DEGRADES SOUND QUALITY!!!

Discuss music production with Ableton Live.
olafmol
Posts: 379
Joined: Fri Jan 31, 2003 11:57 am

Post by olafmol » Sun Sep 18, 2005 4:25 pm

did you disable normalise?

glsimonsen
Posts: 58
Joined: Fri Jan 14, 2005 9:59 pm

Post by glsimonsen » Sun Sep 18, 2005 5:54 pm

atomic,

oh yeah, there's a difference. a BIG difference, not subtle at all. perhaps, it works better on a pc. i'm using mac.

leisuremuffin
Posts: 4721
Joined: Tue Apr 06, 2004 12:45 am
Location: New Jersey

Post by leisuremuffin » Sun Sep 18, 2005 8:15 pm

weezy wrote:I am definetly going to try the resampling within live.

But basically what I do is I load the sample into live, warp it, I usually use "beat" and "1/4" (because I can hear it produces less "extra junk" than setting to 1/16 - I usually set it at 1/4 regardless of how long the sample is)
You are not getting the point of beats mode AT ALL.

Transients are the peaks or attacks in your sounds. In the case of somthing that is a beat, they would be the begining of each drum sound. Set the transients parameter of beats mode to match where the transients fall in your loop. for instance, if your loop is just a kick snare kick snare on every 1/4 note, 1/4 note transient setting will work fine. However, if you have a high hat that is on every 1/8 note, you now need to go to 1/8 transient setting or the length of the grain will be too long.


as far as all this BS about quality of render, I do believe it could possibly happen with a 3rd party plugin, however, if you are using all built in stuff from live, no F'ing way. It simply is not possible for it to be any different. Did you tell live to normalize your render output? That could change the sound. Did you play the rendered file back in live? Is it warped? Does it now have more effects on it by mistake? Did you render as loop?



.lm.
TimeableFloat ???S?e?n?d?I?n?f?o

atomic
Posts: 403
Joined: Sat Oct 30, 2004 9:45 pm
Location: Toronto, Can

Post by atomic » Sun Sep 18, 2005 8:52 pm

glsimonsen wrote:atomic,

oh yeah, there's a difference. a BIG difference, not subtle at all. perhaps, it works better on a pc. i'm using mac.
What exactly is the difference?
ABIT NF7+AMDXP@2500, RME Multiface, TC Powercore Element+Virus, UAD-1, Nord Lead2, Reason2.5, impOSCar, Microtonic, Reaktor5 and some other stuff...

DJRetard
Posts: 473
Joined: Fri Jun 17, 2005 8:48 am

Post by DJRetard » Mon Sep 19, 2005 10:25 am

???? whats the difference between "real time rendering" and recording the master out.

Also, why does the render function sound perfectly fine to me?


.lm.[/quote]

I dont know, but there must be something in it because Logic pro tools cubase nuendo SAw all have real time rendering options. Then again I could never do a real time render in live due to the cpu being maxed out constantly.

supster
Posts: 2133
Joined: Mon Sep 20, 2004 6:26 am
Location: Orlando FL

Post by supster » Mon Sep 19, 2005 10:35 am

weezy wrote:II usually use "beat" and "1/4"

you do realize this is one of the best way to get glitches and 'slipping' sound from just about evertying but the most simple material right?
.
--
NEW SPECS: Athlon 4200+ dual; A8N-SLI m/b; Win XP Home SP2; 1 GB RAM; 2x 7200 RPM HDD: 1 internal, 1 Firewire 800 (Firewire is project data drive); M-Audio Triggerfinger

josh 'vonster' von; tracks and sets
http://www.joshvon.com

supster
Posts: 2133
Joined: Mon Sep 20, 2004 6:26 am
Location: Orlando FL

Post by supster » Mon Sep 19, 2005 10:40 am

leisuremuffin wrote:
as far as all this BS about quality of render, I do believe it could possibly happen with a 3rd party plugin, however, if you are using all built in stuff from live, no F'ing way. It simply is not possible for it to be any different.

yeah, i really think most of the people complaining about this really dont know what they're doing. honestly ..

ive never had a problem with render that degraded sound qualtiy just because of the render. there was always some other factor that messed with the result.
.
--
NEW SPECS: Athlon 4200+ dual; A8N-SLI m/b; Win XP Home SP2; 1 GB RAM; 2x 7200 RPM HDD: 1 internal, 1 Firewire 800 (Firewire is project data drive); M-Audio Triggerfinger

josh 'vonster' von; tracks and sets
http://www.joshvon.com

minimal
Posts: 939
Joined: Tue Jun 01, 2004 9:57 am
Location: zh - switzerland
Contact:

Post by minimal » Mon Sep 19, 2005 11:13 am

... does someone has ever posted two audio examples to make this thing clear?
In order to be "scientific" we need an A/B comparison. I do not have any webspace, but lots of ppl here do...

supster
Posts: 2133
Joined: Mon Sep 20, 2004 6:26 am
Location: Orlando FL

Post by supster » Mon Sep 19, 2005 11:19 am

minimal wrote:... does someone has ever posted two audio examples to make this thing clear?

no, the thread needs to go on at least 4 more pages without a single example.

then when its finaly posted:

1) there will be no audio quality difference whatsoever

2) it will be completely ignored or argued with for at least another 7 pages

3) the thread will eventually be two guys talking yak hunting in peru back and forther for another two pages

4) the thread will die

5) 2 weeks later another newbie will post the same exact thread.

6) repeat
.
--
NEW SPECS: Athlon 4200+ dual; A8N-SLI m/b; Win XP Home SP2; 1 GB RAM; 2x 7200 RPM HDD: 1 internal, 1 Firewire 800 (Firewire is project data drive); M-Audio Triggerfinger

josh 'vonster' von; tracks and sets
http://www.joshvon.com

minimal
Posts: 939
Joined: Tue Jun 01, 2004 9:57 am
Location: zh - switzerland
Contact:

Post by minimal » Mon Sep 19, 2005 11:21 am

supster wrote:
minimal wrote:... does someone has ever posted two audio examples to make this thing clear?
5) 2 weeks later another newbie will post the same exact thread.

6) repeat
.

ad libitum
ad libitum


but those damn audio files....

quandry
Posts: 1611
Joined: Tue Mar 18, 2003 2:31 am
Location: San Francisco
Contact:

Post by quandry » Mon Sep 19, 2005 12:22 pm

weezy wrote:Because everybody raves about this software, but if your serious serious about ur music, you can't be satisfied with this result - and I think there are alot of people out there that agree. There's no excuses when your playing your tracks to people, its either on point then and there or go home!
hmm, I guess "ur music" and my music are different--"my tracks" are things I made myself playing instruments and programming parts myself, and don't need to be stretched.....I'm "serious about my music"--I create it 100% by myself, no loops or pieces of other people's tunes"--and I couldn't be more satisfied with the results from Live. Maybe you should say "serious about DJing" and "when you are playing other peoples tracks to people".

Maybe try stuff that doesn't need to be warped, or try out the 4 other types of warping, and try adjusting all of the parameters to get the sound you want. Especially with the new complex warp, it seems like most things can be warped in some way that doesn't degrade the sound much. Live can pretty much do it, you just need to know how to strech what type of material.
Dell Studio XPS 8100 Windows 7 64-bit, 10 GB RAM. RME Multiface, Avalon U5 & M5, Distressor, Filter Factory, UC33e, BCR-2000, FCB1010, K-Station, Hr 824 & H120 sub, EZ Bus, V-Drums, DrumKat EZ, basses, guitars, pedals... http://www.ryan-hughes.net

leisuremuffin
Posts: 4721
Joined: Tue Apr 06, 2004 12:45 am
Location: New Jersey

Post by leisuremuffin » Mon Sep 19, 2005 2:38 pm

DJRetard wrote:
???? whats the difference between "real time rendering" and recording the master out.

Also, why does the render function sound perfectly fine to me?


.lm.
I dont know, but there must be something in it because Logic pro tools cubase nuendo SAw all have real time rendering options. Then again I could never do a real time render in live due to the cpu being maxed out constantly.
Yeah, i hear ya. It must be tough to afford a decent computer when your one of "the most liscensed producers in dance music." Whatever buddy.


.lm.
TimeableFloat ???S?e?n?d?I?n?f?o

ashley_k
Posts: 349
Joined: Thu Oct 07, 2004 6:37 pm
Location: Kent, UK

Post by ashley_k » Mon Sep 19, 2005 4:17 pm

So when exactly does the yak hunting season start...

weezy
Posts: 4
Joined: Wed Sep 14, 2005 12:30 am

Post by weezy » Mon Sep 19, 2005 4:24 pm

No I'm not talking about using it to dj, I'm talking about playing beats for industry people (so when i meant playing it for people and it has to sound on point, no excuses - you look stupid when you have to say "oh it should sound actually like that or my computer software kind of messed that up there...").
If I dj, I use vinyl or if its there already, I'll use serato. I understand maybe when it comes to producing people can entirely switch to software but come on not djin! Well atleast hiphop djin...honour the tradition of djin - all I have to say.


But for ableton, because I have honestly tried all the different combo's when warping, say I have a 8 bar sample which has everything, kicks/snares/hihats/horns etc. etc. what is the best combination of (beats/complex etc. and transients 1/4 or 1/16 etc.) to use?

And yes of course you wouldn't have trouble time strething your own created music cuz the sounds are already in your sound bank all seperated. But with samples you don't have that luxury.

am is are
Posts: 107
Joined: Tue Jul 26, 2005 4:44 pm
Location: New York
Contact:

REALITY CHECK

Post by am is are » Mon Sep 19, 2005 9:46 pm

anytime you are working with sound and you stretch that sound, the quality of that sound will change. period. that's reality. of course if you are working with analog (tape) the timbre and pitch will change as you speed the sound up or slow it down.

in the digital world the quality of the sound will always change.

2 things you can check: in the clip info window, in the middle SAMPLE section, there is a window for setting the original tempo of the clip you are working with. this tempo needs to be accurate (the same as the original tempo of the clip) to get best results. for pristine results, your project tempo (upper left of LIVE window) should be the same. in that case, and only in that case, LIVE is not doing any stretching of the sample and so it should sound exactly the same as your original sample.

in the transients pulldown, below the TEMPO setting in the SAMPLE section, it's very important to experiment and choose the best setting for each sample -- there can be a huge difference between the different settings.

if you don't know the exact original tempo of your sample, use the tap tempo function set to a keyboard or midi key and tap along with your sample's downbeat. do it a few times and then set that as the SAMPLE tempo.

you might also try using the COMPLEX warp mode rather than BEATS if there's other information in your sample other than purely rhythmic/drum stuff.

but remember, LIVE was created as a live performance sampling instrument. it's not a time-stretching utility and, yes, there are definitely better pitch/time-stretching algorithms out there. download Celemony's MELODYNE demo and check that out. it's a better algorithm. NOTE that it can't stretch stuff live. you have to stop the sound and let the program analyze the file.

maybe the best algorithm i've heard is MPEX Time Scaling which somehow takes into account how human beings actually HEAR in order to work it's magic. It is also NOT a live-stretching algorithm. it takes some time to do the analysis and stretching but it works pretty great. MPEX Time Scaling is included with SONAR Producer Edition.

one more tip: if you want to use a sample and want it to sound pristine, you need to avoid stretching it. period. so choose samples that are already at the tempo you need.

this isn't magic, it's math, and there are finite limits to accuracy when interpolation is involved. maybe that's not the answer you want to hear, but it is something you have to accept. Either invest in a program/software with the best stretching algorithms or accept LIVE for what it is. a pretty amazing live performance software that gives you many of the benefits of working in a DAW.

peace

Post Reply