Transient loss

Tips & tricks, working results, technical support
Cupwise
Expert
Expert
Posts: 1044
Joined: Tue Nov 23, 2010 2:03 am

Re: Transient loss

Post by Cupwise » Thu Mar 01, 2012 10:12 pm

Raynorshyn wrote:Ok... I'm still a hair confused here:

Timp says "point the arrows so they are going towards freqd player lengths.

Then increase the timed kernals in order to complete the sample lengths"

Then Cupwise says "if you want to use timed only, the arrow should be on timed"

In Classic mode, or hybrid, which way should the arrow point to use the timed engine on the kernel?
sorry Rayn, fredrikberg is right. i've never really paid attention to the arrow, i always go by the (1), which is what i was thinking of. the (1) will be by whichever one you are using, and the arrow should be as fredrikberg says.

i think this topic is interesting because i haven't seen it talked about much in the forums. i guess because timed is so cpu heavy. i've compared the different versions but not enough to really come to any conclusions. my cpu isn't quite up to snuff so i can't really listen to long timed kernels and do a quick A/B with freqd, to decide if i think timed sounds that much better.

User avatar
misterambient
User Level V
User Level V
Posts: 56
Joined: Sat Mar 05, 2011 3:11 am
Location: The Netherlandz
Contact:

Re: Transient loss

Post by misterambient » Fri Mar 02, 2012 2:20 am

Wow I just tested classic mode with 10 ms on clean, even and odd harmonics, arrows pointed at freqd (click on them) and its an incredible difference!! Doing this just with one preamp, in 44.1 Hz is like you're on stage with the singer... incredible. Prf went from 10% to 40% though (on my i5 760). And it introduces half a second latency. :P 8-)

User avatar
misterambient
User Level V
User Level V
Posts: 56
Joined: Sat Mar 05, 2011 3:11 am
Location: The Netherlandz
Contact:

Re: Transient loss

Post by misterambient » Fri Mar 02, 2012 2:36 am

Sonic bliss!! Moving over with this to my overclocked i7 970 now because it is frying my computer after tweaking 5 instances, just like timp, Cupwise and enrique said it would. :twisted:

jrasia
User Level II
User Level II
Posts: 25
Joined: Fri Dec 10, 2010 4:39 am

Re: Transient loss

Post by jrasia » Fri Mar 02, 2012 4:43 am

Hmm, I noticed a serious lack of bass content when I perform this 'tweak'.

Am I doing something wrong, or is it perhaps with less bass content, you are noticing more transient information now?

Cupwise
Expert
Expert
Posts: 1044
Joined: Tue Nov 23, 2010 2:03 am

Re: Transient loss

Post by Cupwise » Fri Mar 02, 2012 5:20 am

yeah that's one thing to be aware of here, lower kern lengths mean the bass cycles in the impulses are going to be cut short, meaning bass is cut.

you can notice that small changes to kern length below 10ms cause drastic differences in the sound. vst analyser would come in handy to see if the difference you hear is due something like bass being cut, as jrasia mentioned.

dpclarkson
Member
Member
Posts: 320
Joined: Thu Apr 22, 2010 9:53 am

Re: Transient loss

Post by dpclarkson » Fri Mar 02, 2012 10:40 am

How much bass is being cut?
Should be considered that the 'TIMED' tweak could work
better on some sources than others, but not on every source?

fredrikberg
User Level III
User Level III
Posts: 36
Joined: Wed Mar 31, 2010 7:21 pm

Re: Transient loss

Post by fredrikberg » Fri Mar 02, 2012 10:48 am

I guess using split mode and timed at 10ms there
wouldn´t be no bass loss because Freqd player will handle rest of the kernel lenght.

Does anyone know how smaller Prog rate and Offset values affect sound quality of programs.

Decreasing them increases CPU.
Also Increasing them makes sound more dull.

I also found that some of the sounds that lost bass were in fact smeared and with correct settings the coloration were there and the transients were maintained.

dpclarkson
Member
Member
Posts: 320
Joined: Thu Apr 22, 2010 9:53 am

Re: Transient loss

Post by dpclarkson » Fri Mar 02, 2012 10:51 am

Anyone performed this tweak with compressor programs, e.g. 76D?
I would like to know if there are differences in dynamic behaviour also.

fredrikberg
User Level III
User Level III
Posts: 36
Joined: Wed Mar 31, 2010 7:21 pm

Re: Transient loss

Post by fredrikberg » Fri Mar 02, 2012 1:25 pm

dpclarkson wrote:Anyone performed this tweak with compressor programs, e.g. 76D?
I would like to know if there are differences in dynamic behaviour also.

I would say so after some testing. More grab and snap.

User avatar
Tim Petherick
Expert
Expert
Posts: 1772
Joined: Sat Apr 17, 2010 4:07 pm
Location: Bath , Uk

Re: Transient loss

Post by Tim Petherick » Fri Mar 02, 2012 1:57 pm

dpclarkson wrote:Anyone performed this tweak with compressor programs, e.g. 76D?
I would like to know if there are differences in dynamic behaviour also.
Everything should sound better with this!

I wanted to start a new thread with all these things in them, so we could have a more in depth talk about each of the functions.

Tim

Cupwise
Expert
Expert
Posts: 1044
Joined: Tue Nov 23, 2010 2:03 am

Re: Transient loss

Post by Cupwise » Fri Mar 02, 2012 4:06 pm

the prog rate is connected with kernel length, so to get a faster prog rate you need less kernel length. often when you change one the other is going to change automatically along with it. they both effect the end result you get in different ways.

all my understanding on prog rate is, is that you need faster prog rates for more accurate compression/dynamic behavior. my wording might not be 100% correct, but this is my understanding. there may be more to it also. i think the prog rate is one thing (if not THE thing) keeping nebula from handling instant compression/limiting (and therefore also squashed tape or overdriven tube effects).

i have some progs i made from overdriving cassette tape, and i was messing with them and lowered prog rate/kern length down really low, and got 'squashed drums' similar to what you would see if the drums went to the actual tape that hot. the sound was off, but the compression/amp envelope behavior was closer (looked like limited/squashed drums through an oscilloscope). since the kern length is so low though, the sound gets mangled.

so i think basically you have- longer kern length= more accurate frequency response, smaller length/faster prog rate= more accurate dynamic behavior.

the programs you get from 3rd parties and acustica themselves are set to be in the comfortable middle zone, but of course you can tweak those settings yourself for your own experimental uses... you are basically then going away from the intended 'safe' settings of those effects. kind of like overclocking a cpu...

fredrikberg
User Level III
User Level III
Posts: 36
Joined: Wed Mar 31, 2010 7:21 pm

Re: Transient loss

Post by fredrikberg » Fri Mar 02, 2012 4:24 pm

Thanks for the explanation. Do you know how Offset is related to the sound as well.

dpclarkson
Member
Member
Posts: 320
Joined: Thu Apr 22, 2010 9:53 am

Re: Transient loss

Post by dpclarkson » Fri Mar 02, 2012 4:41 pm

Just did a quick test with 76D;
the sound is more focused and the dynamic
behaviour seems less 'messy'.
Can anyone confirm this, just to know it's
not in my head.
Would the dynamics of preamps/tape/tube also
benefit from the evf17-parameter?
Everything seems to sound more snappy, more
real and more focused.
The TIME and FREQD don't also null, so it's
not a complete mindf*ck.

Cupwise
Expert
Expert
Posts: 1044
Joined: Tue Nov 23, 2010 2:03 am

Re: Transient loss

Post by Cupwise » Fri Mar 02, 2012 6:02 pm

fredrikberg: i don't know anything about offset or what exactly it does.

User avatar
Tim Petherick
Expert
Expert
Posts: 1772
Joined: Sat Apr 17, 2010 4:07 pm
Location: Bath , Uk

Re: Transient loss

Post by Tim Petherick » Fri Mar 02, 2012 6:15 pm

dpclarkson wrote:Just did a quick test with 76D;
the sound is more focused and the dynamic
behaviour seems less 'messy'.
Can anyone confirm this, just to know it's
not in my head.
Would the dynamics of preamps/tape/tube also
benefit from the evf17-parameter?
Everything seems to sound more snappy, more
real and more focused.
The TIME and FREQD don't also null, so it's
not a complete mindf*ck.
:D

When I first started tweaking nebula , i didn't really know what I was doing ,But after a while things start to make sense . I think the manual could do with a revision of some sort.
Evf 17 seems to make everything sound cleaner on the transient , less mud. It makes sense that it would sound good on everything. Last year I asked G what it excatly was . He said evf 17 is like a debugged rms 17




Tim

Post Reply