Login

Neb 3 Pro and Neb 3 Pro x64 sound different

Tips & tricks, working results, technical support

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby fuseburn » Fri Apr 01, 2011 1:17 pm

Alright, so this is what I can reproduce:

At every push of SAVE the AHEAD value is divided by the ratio of current samplerate to 44100. That means if you're working on 44.1k, it remains the same (divided by 1). But if you're working on 96k (like me) it is divided by 2,18 (and then rounded), like 6 ms -> 2,75 ms (displayed as 2,8 ms -> 1,3 ms (1250 in XML) etc.
To be even more precise: Nebula seems to do the wrong conversion while reading the config XML when opening the MAST page. The XML contains 6000 - Nebula displays 2,8 ms. You hit SAVE and it writes 2750 to the XML.
User avatar
fuseburn
User Level X
User Level X
 
Posts: 133
Joined: Mon Apr 05, 2010 6:19 pm

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby Mercado_Negro » Fri Apr 01, 2011 2:10 pm

fuseburn wrote:Alright, so this is what I can reproduce:

At every push of SAVE the AHEAD value is divided by the ratio of current samplerate to 44100. That means if you're working on 44.1k, it remains the same (divided by 1). But if you're working on 96k (like me) it is divided by 2,18 (and then rounded), like 6 ms -> 2,75 ms (displayed as 2,8 ms -> 1,3 ms (1250 in XML) etc.
To be even more precise: Nebula seems to do the wrong conversion while reading the config XML when opening the MAST page. The XML contains 6000 - Nebula displays 2,8 ms. You hit SAVE and it writes 2750 to the XML.


Yes, I can confirm this but what's 'AHEAD' for? What's supposed to do?

[EDIT] OK, I found this on the MAST page F.A.Q.:

"AHEAD:
This parameter controls the look-ahead buffer used by compressor and dynamic emulation programs in general. It is expressed in milliseconds. Increasing this parameter will increase latency."

Question to Giancarlo: Could this affect the sound of libraries? If it's so, what value should we use in general?
i7 3770k :: Asus P8H77-V LE :: 16Gb DDR3 @1600MHz :: Geforce GT 520 :: OCZ-Vertex 128Gb :: WD Black Series 1Tb and Green Series 1Tb :: F*******e Liquid56 :: REAPER 64bit and StudioOne 64bit (both latest versions) :: Win 10 64bit
User avatar
Mercado_Negro
Global Moderator
Global Moderator
 
Posts: 1353
Joined: Sat Mar 27, 2010 9:30 am

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby giancarlo » Fri Apr 01, 2011 2:43 pm

it's expressed in milliseconds. It's truncated internally for values grater than 1.5 milliseconds. Don't touch it. There is an hidden issue with that value: samples are truncated at a certain amount of samples, both 44100 and 96000. It means that 96Khz programs have less predelay than 44100 programs. So just keep 1500 and forget about it. I'll try to sort the predelay thingy at a different level, but I can't touch it now. It's not an issue imho if you don't touch things.
User avatar
giancarlo
Founder
Founder
 
Posts: 9149
Joined: Mon Sep 21, 2009 10:40 pm
Location: Italy

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby giancarlo » Fri Apr 01, 2011 2:47 pm

this topic is creating a great confusion. Someone is comparing nebula3 with nebula3 reverb, someone is comparing x64 to x86, someone is comparing nebula loaded at different frequency rates. There is no issue in nebula, unless you find a difference between x64 and x86, which seems impossible (but I can't exclude bugs).
User avatar
giancarlo
Founder
Founder
 
Posts: 9149
Joined: Mon Sep 21, 2009 10:40 pm
Location: Italy

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby Mercado_Negro » Fri Apr 01, 2011 3:42 pm

OK, I ran some tests for fun. Here's the result of a null test:

Image

I recorded the output of the project using a handy option in REAPER called "Bounce". I just played that loop and recorded it:

http://hotfile.com/dl/112908145/e5c8a7f ... t.wav.html

That's pretty much the same I've gotten with hardware in the past so don't worry, it may sound like a big deal, but it's not :)

giancarlo wrote:it's expressed in milliseconds. It's truncated internally for values grater than 1.5 milliseconds. Don't touch it. There is an hidden issue with that value: samples are truncated at a certain amount of samples, both 44100 and 96000. It means that 96Khz programs have less predelay than 44100 programs. So just keep 1500 and forget about it. I'll try to sort the predelay thingy at a different level, but I can't touch it now. It's not an issue imho if you don't touch things.


OK, thanks! I reverted to 1500 in the xml. Cheers
i7 3770k :: Asus P8H77-V LE :: 16Gb DDR3 @1600MHz :: Geforce GT 520 :: OCZ-Vertex 128Gb :: WD Black Series 1Tb and Green Series 1Tb :: F*******e Liquid56 :: REAPER 64bit and StudioOne 64bit (both latest versions) :: Win 10 64bit
User avatar
Mercado_Negro
Global Moderator
Global Moderator
 
Posts: 1353
Joined: Sat Mar 27, 2010 9:30 am

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby ngarjuna » Fri Apr 01, 2011 4:56 pm

giancarlo wrote:this topic is creating a great confusion. Someone is comparing nebula3 with nebula3 reverb, someone is comparing x64 to x86, someone is comparing nebula loaded at different frequency rates. There is no issue in nebula, unless you find a difference between x64 and x86, which seems impossible (but I can't exclude bugs).


This topic is dispelling great confusion, I think it's clear people were already confused (certainly myself included).
User avatar
ngarjuna
Expert
Expert
 
Posts: 778
Joined: Tue Mar 30, 2010 5:04 pm
Location: Miami

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby fuseburn » Fri Apr 01, 2011 6:36 pm

giancarlo wrote:it's expressed in milliseconds. It's truncated internally for values grater than 1.5 milliseconds. Don't touch it. There is an hidden issue with that value: samples are truncated at a certain amount of samples, both 44100 and 96000. It means that 96Khz programs have less predelay than 44100 programs. So just keep 1500 and forget about it. I'll try to sort the predelay thingy at a different level, but I can't touch it now. It's not an issue imho if you don't touch things.


Well, it IS an issue !
Whenever you change ANY OTHER value in the MAST page and hit save, your halving the value of AHEAD (if you're on a different Samplerate than 96k). You don't even have to touch it to change it. That's a bug, plain and simple ;)
User avatar
fuseburn
User Level X
User Level X
 
Posts: 133
Joined: Mon Apr 05, 2010 6:19 pm

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby giancarlo » Fri Apr 01, 2011 8:28 pm

how many msec are you reading in mast page now? you can't go higher than approx 1.5 than don't move it any more. There is some sort of artificial intelligence on the translated value, so don't care about it. The rule inside is very complex, so the parameter has little influence on nebula behaviour, it's there just for troubleshooting
User avatar
giancarlo
Founder
Founder
 
Posts: 9149
Joined: Mon Sep 21, 2009 10:40 pm
Location: Italy

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby giancarlo » Fri Apr 01, 2011 8:37 pm

and please stay on topic, here we are speaking about x86 vs x64
User avatar
giancarlo
Founder
Founder
 
Posts: 9149
Joined: Mon Sep 21, 2009 10:40 pm
Location: Italy

Re: Neb 3 Pro and Neb 3 Pro x64 sound different

Postby fuseburn » Fri Apr 01, 2011 10:18 pm

Just try it. I can reproduce it with Nebula 3 Pro Reverb x86, x64 and Mac OS X (VST) (1.3.467, 1.3.475, no matter)

At every push of SAVE the AHEAD value is divided by the ratio of current samplerate to 44100. That means if you're working on 44.1k, it remains the same (divided by 1).
Let's say you start out with default 1500 value from the XML. Open the XML in Firefox and refresh to see the current values.
Set to 48k, 88k, 96k (no matter, just not 44.1k)
Open MAST page and immediately hit SAVE, don't touch anything. Now refresh Firefox to see the new value of AHEADLENGTH - which shouldn't have changed anyway.

Sure, I started this thread because I thought I found mysterious differences betweeen x86 and x64. Now that we found out that the untouched AHEAD MAST config value was probably changed at every testing step, I have to question my initial statement myself.
User avatar
fuseburn
User Level X
User Level X
 
Posts: 133
Joined: Mon Apr 05, 2010 6:19 pm

PreviousNext

Return to Working with Nebula

Who is online

Users browsing this forum: No registered users and 9 guests