analoginthebox.com wrote: You as as a developer should know the Nebula SRC algorithm that is pretty unacceptable! You could measure it, or use your ears and you will see/hear that a 96KHz -> 44.1KHz converted version gets worse than a native 44.1 version. A simple converter loop confirms that.
Besides that there could appear all kinds of minor issues with converted programs. For the Germanos series, NAT failed completely at deconvolving a simple 96KHz EQ. Giancarlo and I tried to understand this behaviour, but did'nt get a solution. There are issues, period. Just search through the old forum archieves about users reporting strange behaviour...
Nikklas I´m an owner of your products as well as Alexb´s and CdSoundmaster´s stuff, I don´t have any technical knowledge about Nebula´s guts, but I have ears, and 95 % of my mixing sessions relies on CdSoundmaster and Alexb stuff (yours too).... I never listened nothing but good things to my tracks when these are processed by Alex consoles/eq´s or CdSounmaster eq´s/tapes, no sound degradation or whatever, even more, I couldn´t mix without them right now. The same with plate and spring reverb from VNXT. Oh, and I work at 44/24.
Actually all this is very subjective... I think the digital sound excellent...Logically that a machine Studer is undoubtly better. I have I mixed in 96khz, and 88.200 that sounds better, sonically seems joke more true . not like what I heard with 96khz...It is somewhat subjective, I particularly did not like. I really enjoyed the 88.200... Further the illusion of 88.200KHZ ended fast, when I sent for mastering . and ultimately became an mp3 320Kbps...That in my ear made no distinction. Today I record and I mix again at 44,100 khz...I'm seriously thinking of buying a tape machine... why in the end it not will sound like the analog ,That is a sound with depth. nebula is excellent... More is a thought subjective and unique... everyone has their opinions. The songs were memorable in antiquated equipment, that Today is a search and a reference sound. mix in 96khz will be a reality... most believe that it is not for everyone... I believe it makes no difference in this world Mp3... that as the old vinyl it seems that settled in this new universe. Again I say that this is a subjective view of the world. One point of view ... without fascism .
Last edited by Mimi on Wed Jan 26, 2011 2:10 am, edited 1 time in total.
Well, I wanted to stay out of this discussion because it is obsolete! We will release native 44.1KHz and 96KHz in the future...
But @prosodio: I think you got me wrong. I was not saying you'll degrade your signals by using samplerate converted Nebula libraries. I have bought many libraries from 3rd party developers and I am pretty happy with them
I was repyling to the statement "96KHz is always better" which is simply not true if you are using Nebula and convert 96KHz programs to 44.1KHz. The kernels get degraded and sample rate conversion causes inaccuracy and faults. I call this "degrading" because you would get (even ) better results if the samplerate of the libraries matches your project's samplerate.
(...) i never noticed it but if this is indeed true
Yes, that is one of the reasons why we release our libraries at the samplerate that most of our customers are working with. But I repeat again: 96KHz + 44.1KHz from AITB for the future.
If you can handle working at 96 khz, that pesky thing called science works in your benifit when you use a lot of digital plugins.
Alising and in-harmonic distortion is pushed outside of the main bandwidth we can hear even when we fold it down to 44.1.
My untrained girlfriend in a blind test could hear the difference between a mix sourced from 44.1 where all my libraries, plugins ect where played at that rate. I then reopen the exact same project in 96 its like I have a sound stage that is now 'there' she said in her own words. Other friends can hear a difference (positive and not just 'different' )after my blind tests. So if 'average' people can hear it then why wouldn't we want to capture these things at the highest possible source rate for future proofing these libraries.
Do you guys rteally think 24/441 audio or even mp3 are the future?
How many people are spending thousands on new hi-res televisions and home theatre systems for their homes? THATS THE FUTURE AUDIO MARKET.
Not sure whats the big deal about simply having it sourced at 96 khz and being done with it like other
How many people in America now have high quality home theatre systems with Bluray?? My friend (who is no audio engineer at all) was complaining how movies are not mixed in 7.1. He is your ‘average’ guy complaining about the lack of 7.1?! I would have never DREAMEd about this kind of complaint from someone so far removed from the recording/engineering process saying this.
Besides the general public will be sold on this stuff when Bluray and hires audio make a huge splash.
People will rebuy their CDs so they can hear Ziggy Stardust in 5.1 in hi-resolution audio on their Bluray players and home theatre setups.
If you don't see that happening as an audio engineer you may be asking someone in the future 'would you like fries with that'.
24/96 multi-channel audio will be the norm, why release 44.1 libraries when the end delivery format in the next 24-36 months will drastically change
People are so focused on mp3 as the ‘end delivery’ format are walking around totally blind by the future of music and entertainment playback.
People ARE MOVED by higher resolution picture and audio. If they where not why in the heck are people buying/renting Bluray players and sending thousands on home theatre systems then if all they care about is a 128kbps mp3 in the middle of a mega-recession.
Other developers have 'future' proofed their libraries.
Why punish people that are trying to work at the highest possible resolution from start to finish?
A 96khz library loaded from a proper SSD even on my 1st gen mac pro loads in less than one second. Converting 441 libraries when your working at 96khz is a MAJOR PAIN TIME WISE!! I have to wait sometimes 10 full seconds for the SRC to upsample.
If you want to offer 2 flavors and charge more for the 96khz ones I'm all game because AITB stuff is really amazing stuff. Even if it where 1.5x the cost of the 'native' 441 I'd be more then happy to pay that.
Regardless its all good and can not wait for future libraries from all of the amazing developers regardless of SRC.
peter - in your tests did you downsample the examples mixed at 96k?
for me personally once downsampled I found the differences smaller - but still there-at least when using a lot of algorithmic plugins...however until I have a 8 core computer running at 64bit I dont think it worth me personally working at 96. The overheads cost more than the benefit at this time for me. In the future - things may change.
Very generous of AITB to service us with both sample rates given the low cost of the programs.
We mustn't forget the great deal we get or take it for granted with all our demands...
greekpeet wrote:24/96 multi-channel audio will be the norm, why release 44.1 libraries when the end delivery format in the next 24-36 months will drastically change
Than won't happen, they tried that and it didn't work. The SACD was a fiasco, it never replaced audio cds; instead of that, mp3s took over.
It will take at least 5 years until 24/96 COULD became the norm. Until then the computers will surely be powerful enough to handle a full mix with Nebula on each channel. Until then 44.1 is the practical, down to earth approach.
greekpeet wrote:Do you guys rteally think 24/441 audio or even mp3 are the future?
just two things:
it's a matter of fact that at least 70% of all people making music be it professional or just for fun work with 44.1 or 48khz. and that's the present not the future.
and talking about the future, i doubt there's something like a future of the audio industry. and those who will work professionally (as in get paid sufficiently) in some kind of multimedia industry will be few.
ps: made my own 96khz tests with my lynx aurora and it's possible that i will convert at some point but not now.
pps: if it's possible to integrate a better SRC algo, why not do it? for the majority of users. so that's actually four things now