Since we can create our own versions of nebula by copying/renaming the XML and VST, can anyone share their INIs so we can do custom versions? There's no reason we can't have Nebula - High CPU, Nebula - Woeischris Edition, etc,...
woeischris wrote:Since we can create our own versions of nebula by copying/renaming the XML and VST, can anyone share their INIs so we can do custom versions? There's no reason we can't have Nebula - High CPU, Nebula - Woeischris Edition, etc,...
- you still need to edit each preset separately. I guess you could build a high performance preset library by editing manually and using the "save as" option.
I can run about 3 11k presets on a 4.4GHz i7 using 50ms timed kernels. Quite unpractical. Nevertheless, after my last round of tests I'm going to avoid using the "freqd" on the clean kernel for preamps/consoles & tapes.
fredrikberg wrote:It would be great to be able to work in Economy-mode and then at render nebula automatically switch to Normal-mode where you could have all timed to 50ms.
There is a "Render Quality" parameter in the xml-file set to 1.
What is that for?
It's and old parameter, please DO NOT CHANGE. There are similar request for increase quality on render but that cam be dangerous if the sound change to much from online versus offline version.
Dangerous? If the option were there the user could decide for himself how he would work with it to gain the best results. I discovered that not all programs worked with only timed mode, but those who work sounds VERY good with this little tweak.
yr wrote: - you still need to edit each preset separately. I guess you could build a high performance preset library by editing manually and using the "save as" option.
what would be a lot less painful than that would be if there were a patch that you could apply to programs from NAT, as with the internal sidechain patch, that would switch the preset to using timed mode instead of freqd.
such a patch is possible (i'm like 98% sure of it anyway). i could probably do it. here's why i haven't offered to yet (i've been thinking about mentioning this for a while now):
a)the cpu power needed for many instances is huge. i still think it would be better to have a rendering mode, but as enrique points out there could be issues with that also. but to load the program straight up in timed mode for actually working with a mix, with many instances, who has the kind of cpu that can handle that?
b)because of a), most people would have to keep unpatched versions to actually mix with, then on render maybe load the patched versions that use timed. but what if they've adjusted any parameters inside nebula? they'd then have to mirror those adjustments to the newly loaded timed preset. that would be more time consuming/tedious than just going to the 'kern' page and switching the first kern to timed before render (which in my opinion seems like the best option we have at the moment)
c)there would have to be maybe a few different patches. i can think of at least 2, one for 'preamp' style programs, the other for compressors, because preamp style programs use 50ms lengths by default usually, and compressors usually use 10ms. also if any dev changed those lengths themselves for whatever reason, the patch would undo their change (by reverting to the default)
to me the patch seems like it would be kind of pointless for those reasons, mainly a) and b), but if someone still wanted i could look into it and see if i could make the 2 patches...
Nebula is still confusing people everywhere with it's many parameters and odd interface conventions.
I really think it's overdue that we had some simple controls to switch between best, medium and low performance for example.
When I am using Nebula in a more 'offline' way I simply want the best possible quality because the CPU usage is not important to me in that instance, so a simple way to say 'give me your best sonic quality NOW' would be useful (without having to tweak many different values).
Think of it another way, if it were easier to achieve various quality levels then more new users would experience the true power of Nebula without having to become something of an expert first!
I think there are a couple of variables making a quality switch complex- the parameters that "should" be tweaked are different between preset types and sometimes depend on the way a certain library/preset were created. On top of that, "quality" in this case is at least partly subjective, because you are trying to balance dynamic behavior with tone accuracy (while avoiding artifacts).
I think it's safe to say that most preamp/console & tape presets sound better using classic mode and (at least) timed clean kernel. EQ's, on the other hand, are already more problematic- some libraries sound very strange with this "tweak"...