IRs and Nebula programs of guitar cabs - Kalthallen Cabs released

[Kazrog]

Maybe Kazrog. Recabinet: The Nebula Edition

I really appreciate the kind words, but this is not going to happen. Nebula is an impressive piece of technology, however it is very limited in terms of what it can reproduce (6 harmonics, which isn't the entire behavior of a loudspeaker, or most other analog gear for that matter.) By contrast, we've got a new engine in the works that DOES emulate the entire behavior of loudspeakers, will be released later this year, and is also a lot more CPU friendly.

As Slate pointed out when he released VCC, Nebula is great for doing "black box" modeling of devices - as in "I don't know what this magic box does, but I'm going to sample it and simulate it." However, it's not a replacement for a DSP engineer meticulously developing an algorithmic model based on intimate knowledge of a piece of hardware.

I'm not saying Nebula isn't useful, again it's a very impressive technology and is very good at replicating many of the salient characteristics of various types of hardware. I've considered doing a Nebula library in years past, but like many others, I wasn't entirely satisfied with the results over regular IRs, so I turned my quest elsewhere.

I'll keep you guys posted on the future of Recabinet.

 

[hoehlentroll]

shane, that sounds promising, although that promise was often heard in the past. but we'll see that
for the time recabinet was put out it was really sounding nice and state of the art. most of the impulses were kind of dark. maybe you'll look into this issue at your next release? I'm really looking fowad too the release, though

 

[Kazrog]

shane, that sounds promising, although that promise was often heard in the past. but we'll see that
for the time recabinet was put out it was really sounding nice and state of the art. most of the impulses were kind of dark. maybe you'll look into this issue at your next release? I'm really looking fowad too the release, though

I appreciate the honesty. There have been a lot of promises made (even actual release dates) by others regarding next-gen speaker simulation, and to date there have been no actual products released that fit that description. I can understand the skepticism, which is why I don't announce release dates until the new software is already at a release candidate stage.

I've previously announced that such technology was in the works for a future version of Recabinet, but in the process of launching our first plugin last year, that development got sidelined in favor of dealing with customer requests, workflow improvements, bug fixes, audio engine enhancements, etc. Once Recabinet 3.0.4 was pushed live, I felt confident enough in the current state of the software to get 100% back into R&D for the next major update.

Speakers are quirky, misunderstood, and their behavior is understated and under-researched. A lot of things you read (even from the speaker designers themselves) aren't entirely true. This has been a long journey, but I can see the finish line ahead, and it's not far off.

As for Recabinet IRs tending to sound dark, I've also had an equal number of people say that they sound bright. YMMV and it's very much relative to your guitar, pickups, signal chain, etc. In any event, I've thought of some ways to better accommodate the wide range of input signals out there. Lots of cool stuff is on the way!

 

[hoehlentroll]

As for Recabinet IRs tending to sound dark, I've also had an equal number of people say that they sound bright.

really? don't remember comments going in that direction (not on this forum at least). it's like you say, the secret of real sounding guitar lies in the cabinet. every step towards the "real thing" is apreciated. I#m still using recabinet to blend in with some brighter samples and brings in some fatness, which I really like! If you need any beta testers send me a PM

 

[Kazrog]

really? don't remember comments going in that direction (not on this forum at least). it's like you say, the secret of real sounding guitar lies in the cabinet. every step towards the "real thing" is apreciated. I#m still using recabinet to blend in with some brighter samples and brings in some fatness, which I really like! If you need any beta testers send me a PM

Plenty of users from this forum and other metal guys have told me that Recabinet 2 was too bright, and Recabinet 1 was too dark. So, in Recabinet 3, the entire library has a more neutral, open balance to it, and there's built in EQ and blending.

If you want more top end sizzle in Recabinet 3, bypass the low pass filter, also try blending brighter mics in such as the 57, i5, 409, and U87 to taste. The 121 is a dark mic intrinsically, so if I am using it I always blend in something brighter.

Anyway, this wasn't intended as a thread hijack, just wanted to respond to your original idea. PM sent by the way.

 

[Ermz]

As Slate pointed out when he released VCC, Nebula is great for doing "black box" modeling of devices - as in "I don't know what this magic box does, but I'm going to sample it and simulate it." However, it's not a replacement for a DSP engineer meticulously developing an algorithmic model based on intimate knowledge of a piece of hardware.

Hard to argue with that - however, drawing a parallel, given the recent explosion of the Kemper Profiling Amp, and how its 'black box' approach seems to be yielding more accurate results as compared to, let's say, the algorithmic models of the Axe-FX 2, is it not fair to say that in some instances the 'black box modeling' draws more favorable results? While I do believe Nebula has a fair way to go in order to create a truly realistic cabinet model, in some cases, chiefly EQ, I've not found its algorithm-based equal.

As a sidenote, I'm really looking forward to the Recabinet update. I've spoken to a certain someone partnered on the project, and hopes appear to be high. Whatever the future of cabinet modeling is, and whichever form or shape it takes, I look forward to being at the vanguard of the userbase. So good luck!

 

[Skaldir]

Just wanted to add what Kazrog wrote about the 6 harmonics in Nebula isn't entirely correct. By that logic a normal impulse would have 0 harmonics. Nebula has up to 10 possible harmonics to look at from a basic frequency to capture non-linearities.
This is independant from dynamics.

Guitar speakers are not that non-linear as people think, thats why capturing them with a normal IR works so well.
The most non-linearities you can find in frequency ranges its not supposed to playback anyway for guitars. Roughly 0 - 80 Hz

For me its all about A/B comparisons. I remember when doing the shootouts with Ermz we had a Nebula setting to capture the cabs once that was even tighter than the real mic'ed cab. But it was all about to make it as close as possible to the original.

Working with simulations instead of sampling is of course interesting cpu wise. But if it comes closer to the real thing than my latest Nebula method we will have to see.
The "black box theorem" IS the way impulses and Nebula work. I don't know what is wrong about this approach, except it needs a lot of cpu. Well in case of cabs not that much.
Happy modeling

 

[Sludgekicker]

wow folks this sounds kinda fat heard some mixes tho and there are few ones you can easily hear its some kind of program processed DI but even if they not real it still sounds fat and cool.

Recently ive read about SIR2 as well, im curious if its worth a try or go straight for nebula?

 

[bassguy]

Hate to ask the nooby nebula questions, but these can be loaded in the standard version of nebula 3 right? no need for the pro version?

 

[MIAnn]

wow folks this sounds kinda fat heard some mixes tho and there are few ones you can easily hear its some kind of program processed DI but even if they not real it still sounds fat and cool.

Recently ive read about SIR2 as well, im curious if its worth a try or go straight for nebula?

check nebula 2 out for 29 euros

 

[Skaldir]

Hate to ask the nooby nebula questions, but these can be loaded in the standard version of nebula 3 right? no need for the pro version?

Can't be loaded in Nebula free. Only Nebula 3 Pro/Server
But it seems also the commercial version of Nebula 2 can load it.
See MIAnns posts.

 

[Kazrog]

Hard to argue with that - however, drawing a parallel, given the recent explosion of the Kemper Profiling Amp, and how its 'black box' approach seems to be yielding more accurate results as compared to, let's say, the algorithmic models of the Axe-FX 2, is it not fair to say that in some instances the 'black box modeling' draws more favorable results? While I do believe Nebula has a fair way to go in order to create a truly realistic cabinet model, in some cases, chiefly EQ, I've not found its algorithm-based equal.

You're entitled to your opinions regarding relative accuracy, but I'm sure you know that they are by no means universal. There are a lot of experienced users out there who feel exactly the opposite about Kemper vs. Axe FX 2, for example (myself included, especially as of the v5 firmware update.) They're both impressive, though, and I respect the developers behind them immensely. Also, the Kemper isn't 100% black box, it's more grey box actually.

As a sidenote, I'm really looking forward to the Recabinet update. I've spoken to a certain someone partnered on the project, and hopes appear to be high. Whatever the future of cabinet modeling is, and whichever form or shape it takes, I look forward to being at the vanguard of the userbase. So good luck!

Cool. Thanks!

 

[bassguy]

Can't be loaded in Nebula free. Only Nebula 3 Pro/Server
But it seems also the commercial version of Nebula 2 can load it.
See MIAnns posts.

On their website they have:
Nebula 3 free
Nebula 2
Nebula 3 (http://www.acustica-audio.com/index.php?option=com_content&view=article&id=11&Itemid=123)
Nebula 3 Pro
Nebula 3 Sever

So this wouldn't load in the the regular commercial version of nebula 3 (the one I linked to above), only the pro/server versions (or nebula 2)?

 

[hoehlentroll]

none of the free is working

 

[Sodhats]

Are you still going to do the 2x12 one for cleans? I'm immensely looking forward to it if so. Although you can get a pretty good clean tone with the md421 ones.

Also, what's the recommended volume going into nebula? There's a certain point where some high midrangey frequencies start creeping in, and whether or not I like them varies depending on the tone.

 

[TRUIE]

-18 dB FS, it's in the manual

And yeah, more poweramps/cabs would be great !
But I know it's a lot of work, and this library alone is already awesome, and it's even more awesome to give it for free

 

[Sodhats]

O. Haha.

Reassuringly I'd come to more or less the same conclusion by compressing the guitar track and slowly turning the volume knob.

 

[Skaldir]

none of the free is working

Its an either or thing in Nebula.
Either the programs you create work in Free or in the commercial versions.
I think a free program doesn't open in the pro version either, not only the other way around.

Yes its calibrated to work with -18dbFS.

Some guys I don't know spread the word about my IRs in other guitar forums and the people there also really dig them. Very happy with the response so far.

 

[harmonic1]

I went to the minus.com download page for these and have had it sitting open for a while now, but clicking on "download" or the big RAR button just reloads the page each time, no pop-up, save window, etc. I registered an account too, still no luck. What am I missing?

 

[Skaldir]

I went to the minus.com download page for these and have had it sitting open for a while now, but clicking on "download" or the big RAR button just reloads the page each time, no pop-up, save window, etc. I registered an account too, still no luck. What am I missing?

You're missing some great IRs
No, you're doing it right. It seems for some people minus makes problems.
What browser do you use? Maybe try another one.