Login

ATTN all Library Developers - Have a Tech Question

Tips & tricks, working results, technical support

ATTN all Library Developers - Have a Tech Question

Postby RJHollins » Sun Feb 10, 2013 9:37 am

Not that the whole Community isn't welcomed to respond [please do], but I would much appreciate hearing from the Library Developers, in particular, to shed light on this.

In my normal, rambling way :lol:

Let's talk EQ presets.

To start, many EQ libraries include full harmonic, clean [no harmonics], and often times the Pre-amp.

I would like to clarify on the strategy of implementing several presets in an FX chain that would 'most' closely mirror the 'original' harmonic character of 'said' EQ.

Please ... let us temporarily bypass the 'use your ears' ruling, as I would like to hear more on the 'technical' aspect.

Mentioned in library documentation is the suggested chain to be built from a single 'FULL Harmonic' instance, and the remaining plugs to use the 'Clean' presets. As paraphrased, 'this would represent the original harmonic content of the unit' [basic premise] :roll:

Further it's said that the 'Full' preset should be used for 'boosting'.

So the question that presents itself is:

What if we used all preset EQ's of only the 'Clean' for each instance, then follow that with one instance of the Pre-amp?

Not an major revelation, as this stems from the 'MOJO' technique, whereby an 'Algo' plug is fed into a Nebula Pre [example]. This very workable technique is very practical and sonically better than the algo alone.

To the specific interest [at least mine :)], to [as closely] replicate the sonic signature of a given piece of hardware. 2. To look at a possible strategy that may simplify the preset ordering, with 3, to configure the FXChain to represent a complete hardware representation. [if it has 5 bands, then we need 5 Nebs].

Would 5 'clean' Nebs fed to a Full Preamp fulfill the same as using 4 clean + 1 'dirty' :twisted:

AlexB ??? Michael ???? our beloved 'Other Developers ??? !!! Just looking for a constructive conversation/insight on things Nebula to helps expand or discover new awareness.

Thanks for taking the time to read ... let the education continue 8-)
i7-5820k, MSI X99A Plus, 16 GIG Ram, Noctua NH-D14, Win-7 Pro [64-bit], Reaper-64

NVC [Nebula Virtual Controllers]
RJHollins
Expert
Expert
 
Posts: 2633
Joined: Sun Mar 28, 2010 5:53 pm

Re: ATTN all Library Developers - Have a Tech Question

Postby yr » Sun Feb 10, 2013 10:20 am

I think that 5 clean presets + preamp are potentially more "accurate", unless your "dirty" preset (with harmonic distortion) was sampled dynamically. But you need to consider additional factors, such as the general quality of the library, playback engine artifacts, preset src etc, so "accuracy" in this case might be hard to define...
Reuven | post-production & sound-design | scenography |
website | nebula presets
User avatar
yr
Beta Tester
Beta Tester
 
Posts: 438
Joined: Mon Mar 29, 2010 9:04 am
Location: Amsterdam

Re: ATTN all Library Developers - Have a Tech Question

Postby RJHollins » Sun Feb 10, 2013 11:36 am

yr wrote:I think that 5 clean presets + preamp are potentially more "accurate", unless your "dirty" preset (with harmonic distortion) was sampled dynamically. But you need to consider additional factors, such as the general quality of the library, playback engine artifacts, preset src etc, so "accuracy" in this case might be hard to define...


Hi 'yr' ... thanks for responding! And let me add, I meant NO slight in not mentioning all Dev's by name [something I'm not so good with :oops: ].

Definitely happy to hear from your perspective!

As hoped, you have raised the very things that add to this query of mine.

In order to try and 'minimize' the range of variables, maybe we can generalize on a few 'givens' or 'ideals' in order to draw a practical, working model.

If I may ...
The intent of my initial question is NOT to directly compare the specific NEBULA library 1:1 to the original unit. Those comparisons/conversations had gone in other threads [which is great], but I'm not looking to go that way here. For one reason, only the Dev can make a 1:1 comparison, as the users only have the library. Again, let this not be an issue, to consider. Let's agree the Library is 'ideal'. [humor me] :lol:

'SR' conversion and all it's variables. This too I would prefer to neutralize for this, since no matter what SR the project is, the SR conversion 'may' have to happen ... whatever it does still provides the tool that we will use. In other-words, what we have is what we have ... lets call it 'ideal' ;)

Other 'variables' mentioned [all true], yet from the user perspective, there's not much we can do about. Yes we can modify the playback engine [TimeD, etc] ... but these types of 'mods' may or may not be done by the user [his call].

But this does raise a question [as I've not personally altered the NEB engineer on a 'Clean' preset. I have on full Preamps ... little to none with EQ's [something I need to test for myself] ... but modding a 'clean' eq preset ... I just don't know what the sonic result might be [if any]? I'll test it.

Your first observation. The idea of 5 clean presets followed by a full range Preamp. This is the arrangement I'm most curious about. It seems such an obvious configuration, yet I don't recall any mention in the documentation or forum conversation.

Granted ... the user can test these things for themselves. However ... those on the frontline, with the specific hardware right there, and the effort that goes into tweaking the library into a cohesive package, would have a deeper, intimate analysis/observation to this.

To be clear ... the choice AND the track requirement might call for just a single instance, and probably with full harmonics. Maybe several ... maybe all clean.

Looking to assemble a single FXChain [for an EQ] that contains all relevant bands with the same 'juice' the unit was spec'd for.

For others that hopefully join the conversation, I'm not trying to limit the use of NEBULA, nor trying to establish any restrictions of any sort. Far from that ... and that IS one of the beauties of NEBULA ... the mix and match without having to leave your chair :mrgreen:

I confess, I do have a side project that I'm working on [and hope to share with the Community ... it'll be free 8-) ], but I need to consider the optimum as my reference point.

Unless I'm missing something so obvious ... it would seem 'All Cleans' driving the Full Preamp would be it.

Yeah ... unless someone raises the question of driving a specific eq band for the crunch ... then what do we have ???

Much to consider for those that find this interesting as I do. Well ... back to the session! Again, thanks for sharing thoughts here. Look forward to positive conversation.

Sincerely
i7-5820k, MSI X99A Plus, 16 GIG Ram, Noctua NH-D14, Win-7 Pro [64-bit], Reaper-64

NVC [Nebula Virtual Controllers]
RJHollins
Expert
Expert
 
Posts: 2633
Joined: Sun Mar 28, 2010 5:53 pm

Re: ATTN all Library Developers - Have a Tech Question

Postby yr » Sun Feb 10, 2013 12:15 pm

I think I understand your question better- if you are looking for the most accurate way to sample an EQ, I would think that sampling each EQ setting as a preamp and providing both clean and distorted versions (and of course multiple sample-rates:) will give you/users the best results.
Reuven | post-production & sound-design | scenography |
website | nebula presets
User avatar
yr
Beta Tester
Beta Tester
 
Posts: 438
Joined: Mon Mar 29, 2010 9:04 am
Location: Amsterdam

Re: ATTN all Library Developers - Have a Tech Question

Postby fpoitevin » Sun Feb 10, 2013 4:27 pm

Reading this question, I've done what I never did before : compare for AlexB MWEq
- 3 ND EQ + 1 full
- 4 Full
- 4 ND EQ +1 MWC input
I know you don't care about "use you ears" (!) but I'm happy that I've done the test. The last one (4 ND + 1 console input) is by far the best sounding for me. I'll go for this now. Thanks for the idea !..
fpoitevin
User Level VIII
User Level VIII
 
Posts: 86
Joined: Mon Sep 20, 2010 9:48 pm
Location: Grenoble, France

Re: ATTN all Library Developers - Have a Tech Question

Postby Cupwise » Sun Feb 10, 2013 4:36 pm

yr wrote:I think I understand your question better- if you are looking for the most accurate way to sample an EQ, I would think that sampling each EQ setting as a preamp and providing both clean and distorted versions (and of course multiple sample-rates:) will give you/users the best results.


i agree. so far i think the only library that was sampled this way is henry's titan. in my opinion, that one probably gives the most accurate representation of the sampled eq. although it is less flexible than others, so there's a tradeoff.
Cupwise
Expert
Expert
 
Posts: 773
Joined: Tue Nov 23, 2010 2:03 am

Re: ATTN all Library Developers - Have a Tech Question

Postby RJHollins » Mon Feb 11, 2013 1:46 am

fpoitevin wrote:Reading this question, I've done what I never did before : compare for AlexB MWEq
- 3 ND EQ + 1 full
- 4 Full
- 4 ND EQ +1 MWC input
I know you don't care about "use you ears" (!) but I'm happy that I've done the test. The last one (4 ND + 1 console input) is by far the best sounding for me. I'll go for this now. Thanks for the idea !..


Great post 'fpoitevin'.

Thanks for sharing your personal testing/observations. This is what I'm looking to hear about.

I may have given the wrong impression with the 'just use your ears' statement. What I wanted to AVOID was dismissal of the question in whole, and 'just use ears' and move on. Without doubt, this is within the subjective decision of the user. But I didn't want to leave it at that. Thus the reason for calling attention to all Devs, Giancarlo, Enrique, etc.

I also want to get the 'technical observations' from those creating the libraries [a perspective that we 'general users' would not have].

Nonetheless ... what Users hear/experience is also valid ... so please continue sharing.

Thanks!
i7-5820k, MSI X99A Plus, 16 GIG Ram, Noctua NH-D14, Win-7 Pro [64-bit], Reaper-64

NVC [Nebula Virtual Controllers]
RJHollins
Expert
Expert
 
Posts: 2633
Joined: Sun Mar 28, 2010 5:53 pm

Re: ATTN all Library Developers - Have a Tech Question

Postby enriquesilveti » Tue Feb 12, 2013 10:31 am

RJHollins wrote:Not that the whole Community isn't welcomed to respond [please do], but I would much appreciate hearing from the Library Developers, in particular, to shed light on this.

In my normal, rambling way :lol:

Let's talk EQ presets.

To start, many EQ libraries include full harmonic, clean [no harmonics], and often times the Pre-amp.

I would like to clarify on the strategy of implementing several presets in an FX chain that would 'most' closely mirror the 'original' harmonic character of 'said' EQ.

Please ... let us temporarily bypass the 'use your ears' ruling, as I would like to hear more on the 'technical' aspect.

Mentioned in library documentation is the suggested chain to be built from a single 'FULL Harmonic' instance, and the remaining plugs to use the 'Clean' presets. As paraphrased, 'this would represent the original harmonic content of the unit' [basic premise] :roll:

Further it's said that the 'Full' preset should be used for 'boosting'.

So the question that presents itself is:

What if we used all preset EQ's of only the 'Clean' for each instance, then follow that with one instance of the Pre-amp?

Not an major revelation, as this stems from the 'MOJO' technique, whereby an 'Algo' plug is fed into a Nebula Pre [example]. This very workable technique is very practical and sonically better than the algo alone.

To the specific interest [at least mine :)], to [as closely] replicate the sonic signature of a given piece of hardware. 2. To look at a possible strategy that may simplify the preset ordering, with 3, to configure the FXChain to represent a complete hardware representation. [if it has 5 bands, then we need 5 Nebs].

Would 5 'clean' Nebs fed to a Full Preamp fulfill the same as using 4 clean + 1 'dirty' :twisted:

AlexB ??? Michael ???? our beloved 'Other Developers ??? !!! Just looking for a constructive conversation/insight on things Nebula to helps expand or discover new awareness.

Thanks for taking the time to read ... let the education continue 8-)



Amplifier = non lineal generator
Passive filter = NOT non lineal generator
See the electronic schematic and make the routing.
Enrique Silveti.
Acustica Audio customer and technical support.

MBP 11.5 (i7-4870 | 16 GB | 512 SDD)
SP4 (i5-6300 | 8 GB | 256 SDD)
UFX | Lyra2 | USBPre2
VM U15 | VM Win10 | VM OSX 10.12
N4/NAT4 | SPX3 | RX5 | LN2C | Smaart8 | R5 | PT12 | PX10 | NIK5
User avatar
enriquesilveti
Expert
Expert
 
Posts: 2663
Joined: Sun Mar 28, 2010 9:00 pm
Location: Lodi | Madrid | Buenos Aires

Re: ATTN all Library Developers - Have a Tech Question

Postby RJHollins » Tue Feb 12, 2013 6:41 pm

enriquesilveti wrote:Amplifier = non lineal generator
Passive filter = NOT non lineal generator
See the electronic schematic and make the routing.


:shock: :?:

With a 4yr degree in EE, the terms defined are understood, Enrique. 'Wiring up' NEBULA's preset routing with capable software is easy enough.

Trying to understand your message?

Are you relating a 'Clean' preset to a 'passive filter' ?

If we speak strictly of NEBULA equalizers. These are the main libraries that can easily call for multiple instances at a time.

If we are to match the harmonic/non-linearities of the original hardware unit [within the limits of Nebula], would multiple, single kernel [clean] presets feeding into a multi kernel preamp not be an accurate representation ?

To be as clear as I can [not easy for me]. A single EQ preset that contains all kernels and harmonics would be no different than if I patched in the hardware and activated a single band.

But now, lets say we want 4 bands of EQ, all run in series for the same sonic source.

On the hardware, we'd just activate each section all within the same unit.

With NEBULA, we have the choice of having 4 separate units with full color. Or we can choose 4 'Clean' presets that contain the curve, then drive those into a full blown preamp.

What combination would best represents the original unit from the Nebula Users perspective ?

I know that these can be difficult topics to discuss. Subjectiveness is always a deciding factor ... as it should be! 8-) Graphs, charts, measurement, specification provide important documentation.

None of these would override a sonic decision. This is the 'Art'. We also have the 'Science', and this is the particular focus I'm asking about.

Apologies for the 'scattered' writing ... I appreciate those trying to follow this, and look forward to everyone's insight.
thx
i7-5820k, MSI X99A Plus, 16 GIG Ram, Noctua NH-D14, Win-7 Pro [64-bit], Reaper-64

NVC [Nebula Virtual Controllers]
RJHollins
Expert
Expert
 
Posts: 2633
Joined: Sun Mar 28, 2010 5:53 pm

Re: ATTN all Library Developers - Have a Tech Question

Postby enriquesilveti » Tue Feb 12, 2013 7:17 pm

Normally is not possible to sample active sections without the passive sections because both interact. Normally equalizer are serial so you can serial 2,3 or 4 Nebula instances depending the full EQ to emulate. If you have only one amplifier for those 2,3 or 4 EQ section you can use one 3K/5K/7K Nebula instance plus 2,3 or 4 of clean 1K instances. The 3K/5K/7K Nebula instance will emulate the EQ/phase curve plus non lineal distortion from the active section, the 2,3 or 4 of clean 1K Nebula instances will emulate the EQ/phase curve. So to understand how those passive filters are connected with the amplifier section you must see and understand the electronic schematic.
For example a pultec eqp1a is a serial passive filters network (http://www.recproaudio.com/diy_pro_audi ... ematic.jpg) plus a valve active stage (http://www.recproaudio.com/diy_pro_audi ... _schem.jpg) and the PSU can be transformed based or also valve based like the original vintage models.
Enrique Silveti.
Acustica Audio customer and technical support.

MBP 11.5 (i7-4870 | 16 GB | 512 SDD)
SP4 (i5-6300 | 8 GB | 256 SDD)
UFX | Lyra2 | USBPre2
VM U15 | VM Win10 | VM OSX 10.12
N4/NAT4 | SPX3 | RX5 | LN2C | Smaart8 | R5 | PT12 | PX10 | NIK5
User avatar
enriquesilveti
Expert
Expert
 
Posts: 2663
Joined: Sun Mar 28, 2010 9:00 pm
Location: Lodi | Madrid | Buenos Aires

Next

Return to Working with Nebula

Who is online

Users browsing this forum: No registered users and 5 guests