Login

Timed mode how much ms?

Tips & tricks, working results, technical support

Re: Timed mode how much ms?

Postby enriquesilveti » Thu Aug 08, 2013 9:50 am

See this. Is the IR of H1 of a preamp sampled with NAT Platinum 1.3.513 at 96 kHz and 32 bits floats.
Attachments
IR.jpg
IR.jpg (187.74 KiB) Viewed 993 times
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: Timed mode how much ms?

Postby RJHollins » Thu Aug 08, 2013 9:14 pm

Given the above IR ...

What would the settings be inside NEBULA's Kern page?
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: Timed mode how much ms?

Postby Cupwise » Fri Aug 09, 2013 4:48 am

it's because the bass cycles extend beyond the rest of the impulse, in plenty of the samples taken from different hardware. i think the length you need to accurately capture the bass is dependent on the hardware though, because i've seen and made programs with really short lengths for h1 that still had flat bass going down to 10hz (my rayphlex comp only has 2ms for h1 and is flat down to that range). which is weird because that's much less than the length of a single low freq bass cycle, so i think it has to do more with how the hardware handles the bass than it does with the length of the bass cycles. i don't understand how nebula can reproduce a low bass response from only 2ms, but it does/can.

if you have NAT generate a tone sweep offline, and don't run that through anything and just make a program from it without touching it, the impulse you get out of that is very very quick. it fades to digital silence pretty much instantly, and it still has a good bass response. but in some cases, some hardware requires more time for an accurate bass response. i think it's probably something to do with the hardware introducing phase shifts on the low bass, so that extends the data for the bass out beyond the typical impulse length. so you have bass cycles going out up to 50ms to 100ms or even beyond that.

but if the impulses generated by NAT from a hardware sample actually have bass content going out that far after the peak of the impulse, and it's not accounted for in the program's length, you may lose bass from the response. and this can happen with any kind of program, including comps which are typically set to 10ms.
Cupwise
Expert
Expert
 
Posts: 773
Joined: Tue Nov 23, 2010 2:03 am

Re: Timed mode how much ms?

Postby RJHollins » Fri Aug 09, 2013 7:29 am

k ... pushing for some definitive answer [to keep down confusion].

Hi CupWise, we know you've had extensive testing while developing your libraries [given].

Since no one but the dev would have access to the specific hardware unit sampled, we Users have only limited reference points to draw from.

If you can clarify ... if you release a library that has, say, 20ms or 50ms in the original kernel setting ... and then, say the User switches it to TIMED ... are the original ms Lengths still valid to [at least] match the response intent of the original library ? Or does the TIMED mode use a different reference point, and thereby, a different ms length must be determined ?

Any clarification/ insights would be helpful ...

Thanks to all. 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: Timed mode how much ms?

Postby enriquesilveti » Fri Aug 09, 2013 11:06 am

I'm using 20 Hz HPF in NAT. My DA has also a hardware 20 Hz HPF! You can not sample something that your DA can generate plus anyone can hear at 20 Hz? The most important thing about TIMED is your are using direct convolution. No one company is using direct convolution for audio post-production processing.
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: Timed mode how much ms?

Postby Tim Petherick » Fri Aug 09, 2013 5:18 pm

Maybe enrique Can chime in But I thought that nebula will play a flat frequency response on the low end if the kernals are not long enough to represent the actual capture. So if you take a compressor it will be accurate down to 100hz at 10ms kernal length either on timed or freqd. But will flatten out along the low end within reason.


Just uploading some graph's
User avatar
Tim Petherick
Expert
Expert
 
Posts: 1352
Joined: Sat Apr 17, 2010 4:07 pm
Location: Bath , Uk

Re: Timed mode how much ms?

Postby Tim Petherick » Fri Aug 09, 2013 5:42 pm

timp wrote:I came to the assumption that

50ms = accuracy down to 20hz
100ms = accuracy down to 10hz

basically 1sec =1hz cycle 500ms=2hz cycle and so on

but you don't have to worry about it if the frequency response is flat down there. That's why most compressors still sound good at 10ms. If you have a eq you need it to be above 50ms really.....

2.2ms I'm guessing is down to 454hz

If you look at the transient loss page I stated that you need to mod the xml to go above 30ms.



The plots were the same for both players. Preset was studer tape

50ms

Image

10ms
Bump on 100hz

Image
[img]

2.2ms
Bump just above 500hz
Image

As you can see it does seem that 2.2ms is in fact close to 454hz. Then flattens out. You can see that bump move.So in a compressor as long as its reasonable flat it should not matter so much. Although you may well miss the filter effect in the lows of analog hardware.
User avatar
Tim Petherick
Expert
Expert
 
Posts: 1352
Joined: Sat Apr 17, 2010 4:07 pm
Location: Bath , Uk

Re: Timed mode how much ms?

Postby Scottxx » Fri Aug 09, 2013 6:10 pm

So I go with RJHollins...

Again, if a cabinet preset with 2ms by default, turned to TIMED mode without touching anything. Will that loose bass response or it will be the same originals frequencies but with direct convolution? It sounds better, more cristaline to my ears.
Scottxx
User Level 0
User Level 0
 
Posts: 6
Joined: Sat May 29, 2010 1:56 am

Re: Timed mode how much ms?

Postby Tim Petherick » Fri Aug 09, 2013 6:16 pm

Scottxx wrote:So I go with RJHollins...

Again, if a cabinet preset with 2ms by default, turned to TIMED mode without touching anything. Will that loose bass response or it will be the same originals frequencies but with direct convolution? It sounds better, more cristaline to my ears.


Well if it sounds good, then it is good.

2ms Is only going to get the highs right , then as you can see probably flatten out as seen in the graphs. So no color on the low end but perhaps that's why you like it?

Maybe look at the preset you have in a vst analyzer
User avatar
Tim Petherick
Expert
Expert
 
Posts: 1352
Joined: Sat Apr 17, 2010 4:07 pm
Location: Bath , Uk

Re: Timed mode how much ms?

Postby Cupwise » Sat Aug 10, 2013 12:26 pm

RJHollins wrote:If you can clarify ... if you release a library that has, say, 20ms or 50ms in the original kernel setting ... and then, say the User switches it to TIMED ... are the original ms Lengths still valid to [at least] match the response intent of the original library ? Or does the TIMED mode use a different reference point, and thereby, a different ms length must be determined ?

yes. the same time is the same time. it's the same for either timed or freqd. if it's enough to accurately represent the bass in one, then it will be in the other. also, if you increase the length beyond what it's already set to by default, you can't hurt anything, only possibly get more accurate bass response, with the exception of the prog rate changing due to your increasing the length. if prog rate changes then the program will act a little differently. if you send the kern to timed before increasing, then the prog rate shouldn't change if you increase the length.


timp wrote:


As you can see it does seem that 2.2ms is in fact close to 454hz. Then flattens out. You can see that bump move.So in a compressor as long as its reasonable flat it should not matter so much. Although you may well miss the filter effect in the lows of analog hardware.

the graphs are really just verifying that with shorter kern lengths, nebula gives artificial filtering which isn't accurate to the hardware. so if you used a comp or any other program with short length on a full mix or anything with content in the low freqs, you're going to get probably unwanted filtering of the bass which isn't related to the hardware.

further, it's entirely possible that if that program had it's length extended out to around 100mhz, the low end would be even flatter than the 50ms graph.

some hardware actually cuts bass when you use it. but if you get a graph from vstanalyser that starts cutting more bass when you lower the length of the h1 kern, than what the response was before you lowered it, then you are getting less accuracy to the hardware and an artificial filtering effect due to the fact that you are cutting off bass cycles from the impulses. it might still sound ok for what you are using the program for, but plenty of people use compressors on full mixes and so i'd say that this is a flaw and kind of needs to be dealt with by having longer h1 lengths using timed.
Last edited by Cupwise on Sat Aug 10, 2013 1:54 pm, edited 1 time in total.
Cupwise
Expert
Expert
 
Posts: 773
Joined: Tue Nov 23, 2010 2:03 am

PreviousNext

Return to Working with Nebula

Who is online

Users browsing this forum: No registered users and 2 guests