2 Pages V   1 2 >  
Reply to this topicStart new topic
> Discussion On Modelling
Mertay
post Apr 12 2019, 06:46 PM
Post #1


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



https://www.youtube.com/watch?time_continue...p;v=p9rKZJ2m2TA


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 13 2019, 08:12 PM
Post #2


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Thanks for the share!!! I was reading the chat comments and it seems that no matter how good modellers get, there are just some folks that are not wanting to believe it, no matter what. Some folks saying a kemper doesn't sound like the amp, etc. I keep thinking this will change at some point. At this point, I'm thinking it will never change smile.gif

QUOTE (Mertay @ Apr 12 2019, 01:46 PM) *
Go to the top of the page
 
+Quote Post
Mertay
post Apr 14 2019, 12:16 AM
Post #3


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



QUOTE (Todd Simpson @ Apr 13 2019, 07:12 PM) *
Thanks for the share!!! I was reading the chat comments and it seems that no matter how good modellers get, there are just some folks that are not wanting to believe it, no matter what. Some folks saying a kemper doesn't sound like the amp, etc. I keep thinking this will change at some point. At this point, I'm thinking it will never change smile.gif


His comment on how it sits in the mix got me most. Saying, if everythings digital (experience from Garbage-the band I guess) there's no problem but with acoustic instruments it becomes harder work. Friedman seems to have a similar experience too.

Though he's also not 100% sure. First commenting the axe-fx got scary similar to the real-amp side-by-side, but in a mix disappearing (with same console settings) and some-what blames this the axe-fx not replicating the further harmonics (ones we can hear but can't identify) as good as the real amp being the problem.

This is actually very hard to discuss with guitar players (I tried it) as this is really diving deep into the audio engineering realm. They keep saying axe-fx sound great (as if we don't know that tongue.gif) but no one else matched an amp and compared both inside a mix. So the argument is if a mic.ed amp is easier to setup in a mix compared to an axe-fx.



--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 14 2019, 03:06 AM
Post #4


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



A real cab in a real room does generate super and sub harmonics. These are bits we can't hear that impact the bits we can hear. This type of thing can be helped by oversampling in plugins like the UA series that use hardware to assist the cpu. Their guitar plugins, such as the tube friedman plugin, are darn good. It does come down to the engineer in the end. Making the tracks work together and sound as good as possible. Engineers that came up with plugins/modellers, are accustomed to the quirks involved in using them hopefully. Don't get me wrong, I LOVE a full stack turned up all the way. It's got it's own unique sonic signature and has soooo much sound that it has to be tamed in a mix. That being said, it's just not practical as a home recordist to use a full stack at full volume in a home recording setting. So I'm a big advocate on trying to get a live sound using non live/digital gear smile.gif


QUOTE (Mertay @ Apr 13 2019, 07:16 PM) *
His comment on how it sits in the mix got me most. Saying, if everythings digital (experience from Garbage-the band I guess) there's no problem but with acoustic instruments it becomes harder work. Friedman seems to have a similar experience too.

Though he's also not 100% sure. First commenting the axe-fx got scary similar to the real-amp side-by-side, but in a mix disappearing (with same console settings) and some-what blames this the axe-fx not replicating the further harmonics (ones we can hear but can't identify) as good as the real amp being the problem.

This is actually very hard to discuss with guitar players (I tried it) as this is really diving deep into the audio engineering realm. They keep saying axe-fx sound great (as if we don't know that tongue.gif ) but no one else matched an amp and compared both inside a mix. So the argument is if a mic.ed amp is easier to setup in a mix compared to an axe-fx.
Go to the top of the page
 
+Quote Post
klasaine
post Apr 15 2019, 04:31 PM
Post #5


GMC:er
*

Group: Members
Posts: 3.455
Joined: 30-December 12
From: Los Angeles, CA
Member No.: 17.304



What I've noticed specifically with recording is that if all the tracks have been done "in the box", that my real amp'd guitar tone seems to not fit the vibe most of the time. My general impression is that it's just too live or too spacious (if that makes any sense). A lot of engineers and producers and the artists initially would like to have a big or live, analog 'rock' (ish) guitar sound on their tracks. You know, use all those cool mics they have plugged into the high-end pre amps and comps. Gotta justify the expensive gear, right?
I always suggest that they also take a DI'd clean signal so that they can re-amp and/or model as needed later - just as a safety wink.gif.
75% of the time, the DI'd track ends up being the one used.

Amp tones that do tend to work or fit into these types of mixes are usually cleaner parts - small amp, close mic'd in a small dead room with one SM57.

This post has been edited by klasaine: Apr 16 2019, 02:52 AM
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 16 2019, 07:06 AM
Post #6


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Good point per usual!! Guitar can take up a LOT of sonic space. Engineers typically have to use shelving on both ends high/low to get the guitar to sit right and not step on other things. A digital guitar often fits in better with the other digital tracks. One big analogue track can easily stick out. If it's the vocal, thats fine and dandy, if it's one of the instruments on a non instrumental track, then it's not so good. Always a good plan to grab a DI no matter what one is doing. I've seen that one thing bite more people in the arse than anything else.
QUOTE (klasaine @ Apr 15 2019, 11:31 AM) *
What I've noticed specifically with recording is that if all the tracks have been done "in the box", that my real amp'd guitar tone seems to not fit the vibe most of the time. My general impression is that it's just too live or too spacious (if that makes any sense). A lot of engineers and producers and the artists initially would like to have a big or live, analog 'rock' (ish) guitar sound on their tracks. You know, use all those cool mics they have plugged into the high-end pre amps and comps. Gotta justify the expensive gear, right?
I always suggest that they also take a DI'd clean signal so that they can re-amp and/or model as needed later - just as a safety wink.gif .
75% of the time, the DI'd track ends up being the one used.

Amp tones that do tend to work or fit into these types of mixes are usually cleaner parts - small amp, close mic'd in a small dead room with one SM57.
Go to the top of the page
 
+Quote Post
Mertay
post Apr 16 2019, 06:51 PM
Post #7


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



I wonder if thats why I like running processors through preamps.

Noise is a downside but while even cheapest analog saturation from typical soundcard preamps increases (the feel) of dynamics but even best of plug-in saturators compress further.


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 17 2019, 02:41 AM
Post #8


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Yeah, noise can be a problem. Running any signal through a tube preamp can warm it up and make it a bit more "round" which is sometimes exactly what is needed smile.gif

QUOTE (Mertay @ Apr 16 2019, 01:51 PM) *
I wonder if thats why I like running processors through preamps.

Noise is a downside but while even cheapest analog saturation from typical soundcard preamps increases (the feel) of dynamics but even best of plug-in saturators compress further.
Go to the top of the page
 
+Quote Post
Kristofer Dahl
post Apr 17 2019, 09:43 AM
Post #9


GMC Founder & Rocker
Group Icon

Group: GMC Instructor
Posts: 16.251
Joined: 15-August 05
From: Stockholm, Sweden
Member No.: 2



Interesting!

I had the honor to play a hand built "Pure Tone Amps" for a few months - and after that I have slightly changed my view on digital.

My theory is that the kemper does replicate any amp as perfect as you need it to - dynamic/eq/distortion wise. However I get the impression I am actually hearing it's just a snapshot (that's what a profile is : a snapshot of the real amp sound) - like a short sample I am playing over and over. Whereas the tube amp has natural variation that exceed a ~1 second sample.

I did not watch the whole video - but it sounds bit weird to me that kemper would not be able to replicate all overtones - as those can be measured digitally and the developers would easily see if the profiling process was missing some frequencies.

Perhaps the problem they experienced was rather due to dynamics not being replicated on some frequencies? That's harder to hear and can def effect where the sound sits in the mix. You might be hearing and seeing (on a graph) lots of 10k for example - however the attack might be kinda slow on those frequencies which in turn will not yield satisfying results.


--------------------
Go to the top of the page
 
+Quote Post
Mertay
post Apr 17 2019, 10:04 AM
Post #10


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



QUOTE (Kristofer Dahl @ Apr 17 2019, 08:43 AM) *
...


Thats why I left using ir reverbs long ago, on single instrument they sound fantastic but spread-ed in a mix a "static/boxness" starts to occur I really dislike. Ir-cab.s takes the freq. response and can't replicate speaker movement and coloration, its a good tech. but in detail has its weaknesses.

The dowside of sampling is the very upper overtones can't be accurately captured by any machine due to the units self-noise. These overtones get very low on level and blends too much with the noise character so usually filtered out. But most are accurate and there may be an AI that guess and replicates the missing ones for a more natural feel.


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 18 2019, 06:55 PM
Post #11


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



In a mix a good plugin can work well imho. I've been using the OVERLOUD plugins for years and like them better than any live recording of an actual amp that I've ever managed to get. If I had a big empty room that I could crank up a full stack in, then I"d probably go for the stack. Sadly, I don't have that so having tried just about every other option, I ended up where I started back with the overloud plugin. The amp sims are great, I can use 2 cabs and four mics, custom IR if needed, and great reverbs that do a very good room sound which I dig smile.gif

QUOTE (Mertay @ Apr 17 2019, 05:04 AM) *
Thats why I left using ir reverbs long ago, on single instrument they sound fantastic but spread-ed in a mix a "static/boxness" starts to occur I really dislike. Ir-cab.s takes the freq. response and can't replicate speaker movement and coloration, its a good tech. but in detail has its weaknesses.

The dowside of sampling is the very upper overtones can't be accurately captured by any machine due to the units self-noise. These overtones get very low on level and blends too much with the noise character so usually filtered out. But most are accurate and there may be an AI that guess and replicates the missing ones for a more natural feel.
Go to the top of the page
 
+Quote Post
Mertay
post Apr 18 2019, 08:34 PM
Post #12


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



QUOTE (Todd Simpson @ Apr 18 2019, 05:55 PM) *
...


Thats interesting cause TH-U's respire tech. is simply saturation after the cab. smile.gif


--------------------
Go to the top of the page
 
+Quote Post
Madfish
post Apr 19 2019, 06:58 PM
Post #13


Learning Rock Star
*

Group: Members
Posts: 89
Joined: 30-July 12
From: Poland
Member No.: 16.357



Sampling rate of 2f is sufficient to represent the signal of frequency f. See sampling theorem if you want the math.
Unless you go with some low sampling rate (below 44.1 kHz), you should be able to capture everything within human hearing range.

QUOTE (Mertay @ Apr 17 2019, 09:04 AM) *
The dowside of sampling is the very upper overtones can't be accurately captured by any machine due to the units self-noise. These overtones get very low on level and blends too much with the noise character so usually filtered out. But most are accurate and there may be an AI that guess and replicates the missing ones for a more natural feel.

Go to the top of the page
 
+Quote Post
Mertay
post Apr 19 2019, 09:37 PM
Post #14


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



QUOTE (Madfish @ Apr 19 2019, 05:58 PM) *
Sampling rate of 2f is sufficient to represent the signal of frequency f. See sampling theorem if you want the math.
Unless you go with some low sampling rate (below 44.1 kHz), you should be able to capture everything within human hearing range.


Hey madfish, I was referring to sat./dist. sampling which is a very different matter.


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 19 2019, 11:19 PM
Post #15


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Respire is separate from the cab loading section smile.gif I use respire but that' snot what I'm talking about. You can pick two different cabs and turn respire off. You still get the cab simulation. You can use a stock cab on one and load your own IR on the other which I've done several times. Then you get four mics to choose from which you can place on the cone. It's the most faithful plugin I've messed with other than the FORTIN plugins which sound great but are just guitar which I makes things a bit more of a pain in the neck. The overloud plugins have pedals/rack gear/heads/ dual path processing, IR support, and now rig sampling in one plugin. Oh yeah, and it has respire smile.gif

QUOTE (Mertay @ Apr 18 2019, 03:34 PM) *
Thats interesting cause TH-U's respire tech. is simply saturation after the cab. smile.gif
Go to the top of the page
 
+Quote Post
Mertay
post Apr 19 2019, 11:37 PM
Post #16


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



QUOTE (Todd Simpson @ Apr 19 2019, 10:19 PM) *
...


What do you think about what respire adds/helps guitars in the mix? I demo's at the time but don't have it now.


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post Apr 20 2019, 12:20 AM
Post #17


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Respire adds a LOT of bass to the signal. As a result, it can make it really boomy in your mix. Often too boomy. It's intended as a practice tool mostly. If one were to use it on a track, you have to pre eq the signal using your ears to account for the respire. It's a bit more compressed, and way more bassy so you gotta reduce the bass before it hits the respire. Most folks just turn it off for recording.
As for oversampling, it's true that according to the nyquist theory 44.1k gets all the bits we can hear. However, sampling beyond this allows us to capture bits that we can't hear which actually impact the things we can hear. In a real instrument, there are overtones that are not audible that impact the tones that are audible in suble ways. If we sample at a crazy high rate, we allow capture of the bits we can't hear to allow them to impact the bits we can hear. If this had no effect at all, we would not see sampling rates at 192 khz, as all it would do is make a bigger file. It does make a bigger file, but it makes an objectively better sounding file as well assuming a quality source. So there's that smile.gif

But yeah, for home use 44.1k is usually plenty.

Todd
QUOTE (Mertay @ Apr 19 2019, 06:37 PM) *
What do you think about what respire adds/helps guitars in the mix? I demo's at the time but don't have it now.
Go to the top of the page
 
+Quote Post
Madfish
post Apr 22 2019, 01:22 PM
Post #18


Learning Rock Star
*

Group: Members
Posts: 89
Joined: 30-July 12
From: Poland
Member No.: 16.357



QUOTE (Todd Simpson @ Apr 19 2019, 11:20 PM) *
As for oversampling, it's true that according to the nyquist theory 44.1k gets all the bits we can hear. However, sampling beyond this allows us to capture bits that we can't hear which actually impact the things we can hear. In a real instrument, there are overtones that are not audible that impact the tones that are audible in suble ways. If we sample at a crazy high rate, we allow capture of the bits we can't hear to allow them to impact the bits we can hear. If this had no effect at all, we would not see sampling rates at 192 khz, as all it would do is make a bigger file. It does make a bigger file, but it makes an objectively better sounding file as well assuming a quality source. So there's that smile.gif

But yeah, for home use 44.1k is usually plenty.

Todd


All those overtones (above Nyquist frequency) would create aliases in the output signal. These alias frequencies would be very much within the audible range. That is where oversampling might help you out.


QUOTE (Mertay @ Apr 19 2019, 08:37 PM) *
Hey madfish, I was referring to sat./dist. sampling which is a very different matter.


Hey. In that case I missed your point. What is distortion/saturation sampling?
Go to the top of the page
 
+Quote Post
Mertay
post Apr 22 2019, 06:26 PM
Post #19


GMC:er
Group Icon

Group: GMC Senior
Posts: 4.196
Joined: 27-May 13
From: Turkey / izmir
Member No.: 18.294



Pretty much what kemper is doing but my experience is based on this plug-in more (though haven't been using it in a long while) https://www.acustica-audio.com/store/t/nebula , aside ir also it can replicate drive character.


--------------------
Go to the top of the page
 
+Quote Post
Todd Simpson
post May 20 2019, 05:53 AM
Post #20


GMC:er
Group Icon

Group: GMC Instructor
Posts: 18.500
Joined: 23-December 09
From: Atlanta, Georgia, USA
Member No.: 8.794



Rates up to 192khz don't introduce any artifacts I'm aware of? Do you have any information on this? Wads of interfaces will let you record at 192khz even though folks don't usually do it as you have to dither down to 44.1khz at some point.
QUOTE (Madfish @ Apr 22 2019, 08:22 AM) *
All those overtones (above Nyquist frequency) would create aliases in the output signal. These alias frequencies would be very much within the audible range. That is where oversampling might help you out.




Hey. In that case I missed your point. What is distortion/saturation sampling?
Go to the top of the page
 
+Quote Post

2 Pages V   1 2 >
Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 


RSS Lo-Fi Version Time is now: 23rd August 2019 - 06:12 AM