Guitar modeling has arrived... and it is good!!

[donreynolds]

Just downloaded Studio Devil amp modeler pro. WOW! Very nice Marshall sims in itself. Combine it with ReCabinet 3.0 and you have a smoking guitar set up to record with.

Check out the demo at:
http://www.studiodevil.com/products/" onclick="window.open(this.href);return false; at $140. and can get the CD for another 10 bucks for back up.

They have Recabinet demos already in the cabinet list but if you can get the ReCabinet 3.0 it is another ball game right there. YOU WILL SMILE

I like amplitube, but hey this is SO easy to use and absolutely rocks. CPU is not near as bad either

I played my Carvin Koa DC 200 through it, then a Michael Kelly Patriot hot rod and then a Michael Kelly Patriot Active limited.

My projects are going to rock some guitars from now on.

You can hold me to this...I will never mic another cabinet again. Sorry, I know there are a lot of purists out there but really, there is no need to do it any more unless you just want to. Technology has arrived for guitar modeling, just as Digital has arrived for DAWs. You have to be an oscilloscope to even come close to hearing a difference, if there is any??
I'll probably get roasted for this but...
I'm just sayin'.....

[James Steele]

It will be interesting to hear. My issues have primarily been how a sim responds to dynamics and touch. Also not sure how you create controlled feedback absent being in the room with an amp. I suppose you might be able to hold your guitar up to a loud studio monitor, but...

Anyway, it will be interesting to hear. One of my beefs has sometimes been how modeled guitar parts sounnd after doubletracking a rhythm part for example. Still I will give it a listen. Only way to really know I guess would be to A/B the same part on same guitar between the amp and the model.

I'll listen with an open mind. In the past though I've been disappointed with these simulations.

[donreynolds]

I can see how feed back might be a little more difficult but yeah getting close to a monitor my work if you are after some feed back. Some sims have feedback options. Now that you brought that up I will have to try that. I thought the studio devil pro was good byitself but combine it with recabinet 3.0 plugin or even just import their 2.0 impulses. You can import into studio devil or you can use Convolver to utilize the imported impulses. Now, I know that of course if you have the real thing that would be best to use, but you have to get it miked correctly and that takes time. Not to mention with the modeling you can change it on a whim "just to see". If you have a mix going you may find that what sounded good by itself does not work in the mix and can change it very quickly and experiment. And even throw down several different tracks with different amps. I know now many producer/ engineers are completely not using amps or at least supplementing recorded tracks with modeling now because it has gotten so good now.

[wylie1]

I've not tried modelers yet but there is a free one at the bottom of the page.
I'm still trying to find the time to play with iRig.

[Guitar Gaz]

I use Line 6 - Pod Farm2 at present - I really like it and have been using Pod's for a long while. Studio Devil I know - its okay quite good. They are all pretty good to practice with. For recording I am used to the Line 6 sound and have compared it with real amps - in a mix they sound pretty good to me and they allow you loads of fx and amp types you could not normally afford. Of course nothing is better than standing in front of a real valve amp - but these sims are really pretty good. I wouldn't dismiss them even if you are a valve/tube man - you might still find them useful.

[donreynolds]

I agree n the fact that IF you have the real thing that is going to give you the best likeness of course. I am just saying that the sims have gotten so good, that to compare them is very close not if you set the sim up right.

I just think that the pros of a sim far outweigh the cons now. Real similar to the analog vs digital arguement. We know how that turned out.
iI think if you take a good amp model software, combine it with a true speaker or cab impulse you will have a great tone. Then if you want more reality, set up an aux send and set up a room ambiance track with convoluted Reverb to blend in with it. By the way we use impulse reverb now mostly instead of real rooms because it is now so good.

Maybe I am just weird, but It is getting real hard, if not impossible for me to tell the difference any more between a real amp and cabinet versus a modeled one, especially in a mix.

Don't get me wrong either, I have had some nice amps and cabs in my time, ( still have on or two), but when Boss came out with the GT-8 and GT-10, I just use the amps for amplification and many times I just run through PA and monitors and have been quite satisfied when it is set up correctly.

But again if you are worried about recording 96Khz verses 48Khz , then a real amp might be the best route, but for general listening, If you can come very close to the real thing, I don't see a problem. But each to his own. A bouquet is made up of many different flowers but is is still pretty, eh?

[dewdman42]

How has recabinet been working out for you? its on sale 50% off right now, thinking about picking it up. Trying to decide between Recabinet and redwirez. Anything you want to comment about your experience after a year has gone by?

Myself I will probably use the direct recording outs from my Mesa 20/20 and run into cab simulator for recording at home, and even for practicing.

I hear what James is saying, there is still something to be said for pushing a lot of air around with a real cab, but man it sure is getting close, and the cab sim let's me try a lot of other cab sounds.

[Prime Mover]

I honestly don't think we're near being done with digital amp engineering until we completely abandon impulse responses, and have dynamic, complex physical modeling amps that blow everything out of the water.

The thing that bothers me a little about IRs or samples is that they can never be more than their original recordings. You always have to rely on there having been built a real version of that sound. This stunts the creative engineering of new sounds, because we're spending all our energy just trying to replicate something else. With algorithms, you can do anything, and aren't tied to trying to copy something else.

[timriley]

http://www.youtube.com/watch?v=sandnFGWpdo

Think this is what you're talking about Prime Mover. Haven't actually tested this and i'm not really a proficient enough guitarist to comment but it sounds interesting.

[donreynolds]

Recabinet has addressed the very fact you were talking about involving the speaker models missing the dynamics of "pushing air" that real speakers have. The are now simulating the speaker movement.The new update has examples of it. It sounds pretty good.

[Gravity Jim]

I was a huge defender of amp sims, and still use them in demos. But my experience has been that they simply won't do a good job of a sound that is very important in my work: that not-clean, not-breaking-up smooth, liquid overdrive associated with jazz/blues players like Larry Carlton and Robben Ford.

I was always able to get a passable clean sound from them (although nothing as "interesting" as my Egnater amps deliver), and a couple of them did very nice "hairy" overdrive sounds... But that "Dumble in a box" sound eluded me.

And now that I have my mic locker dialed in for this, it's almost as easy to fire up an Eggy and a ribbon mic next to an SM57 and get the sound as it was to plug into a tube pre and launch a modeler. Almost.

I'd love to play a Fractal AxeFX , as I've heard good things about them. But until someone really nails it down, I'm sticking with my real amps.

[magicd]

I honestly don't think we're near being done with digital amp engineering until we completely abandon impulse responses, and have dynamic, complex physical modeling amps that blow everything out of the water.

The thing that bothers me a little about IRs or samples is that they can never be more than their original recordings. You always have to rely on there having been built a real version of that sound. This stunts the creative engineering of new sounds, because we're spending all our energy just trying to replicate something else. With algorithms, you can do anything, and aren't tied to trying to copy something else.

Impulse responses are not used in something like the MOTU Custom 59 amp model. What we do with an amp model is called Physics Modeling. The original schematic is used to recreate the circuit as a DSP process. We then take the real thing and send signals through it to compare the DSP model against the real thing.

Impulse responses are used for something like a speaker or microphone model. We use impulse responses for Live Room G for example.

I've been a guitarist for a long time. I've switched over to computer based processing for my live rig. I wouldn't have done it if there was any compromise in sound or feel. The most challenging part of the switch was to get all the dynamics and levels worked out. I'm also generating VI keyboard sounds with my rig, so I had to balance the guitar sounds with the keyboard sounds. The rig works great for me now.

As for speakers and feedback and so on, what about when the guitar player plays through a real amp, but sits in the control room while the amp is in the recording room? What about the guys who play on stage with in-ear monitors and put their amps in a box somewhere? It's the same deal. When I'm playing live, I send the output of my UltraLite to the PA, and to a power amp and a 2 X 12" Fender extension cabinet. I have no problem getting amp response with this setup.

And here is a tease for you: DP8 has a Vox AC30 and a Boogie Dual Rectifier (along with a whole lot of other cool stuff). You will dig it the most!

Dave

[mikehalloran]

When I saw the subject, I was hoping to read about guitar modeling.

Although all of these amp/speaker modeling descriptions sound wonderful, what is really out there for those of us who don't play guitar (in my case, anymore)? I have been disappointed with everything I have tried so far.

[HCMarkus]

Always fun to hear from you Magic Dave!

[dewdman42]

I think there is some confusion here. Recabinet does not model the amps. Its a "cab" simulator, nothing more. You still have to have a good source going into it, coming from physically modeled or real amplifiers.

I did end up plunking down $50 for it last night, that is a pretty sweet deal and I'm sure I will use it plenty. I may even run my mesa 20/20 tube amp through it since it has direct outs, so that I can get all the tubeness and experiment with different software cabinets.

By the way, Amplitube, Schuffman, Revalver POD and all the rest are all physically modeled amplifiers, they are not "sampled". They are getting closer and closer. However, one area that pretty much all of them did not model well is in the area of the cabinet simulation. The early rack mount guitar preamps that came out in the 80's and early 90's simply used a low pass filter to simulate the effect that speakers have on the sound. I think we can say that is a very poor excuse for a speaker cab. But people wanted more precision, so then they started created IR's of various different cabs, which is a lot closer in terms of recreating the sonic print a cab gives to sound, but in a static way. Again remember we are talking only about cab. The amp and preamp and pedal modeling provide a ton of other dynamics.

The folks at Recabinet are now taking it a step even closer with this new dynamic feature which is in fact a step beyond static IR and starts to get into more dynamics which some people, including myself, may have felt was missing.

I don't think we have seen the end of development for both amp modeling as well as cab modeling. One poster may be right, until someone literally models speakers cabs using pure physical modeling, it might never be quite there for the speakers. To really model those things would be quite a bit more complicated then modeling amps and preamps. You would literally have to model the vibrations of paper cones in metal speaker frames, attached to wood boxes of different dimensions and different kinds of wood, etc. That is what physical modeling is. With the amps you are modeling the way electrical signal flows through electronic components, but the acoustics of speaker enclosures are 100x more complicated to model that way, not to mention the different microphones which each have their own peculiar physical attributes and then the room the cab is in, etc.. all of those physical acoustics contribute to the way a mic'd cab sounds. Its WAY more complicated then physically modeling an electronic device. That is why IR's have been a reasonable trade off perhaps.

I tend to doubt they will ever do it because its a diminishing return. The existing cab simulations are quite close enough for most people, for recording, especially if they are able to simulate some dynamics that rival that of real hardware cabs. There is still room for improvement, I agree, but like some others here, I see the future as being a completely modeled one and even in the present day where modeling of guitar cabs is not perfect, overall I see far more advantages to using modeled guitar rigs then avoiding modeling altogether.

For live playing there is still nothing like the feeling of standing in front of your 4x12 and feeling the air being pushed. I've never played or heard a modeler that rivals that yet in terms of the "feeling". When I run my Vox tonelab through a tube amp and speaker, the strings just respond in a different (and better) way. The complete instrument is a guitar, cable, pedals, amps, speaker. The entire chain represents "the instrument", and somehow when I play through my tube amp and speakers, it feels more dynamic, more expressive, more musical, more fluid. There is no question that *I* experience something more magically playing through that setup then through Amplitube which sounds sterile in comparison somehow.

Even that Axe thing, it can do some amazing things, but I can hear the static-ness in it also. That being said 99.99% of the listening audience can't hear it. And the convenience factor of working with modelers, and being able to re-amp to your hearts content is really really valuable.