Jump to content

Is the Variax more of a profiler than a modeler?


Recommended Posts

A modeler uses more studying what they're replicating and then applying that info to program something to sound like what they're recreating.

 

A profiler uses less of the person gathering info, and the data gathering is more left to the processor through algorithms and convolution techniques instead of someone studying the thing they're recreating and then making the data themself.

 

 

So, since the Variax is highly suspected to be using convolution and sensory techniques to capture algorithms of pickup and body frequency responses, wouldn't the Variax be more of a profiler than a modeler?

In theory, it's using closely the same techniques as a Kemper would go through.

 

What do you guys think?

  • Upvote 1
Link to comment
Share on other sites

There were lots of articles about Line6's modeling methods when they first came out with the Pods.  I think they used the same techniques to characterize guitars as they did on amps. 

 

Someone said they used some type of acoustic sensors. I'm not sure exactly how it would work but if it's any technology that analyzes the frequency response and characteristics of those guitars and forms them into IRs then it's doing closely what a Kemper would be doing to an amp.

Link to comment
Share on other sites

Have to ask Dave Fruehling

http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=1265136

 

http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1265125&filter%3DAND%28p_IS_Number%3A28299%29

http://pedaltech.blogspot.com/

 

0204frejob01.jpg

 

 

 

Dream Jobs 2004

Dave Fruehling: His Job Rocks!
Age:
32
What He Does:
Designs digital guitars that model classic guitars and related products
For Whom:
Line 6
Where He Does It:
Agoura Hills, Calif.
Fun Factors:
Plays with priceless guitars and amplifiers; works surrounded by people with musical talent; sees his "baby" in the hands of rockstars (and gets to meet some); knows his daughter thinks his job is cool; gets calls from his grandmother when she sees one of his guitars on TV
Rock stars are thrilled to meet him. Teenage boys want his autograph. But Dave Fruehling is not some kind of longhaired pop star; he's a shorthaired electrical engineer, a mild-mannered systems architect at Line 6, the electric guitar and amplifier company in Agoura Hills, Calif.
Fruehling didn't set out to be an EE. No way; his dad was an EE--a long-time employee of Motorola Inc.--and Fruehling was into youthful rebellion. Heading off to college in 1989, in his 1976 Pontiac Firebird with Rush blasting on the CD player and an electric guitar on the back seat, he had set his sights on being a rock star and getting his picture on the cover of Guitar Player magazine.
But he soon realized that as an instrumentalist, anyway, he would never find his way into the pantheon of guitar gods. His grades as a music major at the University of North Florida in Jacksonville were dismal. Before long he was spending more time tinkering with electric guitars than playing them. This was nothing new: a week after he got his first guitar at age 11, he took it apart; during high school, he spent some 20 hours a week building sound-shifting circuits described in his dog-eared copy of Craig Anderton's Electronic Projects for Musicians.
In 1993, Fruehling bowed to the inevitable--maybe it was in his genes--and switched his major to EE.
From that point on, he was a straight-A student. "Once I focused on engineering, it became my complete passion," he recalls.

 two-year stint after graduation at hard-disk maker Seagate Technology LLC in Simi Valley, Calif., gave Fruehling engineering experience and enabled him to see the business card (on the desk of a co-worker) of the vice president of engineering at Alesis. Fruehling perked up; he knew that Alesis, in Santa Monica, Calif., put out music-related systems. The timing couldn't have been better: Alesis was hiring engineers to design a hard-disk recorder, and Seagate had announced that it was leaving California.
From Alesis he soon moved on to his present company, Line 6. Formerly Fast Forward Designs, it had consulted for Alesis but was also making its own products--guitar amplifiers, which lined up perfectly with Fruehling's passion.

At Line 6, Fruehling developed the bass guitar version of the POD, the company's flagship project. The POD is a kidney-bean-shaped signal-processor-based device that makes transistor-based guitar amplifiers sound exactly like vacuum tube amplifiers. For Fruehling, working on the POD was a dream come true. After all, to get the modeling down right, he had to obtain and test a dazzling variety of pathbreaking vintage amplifiers (still scattered today throughout the company).
After a brief stint researching the feasibility of building a USB (for Universal Serial Bus) guitar-computer interface (a project that was handed to another team), Fruehling, along with fellow senior design engineer Pete Celi, took on a blue-sky research project. The basic idea was to develop an electric guitar that, through massive digital signal processing, could convincingly mimic the unique sounds of the most legendary electric and acoustic guitars. It would be almost like taking an ordinary violin and giving it a switch that could let it sound not only like a Stradivarius but also a Guarneri or a Ruggieri, depending on your mood.
"When we started," Fruehling recalls, "we had no idea what we were going to make. Were we going to make a guitar? A box you plug a guitar into? An acoustic pickup that doesn't plug into the guitar at all?"
"It wasn't like an amp," he adds. "You can't just plug in a signal generator; you can't create a controlled signal by plucking the strings. And if it could be done, could it be done at a reasonable price?"
The project took two years. The result is the Variax, an electric guitar that looks fairly unremarkable, except for one extra knob that has the names of classic guitars. It models 50 historic guitars, faithfully capturing all their beloved quirks, like the distinct twang of the Fender Stratocaster or the singing sustain of the Gibson Les Paul. The product came out in November 2002, and more than 10 000 have been sold at an average price of US $1000. It is being played by a growing cadre of stars like Pete Townsend, Steve Howe, and Joe Walsh, along with a host of ordinary folks.
These days, Fruehling hobnobs with guitar gods who come to his office or invite him backstage to quiz him about the Variax. He has met several of his idols, including Eddie Kramer, the producer/engineer on Jimi Hendrix's albums, and Craig Anderton, the author of the guitar projects book that figured so prominently in his high school days. He gets a thrill when he sees guitar players on TV with his guitar in their hands.
Each morning, Fruehling spends an hour or two working on a computer interface for the Variax. In the early afternoon, he consults on a variety of internal projects, before getting together with Celi to make plans for the evolution of the Variax. Sometime during the day, he'll play a guitar for an hour or so, sometimes in the guise of testing, sometimes just for fun.
When he wants to jam, there's no shortage of partners; the majority of Line 6's 200 employees play. Many offices contain mixers and high-quality amps, along with a variety of guitars. Fruehling usually has half a dozen in his.
And in July 2003, Fruehling's photo appeared in Guitar Player magazine. "I always thought when that happened, I'd be holding a guitar and have long hair and fire shooting around me," he says. "But I'll take it this way, sitting in my office with scopes on the desk and equations on the whiteboard."
--Tekla S. Perry
  • Upvote 2
Link to comment
Share on other sites

These are new terms, "modeler" and "profiler" in the sense of "one piece of musical gear that sounds like another." Since they're new terms, their meanings are still evolving.

 

The general audio and mathematical concepts are the same in both.

 

It seems to me that the main difference between a modeler and a profiler is:

- with a modeler, the manufacturer does all the tests and takes the necessary measurements of the original gear

- with a profiler, the users can do those tests and measurements themselves (using functions in the profiler)

 

Advantages of a modeler:

- Don't have to learn how to take a profile or worry about getting the best quality measurements

- Don't have to rent/borrow/buy/steal the best gear to make a profile of it

- With good quality modelers, the models are created by experienced professionals

 

Advantages of a profiler:

- Can choose which equipment to profile

- Can share profiles between users (in addition to patch settings)

- Probably other things I haven't thought of

  • Upvote 1
Link to comment
Share on other sites

In that sense. you're right mdmayfield. 

I'm sure modeling uses some audio analysis and capturing techniques, but with something like a Kemper, the Kemper does all the work and less guessing and more straight forward actually getting the sound.

 

I think if modeling was more about the company doing that on their own instead of letting the users do it, and if the company used the same techniques as the kemper does in it's routines, we wouldn't be bickering too much about modelers sounding more accurate than other modelers.

 

 

My view is more or less this:

 

Imagine someone making a cabinet simulator.

 

modeling seems more like if someone used a frequency analyzer and then dialed in the EQ to simulate the cabinet

 

profiling is more of a somebody programming a computer to do a bunch of audio sweeps to gather the information itself and compile it into an impulse response. The processor does the work to compile something to recreate the sound, instead of a programmer. The programmer more or less just makes those routines and technology to use it in that way.

 

With the Variax, if they used sensors to make up impulse responses then it seems like it would be more akin to profiling. It would explain why it sounds so accurate.

Link to comment
Share on other sites

In this context profiling to me would impliy some sort of IR data while modeling implies math.

But modeling can indeed use IR data in its math.

 

This topic reminds me of trying to explain to people the difference between an emulation or a clone of classic video games.

However the differentiation there is black and white.

  • Upvote 1
Link to comment
Share on other sites

In this context profiling to me would impliy some sort of IR data while modeling implies math.

But modeling can indeed use IR data in its math.

 

This topic reminds me of trying to explain to people the difference between an emulation or a clone of classic video games.

However the differentiation there is black and white.

 

Agreed, though I think in a sense, mdmayfield might be correct on what modeling and profiling really means, that profiling is more open for the user to make the amp tone, as opposed to the company "profiling" it at their R&D place and then it being closed for us to use.

 

I was more thinking about the technique as opposed to what the people are allowed to do.

In old modeling, there was less and less convolution based analysis and compilation, and more A/Bing, and tweaking sound functions to get it to sound close. There's no question that modern modeling uses a lot of more direct and accurate data analysis, and one thing that modeling does have against profiling is that profiling can't emulate an EQ section or the change of the gain dail of the amp it's emulating, especially if the thing it's trying to emulate is something that might have drastic changes when you turn those knobs on your amp other than just "turning up or down the gain".

 

 

Also, about the emulation analogy, emulating video games is using the exact ROM of those video games, but running it through something that isn't native hardware, but uses programming to simulate everything that the original hardware does so the game can actually run.

 

Opposed to simulating, which is rewriting everything form scratch and just tweaking it to act like a certain game does.

 

In a way, the Kemper is more akin to emulation, and modelers more to simulation, but again, that doesn't mean that modeling can't rely on techniques the Kemper uses, it just doesn't SOLELY rely on it.

 

In a way, I find it hard to believe that you could gather all the information of a guitar via just using sensors, for the Variax, but the sensors will cover a lot of ground in the first place. I just doubt it's what they used for stuff like the guitar electronics emulation, probably more or less of studying and knowing how those voltages do much of what it does.

The pickups and body responses probably used most of the sensor technology, but I wonder how they mashed it all together, especially the part where you can slide pickups to different positions in workbench.

 

So maybe the Variax isn't a profiler, but it comes pretty close.

Link to comment
Share on other sites

I think you're probably right concerning the voltages.

And I would think the alt tunings are modeling algorithms also.

 

The first time I heard the term profiler in this context was from Kemper.

Wouldn't that be some voodoo if you could plug any guitar into Workbench HD and then save it to your JTV!!!??

 

I'm fairly new to the JTV and it still amazes me every time I dig into WB.

 

What I like about video game emulation is that the hardware is emulated so well, all the glitches of the original game "works".

  • Upvote 1
Link to comment
Share on other sites

I think you're probably right concerning the voltages.

And I would think the alt tunings are modeling algorithms also.

 

The first time I heard the term profiler in this context was from Kemper.

Wouldn't that be some voodoo if you could plug any guitar into Workbench HD and then save it to your JTV!!!??

 

I'm fairly new to the JTV and it still amazes me every time I dig into WB.

 

What I like about video game emulation is that the hardware is emulated so well, all the glitches of the original game "works".

 

I'm sure the alt tuning is just a pitch shifter. You can't really model different tunings, but you can use techniques to make pitch shifting sound more realistic.

Frankly, the downtuning on the Variax is incredible, but the uptuning could use a bit of work. The downtuning doesn't muddy the highs like most pitch shifters do, but the uptuning still does that alvin & the chipmunk squeaking effect to some degree on the guitar.

 

The best thing to do is to make timbre preserving routines and algorithms, and I do think the Variax has that to some degree, but not to the extent of say, using DAW plugins like PitchWheel by quikquak or ReaPitch in Reaper.

 

As for Workbench, that would be amazing, but obviously Line 6 would need to sell the technology to make our own guitar models. The thing with the Kemper is that it's so direct, that you just need to plug in your guitar, then plug in a few other cables, and that's it. 

The Variax probably has way more equipment involved and a lot more work for the person.

 

Emulators have to be pretty much accurate to a T or else a routine ran in the code would cause the game to crash.

The thing people have to know to make emulators is how the processor works and what language it uses.

A lot of reverse engineering. I go on a forums where people make homebrew NES games and it baffles me how they reverse engineer NES assembly code and learned it. I can do programming, but assembly is way too complicated for me to comprehend.

  • Upvote 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...