Head-Fi.org › Forums › Equipment Forums › Sound Science › Would "amp modeling" be feasible for headphone amps?
New Posts  All Forums:Forum Nav:

Would "amp modeling" be feasible for headphone amps?

post #1 of 14
Thread Starter 
In the world of guitar amps, there are many modeling amps that mimic the sound of more expensive amps. These amps have improved tremendously over the years. Obviously, they don't sound as good as their expensive counterparts, but they now sound very close.

A good example is the roland cube series (microcube, 15X, 20X, 30X, and etc.). Each of these amps have a dial that lets you switch between various tones that mimic some legendary (and massively expensive) amps.


AFAIK, there is no such thing as an "modeling amp" in the world of headphone amps. Is there a technical reason why no such thing exists? If guitar modeling amps are any indication, it seems reasonable to assume that a $500-600 headphone modeling amp could create an attractive variety of sounds.

Why not go one step further and start headphone modeling? Using an automated EQ, sound decay effects, and etc., is it unreasonable to say that a relatively neutral headphone could be made to sound similar to other headphone sound signatures?

Again, clearly you won't get exactly the same sound quality, but it could still be tweaked to sound good. And it would be INCREDIBLE bang for the buck (if you could spend 500-600 dollars to closely emulate the sounds of numerous $1000+ setups, possibly totaling over $10,000)
post #2 of 14
I think the idea of headphone amps is different than that of guitar amps.

With guitar it seems the idea is to produce a pleasing tone which may be caused by adding distortion, reverb etc.

With headphone amps the idea is to produce music faithfully without coloration.

I'm sure there are exceptions but I would say that in general, if an amp significantly colors the sound it's a bad thing.

At the end of the day a headphone amp is there to provide headphones with enough power to let them produce sound.
post #3 of 14
Punnisher got it spot on. Audiophiles(most of us anyways) are only interested in the most faithful reproduction of the music, and the feeling of listening to that music live.

There are many other ways to model certain sounds. One way is to simply use an equalizer or DSP unit. This is a HUGE no-no in audiophile circles, I have only hesitantly brought it up a few times and it has been met with scorn by virtually everyone.

I like to play with my music though, and focus on different parts of each piece. Using an EQ is a great way to add colourations to your sound, and I often find myself switching back and forth between different settings. I always gravitate back to the original, but the EQ can provide a very pleasing selection of different flavours from time to time.

I believe in getting an extremely clean amp and source, and then using DSP to alter sound as the operator wishes. A signal that is subjected to such processing could have conceivably lesser quality overall, but the option can always be turned off, or the component can even be taken right out of the system if one wanted to hear their music as cleanly as possible.

I can't afford a full blown one yet (I just play with my portable rig), but it's definitely something on my list for the future. I might even put a review of it up on this site *shudder*.
post #4 of 14
I agree. The goal of music reproduction is different from creation. Guitar players use all manner of toys to modify the sound to their liking. The point of headphone amps is to give you as accurate a portrayal of the sound as possible.

Further, DSP usually doesn't give good results. It might lend something, but it also takes away. When the driver is 1cm off your ear, faults and shortcomings really stand out. This is the same reason headphones are a lot pickier about amplification than speakers are. If there's the slightest fault, you are going to hear it.
post #5 of 14
Quote:
Originally Posted by jcx View Post
there is some evidence that the main determinants of an amp's "sound" are the frequency response and output impedance - when not clipping

it is usually the case that ss amps are flatter in frequency response and have lower output Z than tube amps so it is relatively easy to modify ss amps in the direction of tube amps:

"the Carver Challenge"

diyAudio Forums - Blind Listening Tests & Amplifiers

which suggests nulling a higher power ss amp against your preferred tube amp at levels where the tube amp isn't clipping - although the ss amp's sound may still diverge from what you liked about the tube amp at higher volume levels simply due to Loudness-Curve changes in perceived frequency balance
I acutally expect many reported cases of discriminating headphone amps could involve clipping behavior which wouldn't be easy to match
some would undoubtedly claim the original amp driven into clipping sounds better than your "modeled" version if youv'e used a ss amp with higher dynamic headroom that avioded clipping
post #6 of 14
you could try to play around w/ OzoneMP....but guitar players are looking for a "nitty gritty" sound, audiophiles using tube gear are looking for a musically saturated "soft" sound...not the same goals at all.
post #7 of 14
It would be like trying to get a cheap Fiat to perform like a race spec'ed Ferrari with quality parts all around. It just cannot be done, the premises are just not there.
post #8 of 14
Quote:
Originally Posted by nor_spoon View Post
It would be like trying to get a cheap Fiat to perform like a race spec'ed Ferrari with quality parts all around. It just cannot be done, the premises are just not there.
This. I think what you're asking is pretty close to impossible. I understand the amps that you're talking about(the Rolands) and I don't think they sound that good in the first place. Ask any owner of a Marshall full stack if the sound of your Roland even comes close. Same difference.

You also can't ask multiple performances from the components of a single device. Those thousand dollar setups, also had higher quality components(in most cases). What about DACs in those setups? Would you create a modeling DAC too?

Additionally, trying to match synergy with DACs, or even Headphones sometimes, would be a nightmare. I know the signature of my akgs. I want to build on that. I don't want an amp that sounds best with HD650s. That's going to turn up relatively fruitless. A jack of all trades quickly becomes master of none in the audio reproduction world. I want an amp that is going to accentuate the strengths of the sound signature that I like. And something that's dedicated to doing that, and just that, very well is going to make me alot happier in the long run.
post #9 of 14
Quote:
Originally Posted by nor_spoon View Post
It would be like trying to get a cheap Fiat to perform like a race spec'ed Ferrari with quality parts all around. It just cannot be done, the premises are just not there.
didn't read the Carver challenge article did you?

the diyaudio summary and commentary on the page I pointed to above is the quicker read, original article:

Stereophile: The Carver Challenge


"quality" in audio isn't the same as cars - its nearly impossible to get people here to look at any "performance numbers" or agree that even basics like wether there is enough I,V to avoid clipping with the amp, headphone combo they're gushing about count for anything compared to their uncontrolled subjective impression

in cars horsepower, torque curves, mass, handling/suspension difference mean sports/race cars can be clearly distinguished from passenger sedans by the numbers and the results at the track or on the road - no cornering salom course, race tracks, rallies, pikes peak climbs in audiophlie land = no way to "prove" your sujective impresions are related at all to equipment "quality"

I doubt single digit % of the subjective comparisons - even "consensus" opinion on this site could be demonstrated to be based on objective reality under even minimally controlled listening conditions

too many offer their opinions/subjective "evaluations" based on their memories of meet impressions without both/all of the compared items ever being in their possession within days/weeks much less 10s of seconds that are required to use our much richer short term memory for the comparison - when they've actually heard the items they're posting about at all

then try asking them about how they achieved/verified the DBT testing established 0.1dB level matching required for identical systems to really sound the same
post #10 of 14
Thread Starter 
Quote:
Originally Posted by crossmd View Post
This. I think what you're asking is pretty close to impossible. I understand the amps that you're talking about(the Rolands) and I don't think they sound that good in the first place. Ask any owner of a Marshall full stack if the sound of your Roland even comes close. Same difference.

You also can't ask multiple performances from the components of a single device. Those thousand dollar setups, also had higher quality components(in most cases). What about DACs in those setups? Would you create a modeling DAC too?

Additionally, trying to match synergy with DACs, or even Headphones sometimes, would be a nightmare. I know the signature of my akgs. I want to build on that. I don't want an amp that sounds best with HD650s. That's going to turn up relatively fruitless. A jack of all trades quickly becomes master of none in the audio reproduction world. I want an amp that is going to accentuate the strengths of the sound signature that I like. And something that's dedicated to doing that, and just that, very well is going to make me alot happier in the long run.
I said very clearly that it WON'T sound as good as the original. The point is to emulate some of the qualities of the original, to give the owner a variety of sounds at a cheaper price. I see no reason why tube rolling and amp-specific colorations cannot be emulated. Even certain headphone-specific qualities could be emulated by EQ presets, reverb-like effects at specific frequencies, and etc.

The goal of such an amp is NOT to produce the highest quality sound. The goal is to provide variety to maximize bang/buck while still providing a high standard of sound quality (which might be lower than the standard set by originals)

Look at the example of the roland guitar amps. It's amp models clearly can't compare to the originals (although their sound quality is highly debatable, many professional reviews rate its sound, especially the clean channel which emulates a JC120, very highly). HOWEVER, it does give a very good idea of what the originals sound like. Thus, it is extremely good for practice applications. It's also just plain fun. To get the equivalent variety of sounds WITHOUT emulation, would cost me 10,000 dollars or more. Instead, I get a very close emulation to many different amps on my 30X for under 200 bucks (the emulation is accurate enough that my jam partners can hear the tones and associate them with specific amps).
post #11 of 14
Look into the Smyth realiser system, Jawang. It is the fully-realized development of your idea, albeit aimed more at reproducing speakers than amps.

Problem is, again, your expected price point. There are maybe .1% the potential clients for this product that there are for a modelling amp. That means every client has to pay more to offset the expensive R&D needed to make it work.

The realiser is roughly $3000, but it's the end of the line for lots of folks.
post #12 of 14
This has kinda pseudo-exists - if you have a well-performing amplifier (read: cheap professional) with a clean, linear output, you can use DSP's and their various presets to emulate a tube amplifier. The Carver challenge pretty much put claims like this:

Quote:
View Post
It would be like trying to get a cheap Fiat to perform like a race spec'ed Ferrari with quality parts all around. It just cannot be done, the premises are just not there.
to rest. At the end of the day, differences in amplifiers are at their greatest between tube amps (I hold the position that no two amplifiers with sufficient power, impedance, distortion, and frequency response linearity can be distinguished audibly) because of their high distortion, and distortion is much easier to add than to take away. Guitar amps probably change the sound more than audiophile tube amps, but the means to create a "modular" amp are pretty much the same: take a low-distortion amplifier, add distortion that either replicates or approximates the distortion pattern of well-known amplifiers or tubes, and match levels.
post #13 of 14
Thread Starter 
I'm reading up on the Smyth realiser now, and it looks like an amazing piece of technology.

And yes, that system is almost exactly what I am talking about (in fact, it does more than that, since apparently it can record new sound environments and reproduce them)


I guess a better way to phrase this thread would be to ask whether dedicated DSP units are be viable, especially if they contain effects that significantly alter sound (i.e. tube emulation)
post #14 of 14
OzoneMP can emulate tube distortion/compression, but don't expect miracles

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Would "amp modeling" be feasible for headphone amps?