Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Discrete vs opamps...
New Posts  All Forums:Forum Nav:

Discrete vs opamps... - Page 3

post #31 of 176
Quote:
Rick, they have no experience and they keep trying to say our tubes are colored. Enough is enough ..... do we have to get rough with these guys
Don't tempt me man............
post #32 of 176
Quote:
Originally Posted by K2Grey
It needs to be pointed out that not everyone believes the Gilmore Lite / Dynalo is superior to all opamp-based amplifiers and thus your personal taste is not particularly relevant for purposes of making absolute blanket statements.
I know that and I was expessing my opinion. But what I ended up answering ... or defending ..... was my method of comparison between the two competing designs. All I do know is I hear flaws in op amp based designs that I cant live with regardless what they are compared to .... and the gilmore discrete designs are what I own. But regarding design .... do most well known amp designers in the field of audio consider op amps superior to discrete components? I have never had the impression they do.
post #33 of 176
Quote:
Originally Posted by sacd lover
But regarding design .... do most well known amp designers in the field of audio consider op amps superior to discrete components? I have never had the impression they do.
Nevermind .... the rickmonster came through ....


Quote:
Originally Posted by rickcr42
not true.It is just a way and for some the best way,others not.I personally have and have built every manner of topology there is and at this point in life prefer the simple single ended class-a stage above all others on the sonic merits.It is a taste thing and no more

Back to Opamps and if discrete is better than chip amps or visa-versa

Chip amps are and must be a comprimised solution because they are a generic part and not a highly specialised part.The manufacturer can "suggest" uses but in the end they have no idea what you will be using it for so must design for just about any contigent over a very wide range of possibilities.
This is not inherently a bad thing but the limitations are WE have no access to the inner workings and must accept whatever parameters the substrate designer decided were best.
Sometimes this leads to a good sounding chip but even thenm is and must be a comprimise.

Next.There is no such thing as a class-A monolithic opamp and there can not be.the size of the package and heat dispersion alone would dictate a much bigger area (footprint) AND some means of heat sinking so right there is a huge built in comprimise since it is well known class-A sounds better and that is not just "hype" but reality
Not only are chip opamps not capable of class-A but some not even Class-A/B but are in fact Class-B amps though in the long run class-B may be a better choice than an imperfect A/B stage and here is why

Class-A : Output section transistors are always turned on full throttle so they do not have to respond to music signal but just pass them along amped to a new level.Unless this amp is fully balanced from input to output all the output section transistors will be the same polarity (All NPN or all PNP)

Class A/B : Output section transistors only respond to a signal of the same polarity as the input voltage so a positive going signal activates the "P" channel transistor and the negative going half of the signal activates the "N" channel part.This may sound OK in theory but the reality is there will be a "disconnect" at some point between the P and the N trasistors that is totally dependant on where in the range they are biased into class A and where A/B takes over.Because of the package size of the opamp the "A" zone is virtually non-existant.The chip could not cast off the heat so is not asked to and that means crossover distortion where the + signal goes to a - signal and the transistion zone between the N and P channel transistors are not in sync
The reason why most opamps use this method is because it is more efficient,uses less power and does not require additional heat sinking but NOT because it is a superior way.Cost savings,low heat generation no more.Not merit but cost

class-B : This is where the output transistor is either totally ON or totally OFF just like in class A/B.Where it differs is there is no "cross" to a polar opposite but like class A all the signal is handled by a single device (either N or P).It is like Class-a in that way but unlike Class-a which is always on full boogie ready to rock class-B is off until activated by an actual signal.This may sound bad but in reality to me is a better way than AB though not as good as A.
Any real limitation will be at the leading edge of an audio signal where it must go from "off" to "play" so if it lags a bit then the front end of a note will be slightly truncated but the GOOD NEWS is if designed properly all the notes will STOP at the same time since it will go to "off" as soon as there is no signal.
This means the lowest heat generation of all three types (most efficient) and no crossover distortion because the plus does not hand off to the negative

That is what a monolithic opamp IS.

Discrete Op-amps ARE superior because the user determines the mode of operation at each stage and can optimise each for both the position in the circuit and for end use.If that end use is music then fully class-a operation with no limits is not only a possibility but what every audio specific discrete opamp design I know of operates at !


So there is the "why" of my statement that a properly designed discrete opamp will whip the sh*t out of any monolithic opamp.they can and they do and the only reason to consider a chip amp is portable operation,heat generation or low cost but NOT because it has better sound and that is what all this is about right ?



Told ya I was gonna be bad but when I am bad I am good !
post #34 of 176
Quote:
Originally Posted by rickcr42
I personally have and have built every manner of topology there is and at this point in life prefer the simple single ended class-a stage above all others on the sonic merits.It is a taste thing and no more.
Okay, let's just say, for the sake of argument, that I (and probably a few more readers) don't know what the heck you guys are talking about beyond a very basic level. But reading this far, and looking at my sig, me and my equipment look like they are on the wrong side of this argument.

Since I'm in the process of saving up my duckets for a sizable cash outlay for good sound at the end of the year, I would very much like your opinion on where my money should (and shouldn't) be spent. And since the days of soldering irons and watching o-scope traces are way behind me (military radar tech), I'd need recommendations for some off-the-shelf gear. If I start up with the DIY stuff, my wife is going to leave me.
post #35 of 176
Quote:
Originally Posted by fewtch

P.S. isn't the LaRocco PR-II somewhere close to $1000, and doesn't it use opamps?
Nope...$400-500 range. More if you upgrade the internal components. It does use opamps.
post #36 of 176
A few months back I owned a Maxed out PPA, Maxed out M3 and a GS1... I would take the GS1 over the others any day. In fact an all discrete design is what first lured me to the Lavry DA10 which I think those who have heard the thing would agree that it kicks the DAC1s butt (which is opamp based)... therefore Ide gladly take discrete any day over op-amps based gear in my experience.

Once I graduate and get a "real" job first thing I plan to do is pick up a GS-X or a PPX SLAM XLR because honestly, nothing op-amp based I have heard comes even close.
post #37 of 176
O
Quote:
kay, let's just say, for the sake of argument, that I (and probably a few more readers) don't know what the heck you guys are talking about beyond a very basic level
I can dig that and why i try do flesh out my answers with more than the "tastes great/less filling" type responses which go strictly to taste or preconceived prejudice towards and amp already purchased.no one like to have a guy like me come along after they made a thoughtful purchase of an amp and say "You screwed up man.It was behind door #2 not #1"

The reason I broke down the hows and whys of the all in one package monolithic opamp I.C. was not to slam it as an illegitimate method to quality hi fi sound but to explain why discrete if done well is a better way but like with the tube vs solid state argument it comes down to not just fsonics but finances.
The maybe best opamp known to humans on sonic merit (arguable as are all things with a subjective componant),the OPA627,is around $20 and will be way more than most need in a gain stage.Great sonics for what it is that when mated to a good discrete output stage can whip some serious butt !
A discrete opamp of even equal quality would cost a lot more and that just for starters.There will also be higher requirements on the power supply where even the worst,most current hungry opamp made can be run on batteries if only for a limited time.Try that with discrete Class-a and you go from "how many days will the batteries last in the amp" to "how many minutes !!!! "
So what an audio designer of commercial gear must do is identify the market " price niche" he/she is looking for an openng in then look to build cost and what topologies/methods that leaves.this translates to final product cost and an all out effort would reflect that so we have "use what you must to get the job done and still make a profit" or you would either never make any profit (thus go out of business fast) or if you take the All Out Assault route sell very few products because you priced yourself out of the reach of the average audiophile (which again you go out of business).


so entry to mid level it is opamps,mid to high it is either discrete or tube and ultra high end mostly tube gear with some solid state examples.Each good at what it does for the price point.

Portable or a combination Home/Portable amp means opamp or nothing since,and this is personal opinion area,all tubes I have heard that will run on a low current battery are more special effect than they are accurate or they have serious frequncy response limitations.Same with discrete transistor portable gear-comprimised by either being underpowered (heat again) or battery killers (better sonics).No free rides here folks !


So bottom line my long winded response was more to answer the initial question from my point of understanding rather than to knock one method or another.There ARE differences and they ARE audible but at what price are you willing to go to get an actual audible improvement ?

That is the question we all wrestle with and for some on limited income there is no choice.
post #38 of 176
Quote:
Originally Posted by rickcr42
Next.There is no such thing as a class-A monolithic opamp and there can not be.the size of the package and heat dispersion alone would dictate a much bigger area (footprint) AND some means of heat sinking so right there is a huge built in comprimise since it is well known class-A sounds better and that is not just "hype" but reality
Not only are chip opamps not capable of class-A but some not even Class-A/B but are in fact Class-B amps though in the long run class-B may be a better choice than an imperfect A/B stage and here is why

Class-A : Output section transistors are always turned on full throttle so they do not have to respond to music signal but just pass them along amped to a new level.Unless this amp is fully balanced from input to output all the output section transistors will be the same polarity (All NPN or all PNP)
Huh? What's all this DIY stuff about "CRD's to bias to class A" and all that, then? Are you saying an opamp can't be biased into class A operation?

I don't think I've ever read a more puzzling statement on head-fi before, unless I'm missing something the size of Texas sitting right in front of my nose.
post #39 of 176
Quote:
Originally Posted by rickcr42
not true.It is just a way and for some the best way,others not.I personally have and have built every manner of topology there is and at this point in life prefer the simple single ended class-a stage above all others on the sonic merits.It is a taste thing and no more

Back to Opamps and if discrete is better than chip amps or visa-versa

Chip amps are and must be a comprimised solution because they are a generic part and not a highly specialised part.The manufacturer can "suggest" uses but in the end they have no idea what you will be using it for so must design for just about any contigent over a very wide range of possibilities.
This is not inherently a bad thing but the limitations are WE have no access to the inner workings and must accept whatever parameters the substrate designer decided were best.
Sometimes this leads to a good sounding chip but even thenm is and must be a comprimise.

Next.There is no such thing as a class-A monolithic opamp and there can not be.the size of the package and heat dispersion alone would dictate a much bigger area (footprint) AND some means of heat sinking so right there is a huge built in comprimise since it is well known class-A sounds better and that is not just "hype" but reality
Not only are chip opamps not capable of class-A but some not even Class-A/B but are in fact Class-B amps though in the long run class-B may be a better choice than an imperfect A/B stage and here is why

Class-A : Output section transistors are always turned on full throttle so they do not have to respond to music signal but just pass them along amped to a new level.Unless this amp is fully balanced from input to output all the output section transistors will be the same polarity (All NPN or all PNP)

Class A/B : Output section transistors only respond to a signal of the same polarity as the input voltage so a positive going signal activates the "P" channel transistor and the negative going half of the signal activates the "N" channel part.This may sound OK in theory but the reality is there will be a "disconnect" at some point between the P and the N trasistors that is totally dependant on where in the range they are biased into class A and where A/B takes over.Because of the package size of the opamp the "A" zone is virtually non-existant.The chip could not cast off the heat so is not asked to and that means crossover distortion where the + signal goes to a - signal and the transistion zone between the N and P channel transistors are not in sync
The reason why most opamps use this method is because it is more efficient,uses less power and does not require additional heat sinking but NOT because it is a superior way.Cost savings,low heat generation no more.Not merit but cost

class-B : This is where the output transistor is either totally ON or totally OFF just like in class A/B.Where it differs is there is no "cross" to a polar opposite but like class A all the signal is handled by a single device (either N or P).It is like Class-a in that way but unlike Class-a which is always on full boogie ready to rock class-B is off until activated by an actual signal.This may sound bad but in reality to me is a better way than AB though not as good as A.
Any real limitation will be at the leading edge of an audio signal where it must go from "off" to "play" so if it lags a bit then the front end of a note will be slightly truncated but the GOOD NEWS is if designed properly all the notes will STOP at the same time since it will go to "off" as soon as there is no signal.
This means the lowest heat generation of all three types (most efficient) and no crossover distortion because the plus does not hand off to the negative

That is what a monolithic opamp IS.

Discrete Op-amps ARE superior because the user determines the mode of operation at each stage and can optimise each for both the position in the circuit and for end use.If that end use is music then fully class-a operation with no limits is not only a possibility but what every audio specific discrete opamp design I know of operates at !


So there is the "why" of my statement that a properly designed discrete opamp will whip the sh*t out of any monolithic opamp.they can and they do and the only reason to consider a chip amp is portable operation,heat generation or low cost but NOT because it has better sound and that is what all this is about right ?



Told ya I was gonna be bad but when I am bad I am good !

The problem with class B is that psychoacoustic research shows that the beginning of the note is most important part of the note and that it is the part that allows you to judge timbre.
post #40 of 176
Quote:
Huh? What's all this DIY stuff about "CRD's to bias to class A" and all that, then? Are you saying an opamp can't be biased into class A operation?
That is exactly what I am saying.It can be partially biased into class a but fully ? You would burn that sucker up in a microsecond !
post #41 of 176
Quote:
I don't think I've ever read a more puzzling statement on head-fi before, unless I'm missing something the size of Texas sitting right in front of my nose.
and you are so quite being a wise a*s
post #42 of 176
Quote:
Originally Posted by fewtch
Huh? What's all this DIY stuff about "CRD's to bias to class A" and all that, then? Are you saying an opamp can't be biased into class A operation?

I don't think I've ever read a more puzzling statement on head-fi before, unless I'm missing something the size of Texas sitting right in front of my nose.
True Class A AFAIK has to either dump heat in the voice coils of the speakers or it dumps it into the heat sink. So if it is designed to run in class A for 100w, the 100w must be dissipated somewhere continuously. That's the heat coming off a 100w light bulb!

Yeah, those things get hot and ironically they run cooler if you play music your music louder. It's just the speakers getting hotter in that case.
post #43 of 176
Quote:
The problem with class B is that psychoacoustic research shows that the beginning of the note is most important part of the note and that it is the part that allows you to judge timbre.
Granted but my personal listening says if the leading edge response is good and everything stops simultaneously the resulting sound is much more pleasing to listen to than class A/B crossover distortion.

Remeber what Timbre is and what most music systems do to that paricular parameter.

Systems A and Sytem B side by side,both equal in dollar value,both sound good,both playing the same identical CD with the same source and same final transducer and you have what ? Two entirely different sounding syustems even though they play the same song !

Which is true to the original "timbre" and which if either dead accurate ? My taste goes to getting the sound of my system to where it plays music.where i actually want to jump around like an idiot of dance or tap my feet to the beat not analyze the inner details of if the instrument is or is not dead on to what i remeber which would change in "timbre" from hall A to hall B anyway.

Class B can be very musical as can A/B but when compred to class a there is no contest for me
post #44 of 176
Quote:
Originally Posted by rickcr42
That is exactly what I am saying.It can be partially biased into class a but fully ? You would burn that sucker up in a microsecond !
Which is exactly why I find it humorous that some portable amp builders, that will go unnamed, say their amps are Class A There is a slight chance that they may be A/B bias but likely they are only B.
post #45 of 176
Quote:
Originally Posted by rickcr42
and you are so quite being a wise a*s
Not intended Rick, I just found your statement very puzzling. I don't understand the difference between partial biasing and full biasing of an opamp, but will look it up... thanks for enlightening me on that one.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Headphone Amps (full-size)
Head-Fi.org › Forums › Equipment Forums › Headphone Amps (full-size) › Discrete vs opamps...