| Kuei Yang Wang |
Konnichiwa,
I think it is time to actually do a controlled, like for like test of this, meaning matched gains, identical components etc., as we have plenty "Apples & Pears" comparisons which are unlikely ever to have been level matched.
(And the difference in gain between Peters "inverted" and "non-inverted" circuit is just large enough to be inaudible as increase in level but to be noticable in exactly the way it is being described. See also the classic MF Audio X-10 Tube Buffer Effect - in a level matched comparison it materially disimproves the sound, but if just inserted between CD Player and Pre and A/B'ed WITHOUT correcting the extra gain it has been described and observed to sound IN COMPARISON exactly as the supposed differences between NI & I circuits with different gain.)
I suggest I'll build a "gainclone" using a circuit that offers a "semi-balanced" input with XLR's as jacks. This will be driven by a TVC based linestage and the Ack!dAck.
A 3-Position switch will be fitted to select inverting mode (ground the XLR pin connecting to the positive input), balanced input (switch open circuit) and non-inverting mode (ground the XLR pin connecting to the negative input), in all cases changing only the mode of operation (inverted, pseudo balanced, non-inverted), while leaving gain, components etc EXACTLY identical.
In order to eliminate any personal bias by myself I suggest any London (UK) or South UK based members who feel like having a listen drop by for an organised, blind listening session (weekend days only and I need some time to build the Amp and prepare of course).
Any takers?
Sayonara
PS, of course, this does not have any bearing on the question "Does Circuit A with the Parts Complement A and Part Value A sound better than Circuit B with Parts Complement B and Parts Value B?" Which in the end is a purely subjective question anyway. |
|
|
| protos |
Wow ! Things are getting serious between the I and NI camp.
However since KYW otherwise known as TL is willing to go to such lengths to reach a conclusion I think some audiophiles would be interested to pariticipate in his invite.
I just wish I was closer although I can't rule out showing up via one of those "0 euro" Ryan air flights to London. |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by protos
Wow ! Things are getting serious between the I and NI camp.
|
I don't think there is anything to "get serious" about.
I merely object to the concept of comparing two quite different amplifiers and to draw conclusions about the fundamental topology from this.
On one side there are experimentor guys who try a huge lot but rarely if ever take care to level the playing field sufficiently to allow for direct comparison of topologies. On the other hand you find people who are interested primarily in the fundamentals behind the circuitry.
I seem to remember that a while ago someone build a DAC that was very complex, with asyncronous re-clocking and another, very different one that omited re-clocking and which sounded better. The result was a sudden "public opinion swell" against asyncronous reclocking just as previously it had been discussed as the greatest thing since sliced bread.
Longer ago there where long discussions on battery power and some people upon trying (for whatever reasons) found Batteries not to their liking and suddenly batteries where completely out (they appear in again, let's see for how long).
I go with Schiller/Beethoven:
" Deine Zauber binden wieder, was die Mode streng geteilt:
alle Menschen werden Brüder, wo dein snafter Flügel weilt."
The fashion is non of my concern.
Sayonara |
|
|
| Nuuk |
Good idea T - pity London is a bit too far for me too but I have an 'agent' in your vicinity! ;)
Seriously, I got a bit miffed when people kept asking for comparsions of my OPA627 buffered IGC and the VBICG. There are just too many differences to draw a fair and objective conclusion.
I have seen people swear that one CD player is better than another simply because the 'better' one had a bit higher output. I'm not suggesting that those saying that the non-inverting version sounds better are mistaken but I do know just how easy it is to be fooled in this game.
I can only say to those living close enough to T, should make the effort to participate in this worthwhile experiment. |
|
|
| ronc |
I did an A/B between the IGC and the NIGC but used different values of gain.However the SPL was matched using a signal generator.
Both were battery P/S and both used C resistors.
In the end after having 5 different ppl do a blind test i found that the IGC with an active pre had the same or so close to the same performance of the NIGC with a passive pre (50k shunted pot) that it was hard to tell the difference.
I like the active pre (tda1524a) because i can dial in LF and HF response which is an adjustment for room conditions.
Yes i am not a fan of EQ, but unless you design a set of speakers specifically for a room, EQ becomes necessary.
ron |
|
|
| Peter Daniel |
First of all, upon switching amplifiers I always matched the gain, although by ear only.
Second, I don't care at all which configuration is superior. It may very well be truth that inverting is better, but when implemented in the amp, doesn't allow to use components values that can be used with non inverting topology. And this may as well give the NI amp the advantage.
It's not even about direct comparisons. We sent an amp to the guy in another country. He comments it's ok, but not great. We ask him to send it back and we send him the modified amp. Now he comments it's better than Gain Card. So matching the levels is not the issue here.
It's the topology (to some extent or not at all), the parts values and parts choices.
It may appear than in your controlled test, the inverting circuit comes out better. But so what, if you cannot implement it the same way as non inverting one? Less feedback resistance, better input impedance matching and so on.
If you guys are so serious about it, I might even come to London myself;) |
|
|
| ronc |
I would never state than one config was better than another, just that wih an active pre the amp has more flexibility as to adapting to room conditions.
As i sell to "the common man" and not audiophiles i find that ,in general, most cant hear the last 10%.
Hay its an easy sell, most ppl have never heard a high end (whatever definition you care to hang on that one is) system.With the increased dynamics of the horns and along with the resoluion and accuracy of ANY config GC i find that the general public is blown away.I love either the 2 normal responses i get, one is the open mouth with jaw hanging the other is wide eyed with an ear to ear grin.In both cases i fine the effort i have put into the systems worth while.
ron
BTW! i calculated my profits and i can make more working part time at a fast food resturant.Its more for the ppl that find good sound than me making money. |
|
|
| Mad_K |
Great idea!
-Would you mind posting the test-schematic KYW?
;) |
|
|
| ronc |
Are you addressing me?
ron |
|
|
| millwood |
| quote: | Originally posted by Kuei Yang Wang
A 3-Position switch will be fitted to select inverting mode (ground the XLR pin connecting to the positive input), balanced input (switch open circuit) and non-inverting mode (ground the XLR pin connecting to the negative input), in all cases changing only the mode of operation (inverted, pseudo balanced, non-inverted), while leaving gain, components etc EXACTLY identical. |
that's an excellent way of doing it. Unfortunately, London is a little bit out of the way for me. But I look forward to hearing the results from you. |
|
|
| Peter Daniel |
| On a second thought, a test rig proposed by KYW isn't really fair to compare virtues of both config. Build two amps, the best way you can, utilising the advantages each configuration offers to the maximum in ea. amp. That's exactly what I did. |
|
|
| ronc |
Good idea KWG, as long as the gain ratio was the same.
As i am trying to get the most bang for the buck i dont use higher end componens ie.Black Gates.All i use is standard hammond cabs , +/- 5% C resistors, and nichicon caps.Any pots are 50K alps blue.
Yea i know better components will sound better , but as stated the general public cant normally tell the difference in the last 10%.
ron |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by Peter Daniel
First of all, upon switching amplifiers I always matched the gain, although by ear only.
|
I would not trust my ears to even get it to within 1db. I always use a Voltmenter and test-tone.
| quote: | Originally posted by Peter Daniel
Second, I don't care at all which configuration is superior. It may very well be truth that inverting is better, but when implemented in the amp, doesn't allow to use components values that can be used with non inverting topology. And this may as well give the NI amp the advantage.
|
My point exactly. I have no problem whatsoever with your observation that your implemented NI Amplifier sounds subjectively better than the I version.
I do have a problem when then it is concluded (not by you I might add) that NI is inherently superior after all (eg. the audio orthodoxy was right after all).
| quote: | Originally posted by Peter Daniel
It's the topology (to some extent or not at all), the parts values and parts choices.
|
IT IS THE COMBINATION OF...the topology, the parts values and parts choices.
| quote: | Originally posted by Peter Daniel
It may appear than in your controlled test, the inverting circuit comes out better. But so what, if you cannot implement it the same way as non inverting one? Less feedback resistance, better input impedance matching and so on.
|
I have no quarel with that either. It is as I mentioned before often a question of design aims and design choices. And for example - non of the parts you favour would find their way into my own gear, simply because I find them by far too manipulative sonically (and will maintain the same for Op-Amp's NI), but that is simply a matter of preference and choice.
| quote: | Originally posted by Peter Daniel
If you guys are so serious about it, I might even come to London myself;)
|
We of course do welcome participation from the Commonwealth, but maybe not from the rebellious colonies.
Sayonara |
|
|
| Peter Daniel |
| quote: | Originally posted by Kuei Yang Wang
I do have a problem when then it is concluded (not by you I might add) that NI is inherently superior after all (eg. the audio orthodoxy was right after all).
IT IS THE COMBINATION OF...the topology, the parts values and parts choices.
And for example - non of the parts you favour would find their way into my own gear, simply because I find them by far too manipulative sonically (and will maintain the same for Op-Amp's NI), but that is simply a matter of preference and choice.
|
I guess I need to clarify my previous statements. I NEVER claimed that inverted topology is inferior to non inverted. It may very well be superior. I simply don't know.
What I claim is that GC amp built using the combination of NI topology, parts values and their choices seems to be superior to its inverting counterpart.
Now, I never claimed that the parts I'm using are the best. It's just happens that this is what I was given to try. But I'm always ears to good suggestions and if you think that some other parts are better, I would gladly try them out.
After all, I'm not in all that for the money (although some financial compensation would be nice after all that time spent). I'm simply on a quest to build a really good sounding amp. I made some mistakes on a way, but I always try to learn and improve on them.;) |
|
|
| moamps |
| quote: | Originally posted by Kuei Yang Wang
I suggest I'll build a "gainclone" using a circuit that offers a "semi-balanced" input with XLR's as jacks. This will be driven by a TVC based linestage and the Ack!dAck.
A 3-Position switch will be fitted to select inverting mode (ground the XLR pin connecting to the positive input), balanced input (switch open circuit) and non-inverting mode (ground the XLR pin connecting to the negative input), in all cases changing only the mode of operation (inverted, pseudo balanced, non-inverted), while leaving gain, components etc EXACTLY identical.
|
Hi,
- with only grounding one input, amp gain will change (1+Rf/Ri and Rf/Ri)
- what about absolute phase?
I think that PD's idea (separate amps) is much better.
Do you will measure with scope outputs of amps? I was noticed for some degree higher instability in NI topology (especially when input and output terminals are too close).
Regards |
|
|
| Peter Daniel |
| I also think that NI amp is after all harder to built and more prone to outside interferences. Especially without input coupling resistor (which seems not necessary really), the amp may oscilate and produce increased DC offset in some setups. |
|
|
| ronc |
Rebellious colonies? LOL
When i set up the A/B test i used a signal generator set to 1Khz and an SPL meter at 1.5 meters.Yea i am from the colonies but i find the metric system better, easier anyway.
Its my feeling that if you design for the golden ears the cost goes up and eliminates many of us common folk from getting a better sound system.
ron |
|
|
| Peter Daniel |
| quote: | Originally posted by Kuei Yang Wang
And for example - non of the parts you favour would find their way into my own gear, simply because I find them by far too manipulative sonically (and will maintain the same for Op-Amp's NI), but that is simply a matter of preference and choice.
|
I know what you mean here. It took me quite a long time to finally disregard all the flavour of the month coupling caps (MIT, Hovlands, Jensens, to name just a few) and decide on using some obscure, motor run, Siemens caps. But hey, I never said I know it all.;) |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by moamps
- with only grounding one input, amp gain will change (1+Rf/Ri and Rf/Ri)
- what about absolute phase?
|
Please re-read my post and if neccesary actually draw out what I describe. The key-word in there "pseudo-balanced". No matter what position the switch is in the gain (both noise-gain and signal gain) is always the same and output polarity is always the same. It does rely on the Transformer (volume control) of course to make the work allright.
| quote: | Originally posted by moamps
I think that PD's idea (separate amps) is much better.
|
I do think it is not acceptable, unless implemented with identical components and component values. Otherwise if we take Peters approach I could make my inverting Amp with a BUF634 Buffer and an even lower impedance feedback loop and use any number of other tricks to go beyond the performance (subjective & objective) of any straight Chipamp, which would proove nothing about topology.
Except, I cannot be bothered to invest that much time into solid state stuff so hence the basic "all else being equal BUT the mode of operation" setup proposed. And no, i am not going to use premium parts either nor am I going to batteries in the PSU.
| quote: | Originally posted by moamps
Do you will measure with scope outputs of amps?
|
Of course.
Sayonara |
|
|
| tbla |
| quote: | | I could make my inverting Amp with a BUF634 Buffer and an even lower impedance feedback loop and use any number of other tricks to go beyond the performance |
please explain that - perhaps new thread......? |
|
|
| Mad_K |
| quote: | Originally posted by ron clarke
Are you addressing me?
ron |
-No, KYW aka TL ;) |
|
|
| Pedja |
| quote: | Originally posted by tbla
| quote: | | I could make my inverting Amp with a BUF634 Buffer and an even lower impedance feedback loop and use any number of other tricks to go beyond the performance |
please explain that - perhaps new thread......? |
Diamond buffer (which BUF 634 is) can drive notably heavy loads so if you buffer the IGC with it, you can use 22k/1k or 22k/680R (or even lower values) feedback network for the inverted GC.
Pedja |
|
|
| Peter Daniel |
| But than you will _hear_ the buffer.;) |
|
|
| Raka |
And what does it say?
No, stop, this wasn't meant another thread like "How many words do you hear from track 2 of BUF634 heavy music?", so don't reply :clown: |
|
|
| Peter Daniel |
| It sez: Are you listening to me Raka? ;) |
|
|
| fmak |
I think it is time to actually do a controlled, like for like test of this, meaning matched gains, identical components etc., as we have plenty "Apples & Pears" comparisons which are unlikely ever to have been level matched.
I suggest I'll build a "gainclone" using a circuit that offers a "semi-balanced" input with XLR's as jacks. This will be driven by a TVC based linestage and the Ack!dAck.
-------------------------------------------------------------------------------
I am game, but would propose the simplest of topography, and driven by a resistor based passive of which I have several. I am in central London.
How are you going to use identical components when the NI is direct coupled and the INV has a coupling cap? |
|
|
| Peter Daniel |
| You don't need a coupling cap on INV. |
|
|
| jonclancy |
If this ABX comes to fruition, I would like to be considered. I have no audiophile kit (but some great ARCAM stuff!!). I am not an audiophile, or a classical musician, but I *do* have an ear for music! If this "listen-in" is in the new year, then I would be honoured to come to town.
Alll the best
Jon |
|
|
| Pedja |
| quote: | Originally posted by Peter Daniel
But than you will _hear_ the buffer.;) |
Hmm… yes, I know. ;) Actually it was just a few days ago when I listened to my AD844 common base I/V with and without its output buffer (again it is diamond). And yes, though it is not night and day, there is the difference. Without it the acoustic instruments have more natural colors, but with it the dynamics is better and the bass is deeper and this applies to all recording types I listened to. So that is with and without it. I liked it more with. Just please don’t tell me you like it simple, I like it simple too! And all this happened at the 100kOhm load. Hmm, what will happen on the 10k load? I’ll check this buffer very soon in the discrete implementation as a front end for the IGC.
How's your buffer project?
Pedja |
|
|
| Peter Daniel |
I'm working on it right now. If the parts form digi key come in tomorrow, I will have done tomorrow. I'm very curious about results.
Keeping things simple is OK, but it seems that simplicity is not the best solution, sometimes. |
|
|
| ronc |
Make things as simple as they can be, but not to simple-Albert Einstien.
ron
Quote from my old ME prof"the hardest thing you can do is to make something simple" |
|
|
| Nuuk |
This reminds me of my days back-packing around our lovely planet!
You could always tell the newbies who would have huge bulging backpacks, carrying far more than they would ever need.
The next stage was the 'I'm not a newbie' syndrome where people carried next to nothing in order to differentiate themselves from the newbies. Of course, they managed to do this as they scrounged off other backpackers.
The really experienced backpackers carried neither too much nor too little! :yinyang: |
|
|
| Ropie |
| Minimalism in itself is a pointless exercise, though making the very best of as little as possible is not. |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by fmak
I am game, but would propose the simplest of topography,
|
Which is what I will do.
| quote: | Originally posted by fmak
and driven by a resistor based passive
|
Why bother? TVC is vastly superior AND allows for the simple switchover of mode as described due to balanced output.
| quote: | Originally posted by fmak
How are you going to use identical components when the NI is direct coupled and the INV has a coupling cap?
|
I am going to use EXACTLY the same Amplifier, the only change being the grounding of either one or the other input (signal) side of the pseudo balanced input. Doing this keeps ALL parts, values, gain, polarity etc the same with straight and noiseless switching using a simple "center neutral" DPDT toggle switch.
Sayonara |
|
|
| fmak |
[QUOTE]Originally posted by Kuei Yang Wang
[B]Konnichiwa,
Which is what I will do.
Why bother? TVC is vastly superior AND allows for the simple switchover of mode as described due to balanced output.
-------------------------------------------------------------
Fine, but I would want to compare on my Audio Synthesis Passion V (imoproved) on phono, or my balanced Carver passive.
I am not fond of transformer coupling as this has it's own colour. The question is TVC may sound superior in a system but simplification is best in the case in point.
Note: I can't do it until mid Decemeber. |
|
|
| ronc |
TVC may impart its own color,but the color will be equal.
ron |
|
|
| Peter Daniel |
| The amps will sound different when driven by resistor based passive and TVC. IMO, TVC extracts all the best from those amps, while reistor based passive is more neutral, but not always appreciated. |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by fmak
Fine, but I would want to compare on my Audio Synthesis Passion V (imoproved) on phono, or my balanced Carver passive.
|
Well, that would mean "not in my system". And I'm not making a "Demo" amp, I would merely build my "spare" amp (which is in the offing for a long time)
| quote: | Originally posted by fmak
I am not fond of transformer coupling as this has it's own colour.
|
And I am not (any longer) fond of resistive attenuators, as their own sonic coloration (arguably mostly subtractive) detract.
At any extent, using ANY other methode removes the "level playing field" unless we add a 1:1 balancing transformer into the mix (again).
| quote: | Originally posted by fmak
The question is TVC may sound superior in a system but simplification is best in the case in point.
|
I agree that simplification is best, in which case a TVC is best.
| quote: | Originally posted by fmak
Note: I can't do it until mid Decemeber.
|
Nor can I, probably new year....
Sayonara |
|
|
| Raka |
| quote: | Originally posted by Peter Daniel
It sez: Are you listening to me Raka? ;) |
I wish I could!
:clown: |
|
|
| Gerry |
I havent found any further mention of the proposed IGC / NIGC comparison.
Has it happend? If so, what was the outcome ?
Gerry |
|
|
| fmak |
[QUOTE]Originally posted by Gerry
[B]I havent found any further mention of the proposed IGC / NIGC comparison.
Has it happend? If so, what was the outcome ?
------------------------------------------------------
He went quiet; I don't see why TLC is simplist other then the fact that he uses one. |
|
|
| Gerry |
FWIW, I just came across this explanation why inverted is supposed to be better than non-inverted - sounds convincing to me:
"The Inverting Amplifier <is> the preferred configuration for an OP Amp amplifier. ..... The advantage of this configuration lies in the that the signal enters the amplifier at a zero voltage, zero impedance point, the conjuncture of the two resistors and the inverting input. This means that unlike the Non-Inverting amplifier configuration which accepts the full voltage swing of the input signal, which can cause non-linear effects because of the moving of the input bias voltages and compressing of voltage headroom as the signal swings from one extreme to the other, the Non-Inverting amplifier inputs are comfortably at the same power supply midpoint." http://www.glass-ware.com/audiogadg...its_Detail.html |
|
|
| Kuei Yang Wang |
Konnichiwa,
| quote: | Originally posted by fmak
[QUOTE]Originally posted by Gerry
[B]I havent found any further mention of the proposed IGC / NIGC comparison.
Has it happend? If so, what was the outcome ?
------------------------------------------------------
He went quiet; I don't see why TLC is simplist other then the fact that he uses one. |
Well, I still intend to do this, when I have time.
I actually KNOW the answer to the question already (Pseudo-balanced is slightly ahead of inverted, non-inverted notably behind), but I had actually modest interrest to actually illustrate this publically.
However, in order to keep an absolutely level playing field you MUST have a floating balanced input and in effect a pseudo-balanced input where you merely switch the relevant input to ground.
As this caused already a major stir with any mumber of people compalining about the methode and wanting to change it, what's the point and more to the point, why don't they go and do their own comparisons with their own approaches.
Sayonara |
|
|
|
