No kidding. Its not the transistors that usually limit the HF (well not directly) its the compensation. And 50% (-6dbv) is not nearly enough, sometimes at 10k its closer to 20 or 30 or even 40db down, but a well designed GNFB amp will still have more than 60 times (18 db?) of loop gain.
A good designer will not guess at these figures he will try to determine the pole locations which will tell him the freqs and how fast his open loop gain drops and his total phase lag.
You tryin' to pick a fight, buster?
Seems to me you enjoy elevating yourself above others..... a good designer will know the effects of all these technical assessments on the subjectives, and not weasel out by heaping contempt on those for whom the subjectives are important.
I'm outa here.....
Seems to me you enjoy elevating yourself above others..... a good designer will know the effects of all these technical assessments on the subjectives, and not weasel out by heaping contempt on those for whom the subjectives are important.
I'm outa here.....
OK, we establish that you are not deliberately adversarial. That's good.
Subjectives:
sibilance
listener fatigue ('edginess')
mid and top end clarity
image focus, ability to define spatially the exact location of an instrument or vocalist
front to back imaging, including depth
left to right imaging, including width
euphony - if the amp is 'musical' or not
These words are the darlings of some audiophile reviewers who inform the market, but they are necessary if we are to assess the sound without lapsing in technical vernacular which may not have any obvious link to the listening experience. And they are commercially higly relevant since people, right or wrong, often buy on the strength of reviews.
None of them can be measured, to my notion, until we dig deeper into harmonic profile of the distortion and phase shifts. Therefore their meaningful correlation with the technical terms like THD, slew, HF pole, phase margin, loop gain drops, etc, bear considerable investigation. But this is to some extent new ground - this correlation - and likely to cause lot of fights.
I think it's high time, myself. If a transistor amp sounds terrible with 0.05% thd and a single ended triode sounds wonderful at 2%, then further investgation is required. Yet no one, me included, steps forward to undertake anything resembling an explanation.
We have all been hijacked by the psychology of numerical appraisal.
Subjectives:
sibilance
listener fatigue ('edginess')
mid and top end clarity
image focus, ability to define spatially the exact location of an instrument or vocalist
front to back imaging, including depth
left to right imaging, including width
euphony - if the amp is 'musical' or not
These words are the darlings of some audiophile reviewers who inform the market, but they are necessary if we are to assess the sound without lapsing in technical vernacular which may not have any obvious link to the listening experience. And they are commercially higly relevant since people, right or wrong, often buy on the strength of reviews.
None of them can be measured, to my notion, until we dig deeper into harmonic profile of the distortion and phase shifts. Therefore their meaningful correlation with the technical terms like THD, slew, HF pole, phase margin, loop gain drops, etc, bear considerable investigation. But this is to some extent new ground - this correlation - and likely to cause lot of fights.
I think it's high time, myself. If a transistor amp sounds terrible with 0.05% thd and a single ended triode sounds wonderful at 2%, then further investgation is required. Yet no one, me included, steps forward to undertake anything resembling an explanation.
We have all been hijacked by the psychology of numerical appraisal.
Well said. It is the difference between the "PPM" designers and the Carlo's(DX) of the world. It is the main reason why I have gone modular. I listen , I either like it or I pull the voltage module out and change a value. Put it back , either the smile gets bigger or out comes the module again.
BTW , I know on the simple bootstrap amps .. I have measured close to a volt between bases on the differential. I just measured my GX (luxman) , just under 20mv for 130v p/p. confirmed on the real thing. All my "new" amps show nearly nothing (<5mV) as far as voltage differences between base voltages.
OS
Great if your designing for your self but the one thing Ive learned from this forum is there is very little consistancey in peoples opinion of what "sounds better". Some think tubes sound better than transistors some dont, Some think CDs sound better than LPs some dont, Some think opamps sound better than discrete some dont, Some think ____ sound better than ______ some dont. Im sure we can all come up with a lot of things to put in the blanks. Where does that put a designer who dosnt have the resources to do serious DBT of his designs? Back to the numbers.
Hello guys
Yes, your right Hugh.
This is why so much peoples gone back tunes amps, its more than commercial snake oil to sell tubes amps. So much transistor amp sounds dull and lifeless. For years, in the 80's and 90's the transistor amps makers did say that low thd was the solution and that tube amp was obsolete and lot of audio magazines say same and try to praise bad transistors amps as they was musical sounding.
Jean Hiraga was one of those who found out that the type of thd spectrum was more important than thd numbers.
Since some years there more amps designers who use both their brain and ears to design amps and there is more transistor amps who sound at least musicaly decent.
Bye
Gaetan
Last edited:
Thanks Pete, 20mV is good, that indicates 76.2dB of loop gain, excellent figure.
Another crusade. For every ill-informed, possibly corrupt reviewer there are a thousand unimaginative engineers blindly following the marketing department and the mantra of conventional, safe design practices. Get over it, there's good and bad in every profession.
In truth, engineers have been responsible, right or wrong, for motor vehicles that destroy lives, bombs that threaten humanity, chemicals that despoil the environment, bridges that fall down (Tacoma, 1940, very poor phase margin, huh?), satellite phones that are the tools of terrorism, the list goes on and on. You cannot condemn them for that, any more than you can condemn some audio reviewer trying to set down on paper what he hears, even if, crime of crimes, some hapless audiophile buys an amp that he later comes to dislike.
BTW, that 1% distortion figure on the tube amp, particularly if it's a SET, is conservative, and actually many of them sound absolutely transcendental, and are dismissed at peril. More work is needed to figure out why they sound so damn good to so many people, even the flea powered 2A3s.
Perhaps you also despise tone controls? Plenty of lousy recording engineers around deserve no less, as you may know. Even worse than being romantically misinformed is the cost cutting ethic of modern recording with its MP3 mass market 'appeal'.
Is it reasonable to regard the reviewer in the same way as the police regard lawyers?
Anyway, back to the !@#$ lawns for me, they are too high this spring....
Hugh
Another crusade. For every ill-informed, possibly corrupt reviewer there are a thousand unimaginative engineers blindly following the marketing department and the mantra of conventional, safe design practices. Get over it, there's good and bad in every profession.
In truth, engineers have been responsible, right or wrong, for motor vehicles that destroy lives, bombs that threaten humanity, chemicals that despoil the environment, bridges that fall down (Tacoma, 1940, very poor phase margin, huh?), satellite phones that are the tools of terrorism, the list goes on and on. You cannot condemn them for that, any more than you can condemn some audio reviewer trying to set down on paper what he hears, even if, crime of crimes, some hapless audiophile buys an amp that he later comes to dislike.
BTW, that 1% distortion figure on the tube amp, particularly if it's a SET, is conservative, and actually many of them sound absolutely transcendental, and are dismissed at peril. More work is needed to figure out why they sound so damn good to so many people, even the flea powered 2A3s.
Perhaps you also despise tone controls? Plenty of lousy recording engineers around deserve no less, as you may know. Even worse than being romantically misinformed is the cost cutting ethic of modern recording with its MP3 mass market 'appeal'.
Is it reasonable to regard the reviewer in the same way as the police regard lawyers?
Anyway, back to the !@#$ lawns for me, they are too high this spring....
Hugh
Because people like distortion if its the right type. Recording engineers have known this for a long time. 70s vintage Neve consoles still fetch top dollar because we like there transformer sound (distortion) signature they produce (straight from Rupert Neves lips)
I prefer straight wire with gain. I think I am a rarity amongst audiophiles.
1k and 10K (2V input).
AX is the standard "blameless" (Bode att. 1)
GX is the big amp (120v p/p) (Bode att.2)
Ax has over 100db gain , it is the one that sounds sterile (but good) , and does not even show a couple mV between bases until you go "nuclear" at over 100v p/p.
The GX only has half the gain but sounds nicer in general , it deviates up to 20+ mV when "pushed". So , the "cause and effect" holds true between OLG and the differential. I've never tested this before. Both are stable , run basically the same NFB network , only differ in loop gain.
OS
Attachments
Hello
That just mean that many diyer and audio buyers don't have enough listening experiences and that they are very enthusiatic about their finds.
We need good ears and lot of live concerts and goods audio equipments listening to have some experiences.
Wen I was a young man I was sure that the Quad 303 was the best of the best until I've ears more high end amps and live concerts...
Btw, not all audio magazine are biased wen they do tests audio systems.
And it's not that we like some distortions, our ears/brain system are more sensitive to some type of distortions, even type of thd are less agressive to the ears than odd type of thd.
Bye
Gaetan
Last edited:
To elaborate , I actually can listen to that "blameless" for an hour , switch to the luxman (GX) and subjectively (or maybe even objectively) ascertain the strength's and weaknesses between them. This augments the simulation and the calculations as I can go from the simulator to hearing the circuit at 200W+ in a matter of hours.
Before long you can predict the simulators blabbering (FFT's , bodes , general error reports) into what a particular circuit will actually sound like. There really are some "glaring" differences in what a small change in compensation or just running your VAS or input pair under different electrical conditions might be.
There... I ranted, too.
OS
Before long you can predict the simulators blabbering (FFT's , bodes , general error reports) into what a particular circuit will actually sound like. There really are some "glaring" differences in what a small change in compensation or just running your VAS or input pair under different electrical conditions might be.
There... I ranted, too.
OS
Ahhh , that took a few beers
to belt out ... PS. subjectively , or maybe even psychologically ... A amp will always sound better to me with 0% alcohol. With others , I notice ... they appreciate any particular setup (speakers , amp , source) as their alcohol content increases. For this there is no simulator to predict the causes.
OS
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.