I believe this question is one of amplifier design, but I am not sure.
Let us say that I am listening to the same tracks of music on two different amplifiers of nearly equal power ratings.
Amp 1: presentation is "normal", volume level is easy to set to a comfortable level and instruments are pretty equally matched in volume levels. There is not a marked difference between quiet and loud passages.
Amp 2: volume level for same track seems more difficult to achieve comfortable listening level, seems too soft. But then when there is a passage with wide dynamics this amplifier surprises me with how suddenly loud it is. It is dramatic.
The same sort of thing applies to the separate instrument levels. Where with Amp 1 a guitar and percussion are pretty equally matched in volume, listening to the same track with Amp 2 the guitar and percussion are not at the same level. The guitar is much louder and the percussion is much quieter.
In other words, Amp 2 seems to have much greater dynamics.
How can this be?
I apologize for my inability to accurately articulate my question but am willing to answer any clarifying questions.
Let us say that I am listening to the same tracks of music on two different amplifiers of nearly equal power ratings.
Amp 1: presentation is "normal", volume level is easy to set to a comfortable level and instruments are pretty equally matched in volume levels. There is not a marked difference between quiet and loud passages.
Amp 2: volume level for same track seems more difficult to achieve comfortable listening level, seems too soft. But then when there is a passage with wide dynamics this amplifier surprises me with how suddenly loud it is. It is dramatic.
The same sort of thing applies to the separate instrument levels. Where with Amp 1 a guitar and percussion are pretty equally matched in volume, listening to the same track with Amp 2 the guitar and percussion are not at the same level. The guitar is much louder and the percussion is much quieter.
In other words, Amp 2 seems to have much greater dynamics.
How can this be?
I apologize for my inability to accurately articulate my question but am willing to answer any clarifying questions.
Yes, different amp designs do in fact have different dynamics.
There's a lot of different things that go into this, but just to give my opinion (based on my own builds and design experiments) on the first things that come to mind:
1) The most noticeable dynamic effect, where the complete volume of the signal is compressed, most often is caused by high impedance PSUs. Switch to SS diodes, and regulate the B+, and you get a whole lot of dynamics in the overall volume of the output.
2) The inner dynamics, which are harder to describe accurately, can be improved with reducing distortion. The reason why some things sound louder (and it's not really that they're louder per se, they're more perceptible, more to the front) is that the harmonics that the amp produces (adds to signal, distortion), benefit some instruments more than others.
Best example: SE has a lot of 2H; this brings out voices, especially female voices, to a great degree.
Reduce distortion, and you bring the original balance back; this is fidelity (in best cases, high fidelity).
3) The third dynamics effect is closely related to the two described above, but still I feel separate. Even in amps that have low impedance PSUs, and very low distortion, you can still have one amp be 'slow', and one amp be 'fast'.
I found this when I started driving all output stages with CCS loaded source followers. It makes any amp 'faster', and some amps it makes a LOT faster. The sound has 'impact', and even at same sound pressure level, it's "more powerful". Reasons for this have been talked about on this forum.
These are my main thoughts on dynamics.
There's a lot of different things that go into this, but just to give my opinion (based on my own builds and design experiments) on the first things that come to mind:
1) The most noticeable dynamic effect, where the complete volume of the signal is compressed, most often is caused by high impedance PSUs. Switch to SS diodes, and regulate the B+, and you get a whole lot of dynamics in the overall volume of the output.
2) The inner dynamics, which are harder to describe accurately, can be improved with reducing distortion. The reason why some things sound louder (and it's not really that they're louder per se, they're more perceptible, more to the front) is that the harmonics that the amp produces (adds to signal, distortion), benefit some instruments more than others.
Best example: SE has a lot of 2H; this brings out voices, especially female voices, to a great degree.
Reduce distortion, and you bring the original balance back; this is fidelity (in best cases, high fidelity).
3) The third dynamics effect is closely related to the two described above, but still I feel separate. Even in amps that have low impedance PSUs, and very low distortion, you can still have one amp be 'slow', and one amp be 'fast'.
I found this when I started driving all output stages with CCS loaded source followers. It makes any amp 'faster', and some amps it makes a LOT faster. The sound has 'impact', and even at same sound pressure level, it's "more powerful". Reasons for this have been talked about on this forum.
These are my main thoughts on dynamics.
even motherboard soundcards have enough resolution today to explore such "questions" in a objective manor
need at least attenuators, maybe one of the soundcard interface projects: A community dedicated to helping everyone learn the art of audio. Projects by fanatics, for fanatics! - Search Results for soundcard interface
basic level matching, frequency response testing is the start - Loudness Curves mean even a dB different SPL while listening give different perceptual frequency balance so must be matched to 0.1 dB (~1% drive V)
if you use the same speakers then just Vdrive monitoring should help a lot, have higher resolution than acoustic measurements, but even low cost condenser mics are relatively flat, the same mic should be quite repeatable over a day
there's really no point today in not doing the measurements given the low cost of entry compared to a significant amp build
need at least attenuators, maybe one of the soundcard interface projects: A community dedicated to helping everyone learn the art of audio. Projects by fanatics, for fanatics! - Search Results for soundcard interface
basic level matching, frequency response testing is the start - Loudness Curves mean even a dB different SPL while listening give different perceptual frequency balance so must be matched to 0.1 dB (~1% drive V)
if you use the same speakers then just Vdrive monitoring should help a lot, have higher resolution than acoustic measurements, but even low cost condenser mics are relatively flat, the same mic should be quite repeatable over a day
there's really no point today in not doing the measurements given the low cost of entry compared to a significant amp build
Last edited:
Thanks for the responses.
I should mention that I have two amplifiers that most dramatically exhibit this dynamic affect. Both were built by Jef Larsen of Abraxas Audio. One is a tube rectified SET/SEP KT88 and the other is a SS rectified PP 6V6. The KT88 is the most "dynamic". I guess that if the PSU is key to this then I must say that Jef has the power supply part of the design down very well.
Compression would actually best describe how most amps /that I've heard/ behave compared to these two.
I should mention that I have two amplifiers that most dramatically exhibit this dynamic affect. Both were built by Jef Larsen of Abraxas Audio. One is a tube rectified SET/SEP KT88 and the other is a SS rectified PP 6V6. The KT88 is the most "dynamic". I guess that if the PSU is key to this then I must say that Jef has the power supply part of the design down very well.
Compression would actually best describe how most amps /that I've heard/ behave compared to these two.
Last edited:
No Cap between drivers and output stage.
Direct coupling between the driver stages and output put stage, Push pull & SE amps have excellent dynamics. Class A2 SE and class AB2PP. You can use Cathode followers or Mosfets. Of course you still need a good power supply, not necessary regulated.
Direct coupling between the driver stages and output put stage, Push pull & SE amps have excellent dynamics. Class A2 SE and class AB2PP. You can use Cathode followers or Mosfets. Of course you still need a good power supply, not necessary regulated.
Attachments
On my first amplifier design I used a b+/b- of +30 and -30 volts.
I found it clipped very easily before I got any decent volume out of it.
I quickly learned about transient response.
Some peaks can be massive in music.
I now use much higher B+ lines in my amplifiers.
I found it clipped very easily before I got any decent volume out of it.
I quickly learned about transient response.
Some peaks can be massive in music.
I now use much higher B+ lines in my amplifiers.
Agree with what has been said so far. From my own experience I would add the quality of resistors in the signal path- ones with low temperature coefficient (metal film, metal foil) sound more dynamic. The old carbon-stick resistors would change their value slightly with signal level and work to compress dynamics.
Most of those peaks are removed when mastering. And in the last 20 years they are completely gone, due to the loudness wars.
This is complete nonsense.
If you look at how music is recorded, the peak level can only ever be 0dB.
0dB CANNOT make any amplifier clip.
If it does, then something bad has been done to the design.
QED 😱
The vast majority of recorded music is compressed, taking it from a nominal S/N of about the very best 35-40dB to much less.
Today this is based on ITU "loudness ... LKFS - Wikipedia, the free encyclopedia
Very very few studios will go against the guidelines above on any recording they make nowadays.
They use computer algorithms to be able to verify what they are doing works out on a CD, SACD, DVD-A or the like, and you are playing them back.
Even uncompressed recordings (which is what some of us make) are considered nowadays to be "unlistenable".
The microphones themselves at 48V phantom, will struggle at much below 60-70S:N, which is why proper ones using 130V such as B & K Pro HAVE to use compression, because at 100dB S:N they exceed all the resolution of 16bit audio.
Being as 99% of speakers are actually incapable of reaching the resolution of 16 bit audio, that leaves 1% left to make any audible difference at 24bit.
It's quite staggering the amount of rubbish brought out on audio forums the moment the "black arts" of amps are discussed.
In the end they depend equally much on how things are recorded, not how mediocre the reproduction system is.
Some of us still listen to music recorded back in the days when compression was only applied lightly, as an engineering tool. The average level can be significantly below the peak level. This is how real music sounds in a concert hall.
0dB is a ratio, not a level. However, assuming you mean 0dB with respect to some reference level then whether this makes an amp clip depends on how much gain is present.mullered said:
Resolution may be the key
Many tracks are recorded (specially older ones) with their peaks at silly -6/-10 dB.
Amp designers are aware of this, so most give max. output at few hundred millivolts, and not the digital eq. of (0dB=2V RMS). So they depend on the volume control to facilitate for this.
But the question of volume control is a little psychological as well as technical.
In my experience, the more resolving an equipment is, at low level information, the less you feel the need to pump up the volume!
We tend to wind it up, when we do not hear what we want to hear, similar to asking someone with a muffled speech to talk louder!
I am not disputing most of the argument on this thread, I am suggesting, the amp with better volume, may be better resolving.
Haven't you noticed, we can be happy with a top-notch 8W amplifier, while an average 150W amp may not be enough?
Very true.
Many tracks are recorded (specially older ones) with their peaks at silly -6/-10 dB.
Amp designers are aware of this, so most give max. output at few hundred millivolts, and not the digital eq. of (0dB=2V RMS). So they depend on the volume control to facilitate for this.
But the question of volume control is a little psychological as well as technical.
In my experience, the more resolving an equipment is, at low level information, the less you feel the need to pump up the volume!
We tend to wind it up, when we do not hear what we want to hear, similar to asking someone with a muffled speech to talk louder!
I am not disputing most of the argument on this thread, I am suggesting, the amp with better volume, may be better resolving.
Haven't you noticed, we can be happy with a top-notch 8W amplifier, while an average 150W amp may not be enough?
If you worked in broadcast or professional audio visual enviroments you would know what the levels used are.
They don't vary, EXCEPT in the case of *video mixing desks which habitually have to be fed about -6dB for unrelated reasons.
(*Many tracks are recorded (specially older ones) with their peaks at silly -6/-10 dB.)
0dB level are also what professional sound cards are referenced to, so you have at least some idea what is the correct level to be driving an amplifier to.
I don't use anything else.
No fancy DACs, no CD players on compressed air beds, nothing, just direct digital output direct from the PCM stream via a banal notebook onto balanced XLR lines.
FYI:-
Most of those 300-400EURO sound cards we are using for RECORDING 24-96 audio, so I don't see why there has to be some fancy scheme to do the reverse back to analog.
I was quite impressed with the old ADB card from 20yrs ago (and that was ISA bus!).
An externally hosted image should be here but it was not working when we last tested it.
It's no different with Lynx PCI. You can buy a good s/h DAL or other good pro sound card for 50 USD on Ebay.
You can plug headphones straight into the output perfectly fine, it doesn't need any dedicated fancy "tube amplifer"..
A standard 100 Euro computer CD player from say Pioneer or Plextor is all you need, yet we see all this "hi end" modified DACs, modified OPPO players costing 1000s and 1000s which do not make any difference whatsoever.
In fact the D-A convertors on my ancient SONY professional DAT players do a fine job of converting SPDIF back to analog, but there we open a new subject.
As for music recording and compression, there is no such thing as "back in the days"...because it brings back those fond memories of the incredibly bad S:N levels found using studio tape decks, or PCM broadcast over FM.
The only reason why you think it was "superior" is because the compressors typically used by say BBC broadcast had a stunning technical laboratory behind them, so that compression could remain unobtrusive.
You take that to mean there wasn't any.
I can assure you with the poor channel separation and S:N of FM broadcast they had to use a LOT of compression to make such stuff as the proms sound OK.
The fact is, compared with vinyl, tape or other media of the day, FM stereo broadcast was a quantum leap better.
That goes to tell you how dire and dreadful recorded music from the 60s and 70s really was.
Rose tinted glasses, valve amps and all the other stuff just makes it sound wonderful doesn't it! 🙄
I work in the concert hall (and play too), so I suppose I should know much better how it all really sounds (if of course the wind and brass and strings ever play in tune).
Last edited:
No. I saidmullered said:
DF96 said:
FM can have S/N of 70dB or more. Channel separation maybe 30-40dB? Good enough for real music.mullered said:
60s recorded music may have noticeable flaws. By the 70s things had improved. Later things got worse again.
I'm not sure which planet you have been receiving FM broadcasts on, but in the REAL world you will struggle to get even close to 30dB S:N, and never exceed a channel separation of 20dB, which inevitably in the real recording>broadcast environment from say Opera you squeeze down to a dynamic range of about 25dB.
I should know, we do it everyday here, even last night our live Ballet broadcast went out on fibre-optic with internet HR video.
Would you like to listen to my recordings of the LIVE BBC, Wagner ring cycle on reel-reel from 1978 just to check it all out?
Mind you,- that's light years better than DAB, but there again, nobody wants that anyway.
Last edited:
Unless by "S:N" you mean something quite different from signal-to-noise ratio then I can only assume you have never heard a decent FM tuner. Perhaps you mean useful dynamic range, which requires a significantly better signal-to-noise ratio if noise is to remain unobtrusive during quiet passages? Your mention of a dynamic range of 25dB suggests that you may be making this confusion.mullered said:
Sure you know best!
A soundcard from your laptop is on a par with, say, a Benchmark DAC.
The rest of us are just rich idiots - imagining it sounds better.
Nobody needs a fancy tube amplifier.
Saying that nobody can benefit from a properly designed fancy tube amplifier is another thing completely.
BTW your use of quotation marks makes me want to check my amp to make sure that it really does have tubes in it. Just to be sure.
How can 48vlts phantom not be enough for A mic that puts out 100 milii volts? That's ridiculous.
And if your talking about 0 db full scale (dbfs) say so. It only applies to digital. Analog recorders used VU and didn't care about peaks, tape compression removed those.
And if your talking about 0 db full scale (dbfs) say so. It only applies to digital. Analog recorders used VU and didn't care about peaks, tape compression removed those.
Exactly I couldn't have put it better myself. 😀
For your information my laptop has a professional sound card installed in it...
You know that old fashioned standard called PCMCIA.
There were only 2 Pro sound cards made in this interface type which has direct PCI latency and very low noise.
I don't think you have any valid lessons to give me in how to use it...Unless of course you are a production recording engineer and regularly work in live concert environments making recordings of the kind of quality you can only dream of.
Perhaps you should come and show us how to do it? 🙄
Last edited:
'fraid you don't know what you are on about.
When was the last time you saw any modern recording being done with Analog recorders?
Noise, compression, distortion all by the bucketload.
They luckily died out 3 decades ago, & with DAT we did some absolutely excellent work for 15 years.
Have you ever heard a concert done with DAT or Tascam DA?
The best recording machine we ever had was a pair of Bruel & Kjaer stuck behind the ears (Binaural style)with in a pocket with pocket phantom amp, and a Sony pocket DAT in the other pocket. (pity they are so hard to find even now!)
Zillions of pirate recordings have been done this way., you know the kind of things the "hi end freaks" rave about, but are run of the mill for us.
Long live the pirates! 😉
Dynamics and content come from the RECORDINGS, maybe you all forgot those guys have to have better ears than "hi end audiophools".
Last edited:
- Status
- Not open for further replies.