)
Pavel,
Crest factor is the peak to average ratio.
30dB peak to average is generally not found during orchestral crescendo's (when the level is loudest), but at lower levels.
I checked quite a few recordings even completely uncompressed ones and I have yet to see 20dB Crest factor at very high levels on classical (telarc's 1812 excluded of course). With most current pop music I am struggling to see a 10dB Crest factor. I think the worst I saw so far was around 3dB...
I personally feel that 20dB Headroom over 85dB suffices normally (this goes back to my days as recording engineer in East Germany), a bit more of course does no harm, just like having a big powerful motor in your car does no harm.
This means a pair of "average" 87dB/W/m efficient speakers at 3-4m listening distance needs at least 200 clean Watt's.
For a pair of 97dB/W/m efficient speakers like a pair of Tannoy Monitor Reds or the Crescendo Grande I designed for DIYHiFisupply around 20 Watt suffice.
With something like my 104dB/W/m Magnificats around 4 Watt suffice.
I would say being 3dB short (halve the listed power) is no big deal either, but being 10dB short (one tenth the listed power) is a big deal...
Ciao T
Crest factor is the peak to average ratio.
30dB peak to average is generally not found during orchestral crescendo's (when the level is loudest), but at lower levels.
I checked quite a few recordings even completely uncompressed ones and I have yet to see 20dB Crest factor at very high levels on classical (telarc's 1812 excluded of course). With most current pop music I am struggling to see a 10dB Crest factor. I think the worst I saw so far was around 3dB...
I personally feel that 20dB Headroom over 85dB suffices normally (this goes back to my days as recording engineer in East Germany), a bit more of course does no harm, just like having a big powerful motor in your car does no harm.
This means a pair of "average" 87dB/W/m efficient speakers at 3-4m listening distance needs at least 200 clean Watt's.
For a pair of 97dB/W/m efficient speakers like a pair of Tannoy Monitor Reds or the Crescendo Grande I designed for DIYHiFisupply around 20 Watt suffice.
With something like my 104dB/W/m Magnificats around 4 Watt suffice.
I would say being 3dB short (halve the listed power) is no big deal either, but being 10dB short (one tenth the listed power) is a big deal...
Ciao T
Thanks for the clarification, Thorsten. 200W of needed clean power for 87dB/W/m speaker corresponds to my real life experience.
My screenshot is Telarc, Berlioz, Symphonie Fantastique.
Last edited:
Whatever else. You can easily make a test with 100 pF capacitor vs. 1,5 m of link cable.
Why make a test with 100pF of capacitance when you say capacitance isn't the problem?
You said that an unbuffered opamp output "...will never drive the cable well." Drive the cable in what respect exactly? What aspect of a typical cable is difficult to drive save for its capacitance?
se
About 25 or more years ago, with the introduction of quiet dual jfets from Toshiba and others, I tried to interest Deane Jensen in using them. I did this as repayment of a favor that he did for me in about 1978-79, giving me an Apple compatible cassette with a useful filter design program for my Apple 2 computer. This really saved my socks in making difficult head bump correction filters for my first master recorder for Mobile Fidelity.
I offered him my input, but he rejected it, until a year or so later when my technician who knew him better, was approached by Deane about using jfets. He ended his life relatively soon after.
Parasound was using a Jensen type topology for their preamp that they produced, before my time, and THEY had substituted jfets on the input with great success. This was about 1989 or so, when they did this.
All else being equal, and with source impedances over 100 ohms, jfets are almost always a winner, transformer or no.
I offered him my input, but he rejected it, until a year or so later when my technician who knew him better, was approached by Deane about using jfets. He ended his life relatively soon after.
Parasound was using a Jensen type topology for their preamp that they produced, before my time, and THEY had substituted jfets on the input with great success. This was about 1989 or so, when they did this.
All else being equal, and with source impedances over 100 ohms, jfets are almost always a winner, transformer or no.
Last edited:
I've built the classic Demrow/Cohen preamp with FET's instead of LM394's and it was wonderful. That's why I was surprised that George Massenburg is so down on FET's, possibly just to be different?
They sent 10 right away, even with a new datasheet that looked like they were ready for a rollout. Even Google now turns up mostly only our conversations. How would anyone else even know to ask for samples?
Hi,
I asked PN about the LSJ74 (& LSJ109) around the time the LSK170/LSK389 came out and he said "we are working on it". So I asked him to send a few along if and when they had finished working.
Anyway, I have bunkered Toshiba since (lucked into a massive clearance...).
Ciao T
I asked PN about the LSJ74 (& LSJ109) around the time the LSK170/LSK389 came out and he said "we are working on it". So I asked him to send a few along if and when they had finished working.
Anyway, I have bunkered Toshiba since (lucked into a massive clearance...).
Ciao T
No. Sound.
Rugged output is needed for EMI immunity and minimizing cable influences. Most of the reviewer's objections to JC 3 sound apply to sound of the link level unbuffered opamp output stage. It is up to John if he believes or not, he might recall another hesitation he had some 40 years ago. This is similar, just different issue.
Rugged output is needed for EMI immunity and minimizing cable influences. Most of the reviewer's objections to JC 3 sound apply to sound of the link level unbuffered opamp output stage. It is up to John if he believes or not, he might recall another hesitation he had some 40 years ago. This is similar, just different issue.
Sorry for the OT,
but as a reminder in the "Fremer issue" i think some of the information openly distributed then and still now could help:
Blind Listening Letters | Stereophile.com
John Atkinson later confirmed in a newsgroup discussion this version and added that he also did the test and got 4 out of 5 correct answers for the same/different question.
The whole event made it into the BAS as a short description along an evening feature :
http://www.bostonaudiosociety.org/pdf/bass/BASS-17-02-8910.pdf
Page 6.
Personal conversation with people over the years since then never gave any reason to doubt the basic informations, so i´d still be interested to know if there is another official information about the facts.
but as a reminder in the "Fremer issue" i think some of the information openly distributed then and still now could help:
Blind Listening Letters | Stereophile.com
John Atkinson later confirmed in a newsgroup discussion this version and added that he also did the test and got 4 out of 5 correct answers for the same/different question.
The whole event made it into the BAS as a short description along an evening feature :
http://www.bostonaudiosociety.org/pdf/bass/BASS-17-02-8910.pdf
Page 6.
Personal conversation with people over the years since then never gave any reason to doubt the basic informations, so i´d still be interested to know if there is another official information about the facts.
What the hell does EMI immunity have to do with the ability to "drive the cable well"?
What cable influences exactly? I mean, what cable influences will cause an unbuffered opamp output to "never drive the cable well"? You said previously it wasn't capacitance. So just what the hell is it?
Stop talking in circles.
Oh really? And how exactly were these reviewers able to separate the sound of that from the sound of everything else?
You followed your original statement with "though it may sound silly." You seem to be doing everything you can to remove the "may" from that statement.
se
Steve
You need to get those directional silver grounded antennas mentioned before. They really will help improve things. (Somebody else's wallet etc...)
An output stage not only has to produce power to the output it also has to sink things such as local AM stations, power line noise etc. Shielding helps reduce these issues but does not eliminate them. So a robust output stage will provide greater attenuation. Particularly when there is feedback around that stage! After all that means the input is connected to the output.
Pavel,
John designed that circuit as an add in card. It just got marketed as a stand alone. He has designed stand alone product, so the issue is moot.
You need to get those directional silver grounded antennas mentioned before. They really will help improve things. (Somebody else's wallet etc...)
An output stage not only has to produce power to the output it also has to sink things such as local AM stations, power line noise etc. Shielding helps reduce these issues but does not eliminate them. So a robust output stage will provide greater attenuation. Particularly when there is feedback around that stage! After all that means the input is connected to the output.
Pavel,
John designed that circuit as an add in card. It just got marketed as a stand alone. He has designed stand alone product, so the issue is moot.
- Status
- Not open for further replies.