andy_c said:
No different than if you implemented the same filter digitally. They would both ring like bells. The easiest way would be with an IIR. But most people use a linear phase FIR filter with symmetrical coefficients that puts half of the ringing *before* the transient. Is that an improvement? Not in my experience.
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
"There's no such thing as a free lunch."
Edit: PS -- the Butterworth would *by definition* have no overshoot in the frequency domain. That is why it is called a "maximally flat" filter.
But with linear phase FIR brickwall filters, the overshoot is bounded to a rather small percentage. With an analog Butterworth filter or its emulation using IIR in the digital domain, the overshoot gets worse and worse as the order gets higher. It's unbounded with increasing N.
Edit: I'm referring to time domain overshoot here.
Edit: I'm referring to time domain overshoot here.
I don't expect information above 20KHz to be properly recorded, WITH CD.
Be careful, I have been dealing with this sort of thing for a long time, and I don't appreciate 'cheap-shots' that confuse the issue.
Charles, just read your White Paper, brilliant! I wish that I had your resources, but you have done your 'homework'. Dr. Craven and his deceased associate Michael Gerzon have been invaluable to put an intellectual framework on what we hear, for decades. They had me listening to 78 records, back in 1975, showing them to be more 'natural' sounding than anything produced up to that time, and that included Dolby and LP recording examples.
Be careful, I have been dealing with this sort of thing for a long time, and I don't appreciate 'cheap-shots' that confuse the issue.
Charles, just read your White Paper, brilliant! I wish that I had your resources, but you have done your 'homework'. Dr. Craven and his deceased associate Michael Gerzon have been invaluable to put an intellectual framework on what we hear, for decades. They had me listening to 78 records, back in 1975, showing them to be more 'natural' sounding than anything produced up to that time, and that included Dolby and LP recording examples.
Has anyone seen this paper? It is interesting in that it got a perfect null result
Karou and Shogo “Detection of Threshold for tones above 22kHz.” – Convention paper 5401 presented at the 110th Convention, May 12-15 2001, Amsterdam.
The authors presented 13 subjects with a test signal consisting of a 2kHz tone combined with odd order harmonics, both sonic and ultrasoniics The ultrasonic harmonics were switched on and off at a 2Hz rate. ALL subjects could discriminate the ultrasonics when the combined signal was presented through a single loudspeaker. NONE of the subjects could discriminate the ultrasonics when each ultrasonic harmonic was reproduced from a separate speaker.
Karou and Shogo “Detection of Threshold for tones above 22kHz.” – Convention paper 5401 presented at the 110th Convention, May 12-15 2001, Amsterdam.
The authors presented 13 subjects with a test signal consisting of a 2kHz tone combined with odd order harmonics, both sonic and ultrasoniics The ultrasonic harmonics were switched on and off at a 2Hz rate. ALL subjects could discriminate the ultrasonics when the combined signal was presented through a single loudspeaker. NONE of the subjects could discriminate the ultrasonics when each ultrasonic harmonic was reproduced from a separate speaker.
Charles Hansen said:
If you factor a high-order Butterworth filter into second-order terms, some of those second-order terms, if implemented as a separate second-order filter, can have considerable peaking in their frequency response. However, these are combined with other second-order factors that have sagging in their frequency response in such a way that the combination is maximally flat.
Any second-order filter whose complex poles have an angle greater than 45 degrees with respect to the negative real axis will have peaking in its frequency response. That's because Q = 1/(2*cos(theta)), where theta is the angle of the pole with respect to the negative real axis, and Q is the Q of the second-order filter that implements this complex conjugate pole pair.
Cartridge loading, test records and over 20 KHz content:
There are a lot of strong opinions on these subjects here but being of a experimental bent I prefer real world testing to speculation and simulations based on speculations. Reference Recordings is preparing to release some vinyl again and they are testing various mastering chains. In discussing this I proposed making a test record to evaluate the chain objectively. My ulterior motive was related to a phono preamp I am designing. If I provide adjustments for equalization, resonances etc. a user needs a way to make those adjustments. Thus a need for a test record. Now that I have the opportunity I'm interested in good ideas for test signals that will help make for a better playback experience.
The cutting chain can handle extended response, BUT not at very high level, that's a good way to fry a cutter head.
This may seem a little off of the core Blowtorch discussion focus but attention to detail extends to both ends of the chain.
There are a lot of strong opinions on these subjects here but being of a experimental bent I prefer real world testing to speculation and simulations based on speculations. Reference Recordings is preparing to release some vinyl again and they are testing various mastering chains. In discussing this I proposed making a test record to evaluate the chain objectively. My ulterior motive was related to a phono preamp I am designing. If I provide adjustments for equalization, resonances etc. a user needs a way to make those adjustments. Thus a need for a test record. Now that I have the opportunity I'm interested in good ideas for test signals that will help make for a better playback experience.
The cutting chain can handle extended response, BUT not at very high level, that's a good way to fry a cutter head.
This may seem a little off of the core Blowtorch discussion focus but attention to detail extends to both ends of the chain.
scott wurcer said:
I [and others] have been saying sounds beyond 20khz audible for years.
I have experimented with harmonics effects to 45khz.
beyond might be audible, but 45khz is all i've checked.
1audio said:
The IRE Transactions on Audio 1962 has a good paper on the design of the CBS Labs STR-100 test record. I have the whole set. All my comments are based on actual measurements and I have come to agree with DR. Geddes that this type of distortion, low order and decreasing with amplitude just does not matter much.
People, even 'researchers ' amaze me. It was stated in a relatively unknown book: 'Audio Quality' By G. Slot back in 1971, ironically also published by Philips Paperbacks, shows that it has been known for MANY DECADES that steep filters tend to be audible, if beyond 2'd order. "The effect of the response sounding brighter or shriller than when the characteristic is straight starts when the roll-off is more than 6dB/ octave. Unfortunately, we know of no complete explanation of this phenomenon nor of an exact qualitative analysis" P 52.
Did the laws of physics change, when digital was born?
Did the laws of physics change, when digital was born?
john curl said:
Nothing ironic here, G. Slot worked for Philips. He was my teacher late 60's when I attended the company college. I didn't know about this book. I guess you're not going to sell it...?
Jan Didden
Demian, I want to insert something before everyone gets confused.
You are working at a SOTA (not the company) level and I am VERY interested.
However, many here are either too young or indifferent to understand vinyl reproduction very well. Details that you consider very important, are usually washed out in understanding the general response of a phono cartridge, BUT could be invaluable when setting up a recording/ pressing process, producing SOTA quality reference discs.
We also pay special attention to the phono reproduction process, re. the RIAA curve so that every recording is given a 'chance' to have the highest fidelity possible, for the same reason.
However, in reality, records made over the last 60 years have different actual frequency responses that would probably scare the heck out of most of us, IF we could actually measure it, BUT many of these records, especially some of the very old ones, can sound DARN GOOD on a very high quality phono playback system, so exact frequency response does not to be as important as electrical processing quality, and other factors. It is just a reality that we experience all the time.
Phono cartridge frequency response, either mm or mc is a fairly well understand mechanism, and EVEN MC cartridges, made in 3 different countries had almost the same frequency response from 20Hz to 20KHz. The differences were trivial, compared to other factors such as mistracking or just overall sound, and it was 35 years ago, when I made those measurements. This should be kept in mind in this discussion.
You are working at a SOTA (not the company) level and I am VERY interested.
However, many here are either too young or indifferent to understand vinyl reproduction very well. Details that you consider very important, are usually washed out in understanding the general response of a phono cartridge, BUT could be invaluable when setting up a recording/ pressing process, producing SOTA quality reference discs.
We also pay special attention to the phono reproduction process, re. the RIAA curve so that every recording is given a 'chance' to have the highest fidelity possible, for the same reason.
However, in reality, records made over the last 60 years have different actual frequency responses that would probably scare the heck out of most of us, IF we could actually measure it, BUT many of these records, especially some of the very old ones, can sound DARN GOOD on a very high quality phono playback system, so exact frequency response does not to be as important as electrical processing quality, and other factors. It is just a reality that we experience all the time.
Phono cartridge frequency response, either mm or mc is a fairly well understand mechanism, and EVEN MC cartridges, made in 3 different countries had almost the same frequency response from 20Hz to 20KHz. The differences were trivial, compared to other factors such as mistracking or just overall sound, and it was 35 years ago, when I made those measurements. This should be kept in mind in this discussion.
john curl said:
That book is actually from 1964/5. I doubt if the context is 22kHz anti-imaging filters.
Crossover is specifically not decreasing with level and must be treated separately. A freind described his grandfater's cascaded push pull 45's with interstage transformers and minimal caps. 1930's technology that I'll bet with a decent PS would sound pretty good today.
I am shocked, shocked I tell you, that you do not have Slot's book! Try to find it, used. It's possible, I did it.
scott wurcer said:
Hi Scott,
you might be interested in this one:
http://jn.physiology.org/cgi/content/full/83/6/3548
Oohashi et al. reported just the contrary of the above paper.
His team used gamelan music samples instead of synthetic test tones, seperated the spectra by filtering and presented the filtered and full versions via a pioneer loudspeaker system with additional super tweeter to avoid the mentioned intermodulation effect.
The paper was particular interesting as the team not only did subjective evaluation based on the response from listeners but used EEG and PET scans too to establish a more objective basis for conclusions.
john curl said:
John, now you got me confused allright. If that freq response is either not that big a deal or horrible anyway, why would be fuss about that last 0.1dB RIAA conformity??
Jan Didden
john curl said:
Jan, I use the Abebooks.com used book sellers cooperative web site. There are several there for as little as $10 with shipping. I have found things there that I had given up hope on.
Abebooks has it too, plus another one from Slot:
http://www.abebooks.com/servlet/BookDetailsPL?bi=304517232&searchurl=an=slot&kn=Audio&x=67&y=18
Jan Didden
http://www.abebooks.com/servlet/BookDetailsPL?bi=304517232&searchurl=an=slot&kn=Audio&x=67&y=18
Jan Didden
- Status
- Not open for further replies.