Paragraphic Filter Questions

[nlkleve]

I have been using DC6 for about a year now and I love/hate the Paragraphic EQ.

I do all my restorations using a flat preamp and I find that the preset curves for this filter work very well as a rule. HOWEVER, there are times when there is no curve present for a particular recording. One such case is when dealing with old Decca 78s which use a curve rated 250N-D. (A 250 Hz crossover, no bottom shelf and -6 dB @ 10kHz with a 3 dB/octave slope).

So: How does one make their own -6dB @ 10 kHz, 3dB/octave filter??? I can see no mathematical rhyme or reason to the placement of the data points using this filter.

It also concerns me that many of the slopes of the 78 curves appear to be around 4 dB/octave instead of 6 dB/octave as they should be. This makes me question how accurate it can be to try to adjust the data points visually using this filter.

Norm

[Craig Maier]

It is a myth that Decca (or any other 78 RPM record for that matter) was recorded with any high frequency pre-emphasis. Turnover is the only response modification that is required.

Several other websites on the internet has been promoting a false EQ chart showing the various 78s as having Rolloff which is very misleading.

For an accurate listing of the various 78 RPM record labels and their turnover curves, please refer to the appendix section of your Diamond Cut users manual.

ps - the paragraphic EQ is solving up to 10 transfer function equations each having multiple filter co-efficients and gain factors wherein each and every term is interacting with each and every other term. So, it is not necessarily going to be intuitively obvious what the plotted graph is doing when entering in the parametric data (freq, Q and amplitude). And, the top band (on the right hand side) is always a shelf type filter, which further complicates the matter.

But, the graph is an accurate representation of the filters response.

pss - the DC7 version includes a "closed form" mathematical reversal of the European 250 Hz turnover curve in it Virtual Phono Preamplifier (VPA). You may want to consider trying out that version of the software.

[nlkleve]

Paragraphic EQ: Thank you Craig

Thanks for your response about the Paragraphic EQ. I have been getting better at manipulating its settings and have found that the Make Waves function in DC6/7 can be used in conjunction with the VU meters to get a pretty accurate curve.

I have switched over to DC7. (I did that the day I wrote my post on this subject.) Overall I am very happy that I have switched!

I spoke (through email) with the person who is promoting the "other EQ chart". It seems that he did a lot of research into phono EQ and the history behind it as part of a college thesis he did. He believes very strongly in the values he has posted--which doesn't mean they are correct. He did supply me with a rather extensive history of the use of more complicated EQ curves.

So I am thoroughly confused on this issue! I have tried several 78s EQed with the more complex values and they seem to provide results that are more consistent between different record companies. (Brunswick, Decca, Victor) But there is nothing that says that these companies' records should sound consistent with each other for a number of other reasons.

For now, I will explore the use of these curves with some degree of skepticism and compare my results back to the simpler curves published with DC6/7 as they have always worked pretty darn well!

Thank you again for your help on this subject and I will bear in mind your warning that these "78 rolloff curves" may not be accurate.

Norm

[Craig Maier]

Quoting nlkleve

"I spoke (through email) with the person who is promoting the "other EQ chart". It seems that he did a lot of research into phono EQ and the history behind it as part of a college thesis he did."

--------------------------------------------------------

He is wrong and should get an F on his thesis. I know that chart and it was created by a Forensics Audio engineer by the name of Tom Owen who told me that he created those rolloff settings based on what he thought sounded good to his ear. He created that chart around 25 years ago.

Sorry!

ps - you can find contact info for Tom at this link if you want some direct input on this:

http://www.owlinvestigations.com/cv.html

pss - I would wager that the person advocating that trumped up 78 EQ chart is Gary Gallo - - - am I right?

[nlkleve]

No, not Gary Galo; though I would recommend his "Disc Recording Equalization Demystified" to everyone interested in this.

The web site I have been referring to is: http://www.geocities.com/midimagic@s...t/mixphono.htm. The author's name is Larry Robinson. I disagree with the statement he makes that the acoustic recording process was essentially constant amplitude. This is just plain wrong within the limited confines of the acoustic process' bandwidth.

I also disagree with the notion of equalizing acoustic recordings at all. I'm sure that some EQ will "improve" their sound; but these discs were not meant to sound "modern". The best they can sound is the way they would have sounded when played on the equipment of the day as this is what was intended. A flat EQ comes VERY close to the way a 1920 Victrola sounds. I've compared.

Norm Kleve

[Dan McDonald]

Basically, there are two camps among those who deal who deal with 78s and cylinder transfers :

those who want it to sound like it sounded to someone who listened to it through the playback equipment of the time, which it sounds like the side you are on,

and those who want it to sound like it sounded as performed, which is the side some others (like myself) are on, (although most people probably strike a balance between those two).

My feeling is that I want to hear as much as the music as possible, even if my grandparents or great grandparents would not have heard part of the music on the record, it would have been produced by the musician, and I want to hear it on the transfer. Other people want to reproduce the music as if it were being played on the equipment of the time, and I can certainly see that as a logical stance to take also.

There really isn't a right or wrong - we all do what we want with it.

Dan

[Brian]

Basically, there are two camps among those who deal who deal with 78s and cylinder transfers :

those who want it to sound like it sounded to someone who listened to it through the playback equipment of the time, which it sounds like the side you are on,

and those who want it to sound like it sounded as performed, which is the side some others (like myself) are on...

....and Dan could have added "along with all the recording and mastering engineers over the years." Like Craig said, the choice to EQ or not is yours... and neither choice worth criticizing. However, the realistic recordings we enjoy today indicates the vote of history is overwhelmingly in favor of striving for sonic transparancy.

[Craig Maier]

Quoting Norm:

"I disagree with the statement he makes that the acoustic recording process was essentially constant amplitude."

-------------------------------------------

You are correct. In order for a system to be "constant anything" would imply that a closed loop process is being used. In other words, constant amplitude, constant velocity or constant acceleration are all parameters that can be held constant via a servo control loop containing a differentiator or integrator within that process.

Acoustical recordings are made using only a subset of the energy of the acoustical signal received by the horn. No outside energy (amplification) is provided. A diaphram producing mechanical deflections when subjected to air pressure variations is not contant anything because there is no closed loop circuit (or error amplifier) present to force it to be that way. It is simply an open loop system which will exhibit a non-linear band pass response with all of the attendant resonances that the system will impart on the process.

Constant velocity and constant amplitude came to us via the wonders of the principles of electronic signal amplification, analog filters and most importantly, feedback control loops (thanks to Blackman of Bell Labs fame). So, constant anything associated with sound recording did not hit the scene until after roughly 1924 and the birth of the elecrical era of sound recording.

[nlkleve]

Good Lord! What did I start?

First off, I didn't mean to indicate that I preferred acoustic 78s with all their noise and scratches! I too want to hear all the signal that was etched into them. I just feel that it is better to leave the EQ flat on them. Others may want to try to balance the EQ a little and that is their choice.

My comment about the acoustic process being "constant velocity" stems from the fact that this is pretty much what we experience in our world: A high-pitched sound needs far less amplitude in order to sound loud to our ears than a low-pitched sound does because our ears hear the work being exerted on them; and work is a function of velocity in part. Basically, a pitch that vibrates our eardrum twice as fast as another one will take half the amplitude to sound as loud as the lower one. Hence tweeters don't need to move a lot of air, but woofers do.

The acoustic recording process is a mechanical adaptation of the same process we use to hear sounds. Therefore, it would be expected to be roughly constant velocity in nature and certainly not constant amplitude at all. Of course, in nature, very few things are perfect.

Norm Kleve

ps: Thank you all for your comments. I appreciate all your input!

[Dan McDonald]

Good Lord! What did I start?

Some of us (especially me) just like to chime in whenever something is interesting. Others (like Craig and Rick and Brian and some more) have a lot of expertise and scientific knowledge or experience that they share.

[Brian]

Some of us (especially me) just like to chime in whenever something is interesting. Others (like Craig and Rick and Brian and some more) have a lot of expertise and scientific knowledge or experience that they share.

Dan is way too kind when he puts me in the same category as Craig and Rick --- but one of the things Dan and all us "regulars" here do seem to share is enjoyment discussing audio and learning from each other.

I also don't think you'll find another group that is more forgiving and has less "flaming". There is a ranking of communications mediums based on ambiguity - or how likely it is that messages will be misunderstood. E-mail (and other short messages like we use here in the group) ranks #1 as the most likely to be misunderstood. So, if what we're reading doesn't seem right, most of us here will first assume it's a communications problem - where in other groups the defualt assumption is the #@!% ass#$%) doesn't know what he's talking about the survival of the free world requires us to set him straight in the most unforgiving terms in our vocabulary! :-) Or, as my old first sergeant used to say, "Lieutenant, if you're going to chew somebody's ass -- make sure you're going to chomp down on flesh - otherwise it's awfully hard on the teeth!".

Welcome to the group Norm.
Brian

[DougMac]

Galo's Humor

No, not Gary Galo; though I would recommend his "Disc Recording Equalization Demystified" to everyone interested in this.

Do you know much about Gary Galo? If not, some investigation might be in order.

Doug

[Brian]

Doug,
I was unable to open your Gallo attachment. (Not sure I really missed all that much! :-)
Brian