WTB Lathe-Seeking Advice + Knowledge [+ crucial digressions]

[JayDC]

It is my understanding that, for this, I would want a presto or Rek-O-kut lathe. It is also my understanding that the resulting sound quality will be laughable at best

this is completely untrue.. you can get a great sound from a presto... used it for many club cuts "dubplates" during the past couple of years... my 1-c head will cut a 12.5k tone easy (tested) and goes as low as 20hz (tested)... it's all how you use it, it takes a certain mix of skill and thought to get high volume level cuts that sound great, but it's not impossible..

to say you won't get quality from a presto is like saying you can't cut wood with a hand saw... you can, but it just takes more effort than a power saw...

[cohearent]

A pretty heated debate on half speed...
Let me add to it.
First of all the Neumann SX74 has almost no output at 30kHZ. It falls off very quickly above 23kHz.
I have measured many of them. There is no 60 to 120kHz on ANY phonograph records, even cut half speed Get over it.
I can say with all certainty that what Stan Ricker saw was not bias , but some other oscillation in the original recording. Recording systems have been plagued by this since the dawn of electrical recording. It IS possible that what he saw was something beating against the bias but at a lower frequency. I have seen ultrasonic stuff many times cutting full speed. It's not music, it's noise, so what is the big deal???
Any theoretical improvements in half speed cutting are completely lost in playback. Let me explain.
The biggest problem in cutting music is too much high frequency information. Vocal sibilants are the greatest offender (because they are much longer in duration than percussive transients), as are overly loud cymbals, muted trumpets, etc. (30kHz response would not be a help in cutting Harmon mutes, it just increases the groove curvature which adds to mis-tracking). As soon as the velocity reaches, say, 30 to 40 cm/sec it is going to mis-track. (At 50cm/sec it's all over on the OUTSIDE of a 12" 33, the stylus moves so fast it cuts back on itself and the back edge wipes out what the front just cut. And it gets worse as the groove moves inward.)
In order to cut these things and MAKE THEM PLAYABLE Neumann, Ortofon and others built very fast HF limiters to reduce these peaks to the point where they will play back cleanly. These devices don't work at half-speed. There isn't enough energy at half the frequency to make them react properly no matter how much you fiddle with the time constants. (You could do it digitally, but the audiophiles want their records all analog, thank you.) Remember that cutting half speed does nothing to improve this. It's a playback problem. If it did cut more accurately, which it doesn't, the resulting groove would be worse, not better. (The Neumann paper, which many are in denial about, pointed out that lacquer springback is worse at half speed.)
And yes, the octave loss on the bottom is audible. Particularly from analog tape. And the cutterhead is running out of feedback and begins to sound underdamped. You make the comparisons, I have.
I have done full speed shoot-outs with half-speed cuts, repeatedly, and have never lost one yet.
And many engineers will testify to this! Ask Allan Sides or Jack Joseph Puig.

Best,

Kev

[mossboss]

And dear Sir Kevin
Thank you for an insightful post again
Confirming the Neumann paper as well as making the point of spring back on the cut, which in my view, was most likely the biggest objection in so far as they where concerned, as other technicalities could have be dealt with, not that they wanted to travel down that path either was the reason for that paper
I never had much time for half speed and will confirm most points in your post however i could not have put it as succinctly or as well as you have
I never thought any one with a well tuned as well as to spec cutting lathe doing at proper speed would ever loose in a shoot out with a half speed cut either
My own straight out comparisons always leaned on the proper speed cuts weather it was playing back the lacquers or the resulting records from both half as well as normal speed cuts
Being mindful of the fact that our game which it is full of smoke and mirrors, with claims of cuts at plus 12 dB or more one wonders what these people are doing
It seems that standards have gone out the door as every one wants to be "different"
Not sure about it if one needs to be diplomatic? Sensitive? Accommodating?
Argh forget it, the facts are the facts they will always remain as such
Good to see you chipping in.
I am sure we will hear more about it on this very thread no doubt
May be what they may say about it is something like "opinions been like whatever it is that everyone has one" forgetting the facts
Argh well
But having it from the very people who made these machines apart from been verified by countless of others they persist in flogging this very dead horse so no doubt we will have some more fruitless debate on the subject.
Thanks again
Best regards

[opcode66]

I'm still trying to find some things after the move. So, I have no idea where my copy of the half speed bulletin is at. I also don't remember I had a physical copy or a digital scan. Needless to say it does exist and I quoted from it back in March of 2011.

https://www.lathetrolls.com/viewtopic.php?f=11&t=2584&hilit=half+speed

Excerpt below:

The most significant points in the half speed cutting bulleting:

“We know of no technical reason which would move us to recommend this process to anyone.”

“those which tend to emboss or “push” the lacquer – appear to be increased at half speed.”

“We furthermore believe that the loss of one octave at the bottom of the range is significant enough to cast doubt on the feasibility of this technique.”

The other interesting part is at the end. They show you the eq curve you need to apply in order to use the RIAA built into the amp rack while cutting at half speed. No mod to the RIAA in the amp or pitch computer are necessary as long as you build the equivalent passive eq circuits and feed your preview and program signals throught them.

[mossboss]

Todd
I emailed you a digital copy long ago. C.

[opcode66]

Well, then I might no longer have it. I lost a fair amount of emails a while ago. That was one of them. Unless I saved it to disc from my email I might no longer have it.

[opcode66]

I found it! Do you mind if I transcribe and post it?

[cohearent]

It's great with me if it's OK with Steve E. Maybe in the technical archives? -Kev

[mossboss]

Hello Todd
In view of this debate I would have thought that it be very appropriate
I asked you to hold it back so we do not have another never Ending debate about half speed as I was sure that it would have caused one
Its existence has already been confirmed already by Kev, interestingly enough also referring to the doubters been in ."denial" so we would be doing them a great service in the matter
I am sure they would like to insist that half speed it's a great improvement but as some/ most of us know it is not
Nevertheless the pundits of this technique certainly would like to present themselves as wiser than the very people who produced these machines, suggesting that there is no need or requirement for half speed cutting, in actual fact it is quite the opposite, detrimental to the cut
There is no doubt in my mind that in their endeavours the people at Neumann would have tried out half speed double speed and any other technique feasible so as to improve on an already well working system
I do maintain that Gotham would have pushed the point within the sales dept at Neumann as i am sure the engineers had the view as it was expressed in that document
So it seems that it was a case of them been pushed in making some of the components to archive half speed despite the circular as well as their reluctance to go down that path.
It also seems it was a case so as to get sales from the existing user base, so they most likely relented
I have no evidence of this apart from plain educated guess,
Most likely went something like,
Hang on, these people are asking about it, they want to do it, they want to give us their money, so why don't we just satisfy them! suppress the document circulated, don't worry about it, forget the facts give them what they want
The people cutting on the few Neumann machines here in Australia in the past,never embraced half speed cutting that I know of, if there are any they would certainly be in the rare basket
Any one here knows of any releases or record issues by major labels on half speed?
One company EMI here went down anothere route, the " maxi cut" by modifying some of the existing electronics it was done down under not in the UK and it seems that it was not adopted elsewhere
He is a link to the thread
http://forums.stevehoffman.tv/threads/emi-maxicut-record-mastering-vinyl-cutting-process.102348/page-2
And so it goes in our game which is littered with all sorts of true's half true's and straight out untruths
Such is life than
Any way Todd,
It would be nice if the document it is intact as it was scanned, so if it is possible to post it up as such than it would have the original heading as well as logos etc else we may be accused of "forgery"
I am still looking for the original it is somewhere in a folder with all sorts of Others interesting correspondence which has been buried amongst other stuff
I am sure it will surface as we are sorting out all sorts of things after a major move
Thanks
Cheers
C.

[cohearent]

Just to clarify one point: My comment about "denial" was in reference to the existence of the Neumann document, not the ongoing "debate" here. -Kev

[mossboss]

That's quite clear Kevin. The document has now been located so it would posted as a PDF as well as a transcript By Todd ASAP
Cheers

[opcode66]

Here it is. The PDF scan. And, the text transcribed so it is searchable here on the site. With any hope, this can put some of this back and forth to rest.

----------

Gotham Audio Corporation

Engineering Bulletin 0181-ENG111-p6

RE: Half Speed Cutting of Masters
From: Stephen F. Temmer, President
Date: 27 January 1981


Hardly a day goes by without someone calling us about "half speed cutting". The conversations vary little one from the next. Here are some of the salient features:

1. "I realize it's a lot of bunk, but my management insists on going into it."
2. "It seems to make sense that anything done more slowly, is also done more carefully."
3. "All we need is a half speed adapter for the lathe."
4. "We understand we have to install some new playback amps for our preview tape machine."
5. "We understand you sell a half-speed converter for our cutting rack."

We've given all these examples in the hope that some of them will "strike home". Here is the NEUMANN/GOTHAM position on half-speed cutting relative to the above points:

1. Half-speed cutting has been around ever since the advent of stereo disks, when cutterheads were unable to absorb the heat generated by the cutting levels then in demand. Our 1966 issue of "SOUNDWAYS" magazine had a page of copy on the subject. It is our present considered opinion that both the cutterhead and its drive system are sufficiently generous in their power handling capabilities as to make half-speed cutting for this reason totally unfounded. We furthermore believe that the loss of one octave at the bottom of the range is significant enough to cast doubt on the feasibility of this technique. We know of no technical reason which would move us to recommend this process to anyone.

2. No aspect of disc cutting appears to present any reasoning analogous to this. On the contrary, the non-cutting aspects in lacquer - those which tend to emboss or "push" the lacquer - appear to be increased at half speed.

3. This is not true. You do need a half-speed capability for your lathe, but you also need equalization specific to this task. If you own an AM 32(b) model, you will need a ZA 33(a) Half-Speed Converter. One of these is still on hand. When it is gone it can be simulated from standard parts. If you own a VMS 66/70 Model, you will need a PS 66 Plug-in Programmer corresponding to the rpm and diameter involved: eg. 33 1/3-12". These are available and cost close to $ 1000. If you own a VMS 80 Model, the half speeds of all standard speeds are included on every lathe.

4. There is nothing to be installed in your preview tape machine! All it must be capable of, is the half speed of the recorded tape speed.

5. Your cutting rack, be it a VG 66S, VG 74 or SAL 74, is fine the way it is and needs no further equipment installed in it. Normal RIAA is used in half speed cutting.

At this writing we do not have available the equalizer which is required to be installed in all four channels (two preview and two playback) of the Transfer Console. We enclose a graph which shows the required equalizers. Note that it ends 7.5 KHz and should drop off as soon above this frequency as possible. Four of these are needed, since both the cutting system and pitch/depth control systems must be fed the identically processed signal.

Note also that it is necessary to use special half-speed models of any noise reduction system used!

[opcode66]

I have a bulletin form Neumann where they discourage half speed cutting for two reasons. It doesn't really have a tremendously noticeable effect. And, you loose the lower octave. Straight from the horses mouth.

Hey Todd,

While I can certainly appreciate attempting to further knowledge on this valuable forum, there are a couple of things going on here that I wish to take exception to.
Firstly, (is that a word?) Halfspeed mastering does indeed have a “noticeable effect”. I myself have cut over 60 kHz to lacquer with our system. I never was a math major, but I believe that’s over 120 kHz in real-time playback.

Well duh. That is about the only advantage that I can see is the ability to cut higher frequencies at higher volumes without worry of smoking your cutterhead.

Secondly, “loosing the bottom octave”. Sweet Jeebus. I’m not even going to address this.

Propagating this kind of misinformation only fuels the ignorant reviewers and internet trolls (sometimes the same people, I’m convinced!). Look, the interweb is like a vast echo chamber. Lazy people reiterate poorly researched information as “facts” and this garbage gets endlessly repeated.

I appreciate how confident you are in your assessments. However, I am not ignorant. I am also not lazy. Would you like me to explain how this works? When you run audio at half speed you shift everything down one octave. That is a fact. Since the cutterhead on a Neumann can only reproduce frequencies of 20hz to 20khz then it is obvious that shifting one octave down will truncate audio data. That is not speculation. It isn't even a hard concept to grasp. 40hz becomes the new 20hz at half speed. And, 20hz becomes 10hz. 10hz is not reproducible by the cutterhead. Therefore that data is lost.

Look, you can post whatever you like on your attempts to cut Halfspeed. That’s your prerogative. However, just because I’ve driven a car over 120 MPH, I’m not going to speculate what it’s like to drive at NASCAR events. I leave that up to the professionals.

I am a professional record cutter. I cut on a VMS 70 fully restored. VG 74 B amp rack. SX 74 cutterhead. I've repaired and modified my own gear. I know the VMS system better than a lot of folks who have operated one for much much longer than I've owned mine. I've automated my lathe. I've created halfnut clones for sale. I've had my own diamond cutting stylus produced and am selling them. So, to say I'm not a professional record cutter is clearly not valid.

I’d love to see that bulletin from Neumann that you refer to, but I don’t think that it’s “straight from the horse's mouth”. I suspect that it’s from somewhere south of there.

The scan is posted. I've transcribed it to text so it is now searchable on the site as well. Please read it for yourself. I wouldn't mind an apology while you're at it...

[markrob]

Hi Todd,

You stated:

I appreciate how confident you are in your assessments. However, I am not ignorant. I am also not lazy. Would you like me to explain how this works? When you run audio at half speed you shift everything down one octave. That is a fact. Since the cutterhead on a Neumann can only reproduce frequencies of 20hz to 20khz then it is obvious that shifting one octave down will truncate audio data. That is not speculation. It isn't even a hard concept to grasp. 40hz becomes the new 20hz at half speed. And, 20hz becomes 10hz. 10hz is not reproducible by the cutterhead. Therefore that data is lost.

You also seem very confident of your yourself in your assessments. Also a bit condescending given your level of expertise.

In fact, a moving coil head has amplitude response down to DC. If you apply a DC current to the cutterhead coil, the stylus will deflect proportional to the current. The cutterhead feedback loop is closed vs. velocity. Given this, the velocity response does indeed fall off at low frequencies once you reach the point that the feedback goes away. It has to since the there is no output from the feedback coil if it is not in motion and a constant amplitude response equates to a falling velocity response at -6db/oct. While you claim not be to lazy, its clear you do not, in fact, understand or have taken the time to learn the physics of a moving coil cutterhead. As I stated in my earlier post, the head is not the limiting factor on how low in frequency you can cut. It is limited by the driving electronics. Please correct me if I am stating anything here that is in error.

Thanks for posting the Neumann document. It clearly states their OPINION on the subject. It also provides some info to back up the claims. I think in the context of the state of the art, this advice makes sense for the average user. I also think that at the end of the day, the benefits of half speed mastering might be small given the state of the art that Neumann has achieved. In spite of this document, they have provided a solution for end users to do half speed mastering in a way that allows for easy switchover from full to half speed modes. To do this, they insert correction filters to deal with the EQ errors that result from playing back the tape at half speed and using the standard RIAA EQ. This is a very practical way to approach the problem. However, it suffers from the fact that they do not attempt to extend the low frequency bandwith of the electronics chain. Might this not result in the "lost ocatve"?

I can't speak for MFSL and I am not, in fact a professional cutter. But, I believe (my OPINION) that if you were to custom modify and/or re-design the the entire electronics chain from the tape machine to the drive amplification, you would not suffer from a "lost octave". This would require a commitment to half speed mastering as the method of choice. I suspect that this is the path that MFSL has taken. I be interested to hear from them (if we have not scared them off the forum) on the matter.

Mark

[cohearent]

Hi Mark,

No, the cutting electronics are not the limiting factor. The SAL74B is 3 dB down at something like 2Hz.
Because a velocity device responds to DC doesn't mean its output is flat anywhere near there. Any loudspeaker, including a tweeter, will respond to DC, but you wouldn't ask it to produce 20Hz. The cutterhead resonance is about 1kHz. If you look at the response without feedback it falls off rapidly either side of that frequency. Even factoring in feedback, Neumann incorporated shelving EQ at the top and bottom to compensate for the head running out of feedback above 17kHz and below 20Hz. I believe that becomes an issue at half speed.
The other big issue is the tape machine when cutting half-speed from analog tape, which is what Mo-Fi is doing, since you mentioned them.
And yes half-speed bottom sounds under-damped and lacks punch to my ear and many others that have made the comparison. I have seen half-speed hyped for 40 years as some kind of audio pancea, and it isn't, for all the reasons I have mentioned, here and in my first post on this topic. I'm not trying to take a potshot at anyone.

[markrob]

Hi Mark,

No, the cutting electronics are not the limiting factor. The SAL74B is 3 dB down at something like 2Hz.
Because a velocity device responds to DC doesn't mean its output is flat anywhere near there. Any loudspeaker, including a tweeter, will respond to DC, but you wouldn't ask it to produce 20Hz. The cutterhead resonance is about 1kHz. If you look at the response without feedback it falls off rapidly either side of that frequency. Even factoring in feedback, Neumann incorporated shelving EQ at the top and bottom to compensate for the head running out of feedback above 17kHz and below 20Hz. I believe that becomes an issue at half speed.
The other big issue is the tape machine when cutting half-speed from analog tape, which is what Mo-Fi is doing, since you mentioned them.
And yes half-speed bottom sounds under-damped and lacks punch to my ear and many others that have made the comparison. I have seen half-speed hyped for 40 years as some kind of audio pancea, and it isn't, for all the reasons I have mentioned, here and in my first post on this topic. I'm not trying to take a potshot at anyone.

Hi Kevin,

My understanding of the open loop response of a moving coil head is that it rises at +6db/oct below system resonance and falls off at -6db/oct if you consider its response in the velocity domain. If you look at the response in terms of amplitude, the head is flat from DC to system resonance and falls off at -12db/oct (just a different way of looking at the same response). You can convert from amplitude response to velocity response by taking the first derivative of the amplitude response (v(t) = ds/dt). This is what the feedback coil does as it only responds to velocity.

You can't really cut at constant velocity at very low frequencies as the excursion required increases to nonsensical levels. The RIAA curve reflects this below 500hz as it, in effect, transitions the cutting method from constant velocity to constant amplitude below this point.

Your comment about the fact that a speaker responds to DC but does not produce acoustical output down to DC is correct, but when cutting, we are not moving air via a piston, but instead, deflecting a stylus from side to side. The head is controlled via velocity feedback loop with the system resonance is in the 1Khz range for the Neumann head, you can only apply 20-30db?? of feedback around the loop before it becomes unstable (this is due to excess phase shift at the high frequencies from secondary resonances, feedback time delay, electrical roll-offs, etc). With only 20db of loop feedback, the head would start to loose control at around 100hz on the low end and 10Khz given a 1Khz system resonance (these limits move as you increase the feedback). So as you stated, there needs to be some corrective EQ applied to flatten the head response vs. velocity. Or you can use the fact that the head response is falling at -6db/oct as part of the RIAA curve. Then you only need to add EQ to get the 50hz turnover and some some additional EQ to get the 500hz turnover correct. If you are dealing with half speed, you have to re-think the correction since the head still has the same closed loop response. Not sure what MFSL does to handle this.

I'm not sure why there would be any under damped response of the head at low frequencies once its out of feedback control. The main resonance is damped by the feedback. Is there some other low frequency spring mass system in effect down there that am not taking into account?

I don't think you are trying to take any pot shots here. I also respect the fact that you have done some comparisons of the two methods. Are the differences you detected due to a fundamental flaw in half speed mastering that cannot be overcome or due to poor implementation?

Mark

[cohearent]

I think it is a fundamental flaw, not implementation, since the effects I heard in doing comparisons were basically the same, and in two entirely different systems using different electronics. JMHO.
As stated in "the paper" half speed was developed in the days before high power amplifiers existed to cut high level 7" 45s, 1/4 the power, inverse square law,etc. That made perfect sense. And by the way, the way they high frequency limited them for cutting was to make a tape copy with the limiting on the copy, then transfer the copy to lacquer half-speed. Other than doing it digitally, that would be the only way to get a really sibilant tape on disk, at half speed, and have it play back cleanly at full speed. In other words, the sibilants were actually worse, on the half-speed cut, if not corrected before cutting.

[mossboss]

Well as I thought, this was quite an innocent thread when it started
It somehow went the half speed way for better or worst we even touched the DMM and we did get some interesting input but no follow up apart from a very nice way out of the debate that "DMM may have"
I emailed this paper to Todd more like 2 if not more years ago and I asked him to not posted up as it would open a can of worms on the never ending half versus normal speed cuts
I have always maintained that half speed with a 74 was not an improvement rather the opposite however having followed the half speed wars for years I could see that the proponents of the technique would not let go regardless of whatever argument was presented to them
As Mark as well as Kevin are exchanging views on the point it is no different to me, any way, the arguments presented have been around since Adam was a boy
At the end of the day it will remain as such a to and thro argument that really comes down to the simple fact that people wish to stick with a particular point of view regardless of whatever opposing view is presented to them including circulars from the very North, He
At least this paper certainly created a quite interesting thread for the people here who may want to investigate the merits if they wish to go down that path
There are enough technical reasons presented here for one to be sufficiently informed
By the way Mark, Why would any one be scared off this forum?
We are not doing any more than having a civilised debate on a subject as old as a VMS 74
As the circular states there is no need for the new systems from Neumann to go down that path following an interim solution on a previous system that was not capable of delivering the goods
In so far as MF is concerned I have this to say
Assuming that these people have overcome the limitations by some means not obvious to us here or not known by the rest of the cutters out there, well that's great, they most likely want to keep it to themselves which is fair enough in my view
If however there is no such situation than the points raised here still stand in the absence of valid reasons of what they have done to overcome the shortcomings
At the end of the day building a better mouse trap is what is all about. If they have it good luck to them Should we all go back to the drawing board redesign a cuter head so as to do it?
And assuming this be the case and they still cut with a standard SX 74 head, well! What can one say?
We than have the same situation as Kevin says for the last 40 years
The electronics are more than good enough as it is known by all technical people who went down the half speed route as well as confirmed by Kevin here
So where is the solution to the fundamental problem?
And Mark I will accept the withdrawal of your remarks at an earlier post, i did not reply as i could see no point He he

Cheers