We are the only two people on forum who have been to Malvern concert hall, and that too to watch the same concert. Difficult to forget
Natural Sound
- Thread starter PeterA
- Start date
We are the only two people on forum who have been to Malvern concert hall, and that too to watch the same concert. Difficult to forget
How should one sort through the myriad of variables between tables that contribute to SQ in order to isolate and conclusively determine that "stable accuracy" is the (or even "a") major contributing factor?
I'll interject with a third parameter. Which I was trying to maneuver Peter towards.
Repeatability.
This last is what ddk targets within hard boundaries use of a RR establishes. Getting the best possible sound across the widest possible number of variables collectively or individually presented by LP. Diverting focus away from gadgets (or playing cards) to internalize what is requisite to simply putting on music and enjoying it to the utmost. Listening not hearing.
Possibly this is much closer to the discussion Peter was attempting. Which is why I'm pursuing it again. No less that a Eureka moment discovering where his first attempts at card trick or the magic belt fell short of David's prescience in these matters could be illuminating. At least topically for this system thread.
For DD it is very important the electronic control system and platter mass. Old concepts created "hunting" where the turntable was always changing speed in tiny increments of under and over correction. This would look fine on a normal tachometer but in fact is damaging to the sound...kind of an analog "jitter". By the late 70s to early 80s, JVC, Kenwood, Pioneer and others had developed more sophisticated control systems that also relied on mass (Kenwood L-07d had about 7kg platter and Yamaha GT-2000 a 6Kg one with option for 18KG one) to create smoother transitions with far less over and undershoot.
When I meant categorize I meant the four quadrants shown by micro, which I know very well as it is kind of the cornerstone of analytical test method conception and I am an analytical chemist. We talk about accuracy (in this case as close as possible to 33.33333 or 45 rpm) and precision (how is it fluctuating from instant to instant and by how much?). It seems that the variability over short time periods is likely more damaging to tone reconstruction and timing than being off absolute speed a bit (obviously if this is off by a lot things will sound wrong...think a 33 record played at 45) but very stable from instant to instant. Variability over long time intervals also seems to be less damaging to the sound quality and this is what the later DD designs from Japan (think early 80s) seemed to aim for... along with some inertia thrown in for good measure to reduce the need for sharp corrective action of a servo.
This was theorized, but was audible speed "hunting" ever proven or measured? A tachometer would not measure it but demodulating a 3150Hz tone from a test record might show this, depending on the amount frequency shift and the rate of shifting, by looking at the amplitude of the demodulated signal as well as the frequency spectrum (FFT). I don't know if "hunting" was ever proven, but if you have data on this, I would be interested.
BTW, I would categorize high speed hunting as a third separate issue, apart from accuracy or stability.
That's what I understood by your question, but I don't think you can make generalizations about any type of drive system because of the number of variables involved. I outlined what I see as some of those variables and the few generalizations that one can make, but I don't think you can draw any concrete conclusions without more specifics.
This is easier to isolate on a particular system like Peter's. If you keep all other elements the same (pulley, belt, platter, plinth, tone arm, cartridge, electronics, speakers) and make changes that improve speed accuracy and stability, any perceived improvement in SQ should be highly correlated with those changes.
Brian, one way to start would be to measure a bunch of tables. If they all measure the same or within a very tight band that is both accurate and stable, then you can illuminate speed as a factor when assessing the differences between the various turntable designs.
in other words, if speed performance is very similar(pick the tolerances) then differences one hears must be a result of different factors. One example is my turntable and David’s AS 2000. Speed performance is very similar. They sound similar but different. The increased mass and different materials may account for these differences. I am sure designers take all of this stuff into account.
Given two different tables from different manufacturers that is difficult to do by listening alone, if it is even possible. Different speed related specs might give some indication but there are numerous factors that can contribute to listening differences. In my case I had one version and its immediate successor to evaluate. The major difference between them was the motor and motor controller system.
I discussed the topic with the GPA engineers. They knew the differences in specs ahead of time, but were equally astonished with the sonic results, which were unexpected to the degree observed. We spent time talking about what could be the cause of the difference in sound. The new table was quieter with the cause of that attributed to the new motor and controller software. Alvin Lloyd claimed motor noise is impacted by how the motor is controlled. At the end we concluded that the majority difference should be a result of improved speed accuracy and speed stability first, and quiet operation second. Other minor changes were ruled out. So in part it was a process of elimination and a consideration of what was changed.
It certainly sounds like confirmation bias could be a factor there. Wasn't it the case that they were not entirely forthcoming with respect to the full set of changes between the revisions?
I don't understand what you are referring to.
I'm under an NDA but what little proprietary info I know does not change anything I described above.
Is there a point you want to make?
No, I am addressing the proper reviews by known reviewers.
Threshold of what?
Yes, several times. Using test records and SpectraPlus. I did not keep the spectra, the EMT 927 was the poorest measuring but sounded very good. In fairness I should say they my particular EMT measured exceptionally well for such turntable, significantly better than the manufacturer specifications.
Why do you insist on using the vague concept of speed stability, that GPA carefully avoid, wisely referring only to accuracy and peak or RMS variations? If we go through their site and papers we see they refer to stability as a mechanical or overall general concept, not as a speed epithet.
Just my opinion, but considering that upgrade from 1.5 to 2.0 needs such large amount of parts I can't taken as proved that speed was the major factor (quoting from the GPA site)
Standard upgrade to 2.0 includes:
- Motor replacement/upgrade
- Main bearing cleaned and modified
- Thrust bearing system cleaned and bearing upgraded
- Phosphor Bronze flywheel upgraded
- Computer rebuilt with new circuit boards and custom software
- Power supply replaced with new upgraded version
- Spindle and plinth re-polished
Data gathering. I've been interested in finding correlation between this level of speed precision and accuracy to SQ for quite some time. I still haven't seen any convincing data, but thanks for the anecdote.
If I'd known it bothered you so much I would have started sooner. I don't know what "speed epithet" means.
As examples these measurements are a across the side of a record. That level of accuracy for what? Roughly 20 minutes? Sure seems like accuracy that holds steady across a side is stable accuracy. Call it what you.
How would such correlation be expressed? What would convincing data look like to you?
Short of some sort of test bed where the parameters of interest (and only these) can be varied both systematically and randomly, I can't envision it. Too much has changed between 1.5 and 2.0 for sure, though. And IMO the only viable (marketing) justification for its MSRP and its entire point of differentiation is its extreme speed performance, so it'd be awfully inconvenient if that wasn't deemed largely responsible for its SQ.
I remember Burning Amp (2018 - was it?) where Nelson was giving out 2nd harmonic generators as door prizes. In fact, I think one of the SIT amps had these built in (the SIT-2 maybe) with a knob on the front panel allowing the listener to dial it up and back. This was a great way to isolate and listen to the effects of both + and - phase 2nd harmonic distortion with all else being equal.
1. Fair enough. The reviewers did their own testing or measuring of speed performance. You didn’t answer the question about what measuring technique they used. Nor within what tolerance
2. Threshold for speed accuracy and consistency. You did not answer the question Whether the review were’s discussed what degree of tolerance is audible or matters.
3. Relative speed performance between the three turntables does not surprise me but you did not answer the question about what the speed results were. When you say poor speed performance, what do you mean? How did it sound better than the other two turntables? In what ways?
I must say you ask for a lot of specifics from others and then don’t provide your own.
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |