How important is low W & F performance anyway?

The Monaco Gran Prix 3.0 says their DD table W&F is so low it can't be measured (odd, because the previous DD table published a 1PPM W&F spec even though I've never seen an instrument or software with that kind of resolution).

Maybe the reason these numbers are meaningless is because the W&F of the recording process would most likely swamp any W&F contribution from a decent table.  Even the first DD example (0.8%) which is undoubtedly worse than the cutting lathe would not add a significant amount of temporal distortion.

Are we witnessing what happened in the 70's & 80's with amplifier distortion figures that got vanishingly smaller yet the sound was indistinguishable between competing components?

I have SOTA's Quasar with RoadRunner and Eclipse Motor/ Controller.   It is an awesome table.  Sounds fantastic.   It's obvious speed stability lends to it's great sound I believe.  If there is W&F it is not audible. 

A great table for the money.  Love it.  

Raises several questions:

• At what level is it noticeable?
• Does system transparency affect noticeability? 
• Does listening ability affect noticeability?

@OP. There are two factors at issue here. The first, if I understand you correctly, is that you claim that your measurements differ from the manufacturer's published specifications - please correct me if I am wrong in that.

But it would be helpful if you could go into more detail about your measurement protocol. I presume you have an NDA with the company you were working for but it would also be helpful if you could provide details of the other two turntables - their manufacturer specification and your measurements.

The second thing is that like everything in audio, singling out specific performance measurements can be misleading because system performance  (i.e. the total performance of the device) is what really matters.

@yoyoyaya :  All of the tables in question have encoders attached to the platter, used for feedback to the motor controller.  This gives a very accurate picture of the actual platter movement and does not suffer from eccentricity or other defects of an LP test record.  I used proprietary polar plot software to measure and plot the frequency stability of the signal from the encoder which uses the 2 sigma method for computing W&F.  I also used Multi-Instrument software which has both RMS and 2 sigma which produced similar results to my software.  I have a piece of shareware W&F meter, a Leader LFM 39 W&F meter and an Agilent 35655A dynamic signal analyzer for spectral analysis.  All of them produced similar results.

The one I worked on used an optical encoder with 360K PPR;  the raw signal produced 50kHz at 33 1/3 RPM so I ran it through a divide by 16 circuit to produce 3125Hz which could then be analyzed by conventional means.

The first of the other DD tables used a magnetic encoder with 2560 PPR;  the raw signal produced 355.555 Hz at 33 1/3 RPM.  As I noted above, their published spec was 0.008% (no method given), but it measured ~0.8%.  It uses a coreless motor but I observed an 88.88 Hz lobing superimposed on the W&F signal which seems to correlate with the 160 magnetic poles on the encoder, in other words, it was cogging because of the magnetic encoder.  The motor controller used by this mfr has development software for programming and tuning the controller with the ability to monitor the motor performance and it showed the same cogging anomally.

The second DD used an optical encoder with 10K PPR; the raw signal produced 1388.88 Hz at 33 1/3 RPM.  Their published spec was 1 PPM (0.0001%) but it measured ~0.16%.  The motor was a NEMA17 BLDC motor with iron core stators which of course, will have cogging between the PM rotor and the pole pieces.  This was clearly visible on the plot.