This has certainly evolved into an interesting thread, and an educational one despite the several contradictory claims which have been made.
I've just done an experiment to try to bring it all down to a more practical level. As I mentioned in my earlier post, I have always set up anti-skating on my mm/mi cartridges so as to produce no visually detectable right or left deflection of the cantilever when it is in the record grooves, compared to when it is not in the grooves (which has always meant straight ahead, with my cartridges).
Using the high output Grado Reference Sonata which I presently have installed (compliance 20, vtf 1.5g, elliptical stylus), I carefully looked for cantilever deflection at approximately 12 different points distributed across the full length of a lightly modulated (chamber music) recording.
At no time did I see the slightest deflection.
The cartridge is mounted on a 1980's SOTA Sapphire turntable with Magnepan Unitrac tonearm. Anti-skating is set for about 2/3 of the value recommended in the tonearm manual for the 1.5g vtf (35 shot pellets in the bucket, instead of the recommended 52). As a point of interest, I had previously had a Grace F9E Ruby mounted in that arm, set up by a very good dealer using an oscilloscope, test record, and Lissajous pattern. Their setting was also about 2/3 of the recommended anti-skating for the 2g vtf used with that cartridge (52 pellets instead of 79). And I have found a similar 2/3 factor to be about right on some other turntables/arms/mm cartridges I have set up in the past.
When I have done these setup procedures in the past, modest deviations from the anti-skating settings I settled on, such as using 30 or 40 shot pellets instead of 35, produced clearly perceivable deflection.
It seems to me that this experiment, indicating no perceivable deflection at any point across the record, supports the following conclusions, as pretty much stated earlier by Neil (NSGarch):
1)While skating force may vary from one point on the record to another, the range of variation is very small compared to the nominal value of the skating force.
2)The considerable cantilever deflection that does in fact result with these or similar cartridges from using zero anti-skating force would seem simply based on common sense to be undesirable.
3)Common sense would seem to suggest that the same two conclusions apply to moving coil cartridges, except that with these cartridges the differences resulting from different anti-skating settings (including 0) may be more difficult to perceive, because of their stiffer suspensions, lower compliance, and (possibly) different stylus shapes.
Can we all agree on this, from a practical standpoint?
Regards,
-- Al