Nittaku Premium balls in

roundrobin wrote: I just did a bounce test on a Stiga Expert Roller table of a Nittaku Premium polyball against a Joola Gold 3-Star celluloid ball, a Nittaku SHA and a XuShaoFa polyball (all 4 balls are brand-new straight out of the box).  Highest bouncing one is XuShaoFa, second highest is Joola Gold celluloid, third is Nittaku Premium polyball (just lower than the celluloid by a tiny fraction of an inch), and lowest is the Nittaku SHA.  Dropping from a height of 3 feet, the Nittaku SHA bounces lower than the XuShaoFa by more than an inch!

Bounce is measured by releasing the ball mechanically. After its bounce on a standard steel

plate the ball is monitored with a digital camera with a calibrated mm scale in the

background. The photos are evaluated. The geometric mean of three determinations then

permits calculation of the maximum height of bounce.

B.7 Bounce Conformity

All 24 balls must rebound to a height of not less than 240mm

and not more than 260mm when dropped from a height of 305mm on to a standard steel block.

B.8 Bounce Regularity (to be examined)

There is no specification for this property. Dynamic tests will be investigated

Debater's data showed the Joola celluloid balls averaging 229.8 cm bouncing off glass instead of celluloid.  Let's imagine that the celluloid balls would be in the middle of the acceptable range (240-260 cm) if he had used ITTF testing surface (say 250 cm).  If that is true then ITTF gets bounces that are 8% higher than Debater does with his apparatus.  My guess is that the Joola celluloid is actually at the low end of the acceptable range because I doubt that using glass instead of steel surface can account for that big a difference.  But let's imagine that they are in the middle.  Then if we make the same adjustment for data on Joola 40+ (multiply by 1.08) then we would expect that ITTF ought to be measuring ~239 mm for the Joola Plastics using their steel plate.  (Debater got a mean of 219.7 cm with his apparatus).  With those assumptions, then the Joola 40+ is probably below or maybe just slightly above the acceptable range for bounce height--but they may actually be quite a bit worse than the January 2016 standards.  From his and my data, they are also above acceptable range for weight (which I can confirm with very accurate laboratory analytical balances) and they do not meet standards for roundness and veer, from Debater's data.  The bounce height for the Joola 40+ is very consistent. I'll give them that.  But low.  It is is statistically significant compared to bounce height measured with cellluloid at a level of P = 10 to the -12th power.  In other words, the difference was systematic, not random.

But the thing I notice most is bounce.  Personally I think weight is much less of a problem.  Roundness is a problem, but one that they can solve.  Bounce, that is a big problem.