We present a beam-split coincidence test of the photon model, previously done with visible light, now for the first time with gamma-rays. A similar new test is presented using alpha-rays. In both tests, coincidence rates greatly exceed chance, revealing the flaw of quantum mechanics. A newly formulated threshold model predicted these new tests and was used to derive equations for effects thought to require quantization. Quantization denies sub-quantum states that are allowed by the threshold model. The threshold model embraces Planck's second theory of 1911, where he used h as a maximum threshold. We extended Planck's theory by similarly treating e and m so that all three constants of the electron (h, e, m, for action, charge, mass respectively) are realized as maximum threshold-constants. We recognize ensemble effects, absent in spreading waves, for how we uncovered those constants. We then use ratios of those constants, like e/m = Qe/m , for the spreading wave. By quantizing the Q's and thresholding h, e, and m, a matter-wave can spread and load to a threshold upon absorption. Therefore, wave-function collapse is avoided. We also identify several false assumptions that made any form of classical model seem impossible. The difficulty of realizing an experimental distinction between thresholds and quanta is why quantum mechanics maintained such a strong illusion. ER 7, 21, 2020.