You might be seeing the results of your digitizing scope sampling near to the peak and sometimes but not always at the peak of the time marks. Try running a faster sweep speed (smaller time/div) on the digitizing scope so that you view only one time mark or at least the rising edge and the peak. The objective is to capture the signal with a lot of samples at and surrounding the peak. If your digitizing scope has and envelope or something resembling variable persistence, use it to display the variations in successive waveforms. This should allow you to view the excursion of any modulation or noise without the influence of undersampling the peak. If the signal has only a little noise, the low persistence of the analog scope and your eye will reduce the small influence of the noise.
