I am reporting what the frequencies of the trigger and ramp generated by the 33622A are and what the time/division of the 11801 is set to. The 33622A frequencies are accurate to more digits than I have shown. ~1.1 ns is the difference in the length of the trigger clock period and the ramp period. The 33622A is *much* more stable than the 11801 clock.
At 512 & 1024 samples the 11801 will allow 1 ps/div. As you increase the number of samples, the minimum time per division increases. The limit is 10 femtoseconds interval between adjacent sample points. So as you increase the number of points in a sweep, the minimum time per division increases.
With a 100 kHz ramp displayed with 1024 points spaced at intervals of 10 fs, the display is only showing 1 out of every million possible samples along the ramp.
With 1024 samples at 10 fs intervals, it makes 4 obvious sweeps with 200 fs jitter between the start of each sweep. It generally shifts slightly every 4 sweeps and 4 sweeps take place faster than I can count. When the OXO heater turns on it moves like mad for about 10-20 seconds before it settles down again. At 5120 samples at 10 fs intervals it makes 2 obvious sweeps. The jitter is tough to estimate at 5 ps/div. but looks to be around 200 fs. There is some indication that there might be 32 sweeps in a set, but I have not been able to find a way to verify that.