Does this tonal difference occur in all H9 delay algorithms? If not, in which ones does that difference occur?
The smoothing out effect of using Time mode with a large amount of Taps has to do more with the fact that the spacing between taps no longer follow a set subdivision as it would in Tempo mode. In Tempo mode the knob position determines a set subdivision, but in Time mode its determined by time in milliseconds. Basically, there will be a higher concentration of taps in Time mode making a reverb, for example, sound less clumpy.
In the example at 53:45 that you’re referring to, I’m creating a chorus-type sound. Notice I have my length set at a very short 16th note with five Taps. Switching to Time mode, I’m essentially shortening the delay times between taps making the chorus sound smoother. Works well with Slurm engaged, which serves to take the edge off the transients.
This effect will happen with any multi-tap delay, especially if there are a lot of taps in a short amount of time. But notice I really don’t mention “timbre” anywhere. Though it may be an audible effect you perceive, the sound changes because switching from Tempo to Time mode we’re shortening the delay times further. That’s it. There isn’t any secret processing going on when we switch the Tap mode from Time to Tempo or anything like that.