Who doth awaken me from my slumber?! 
I personally use the Tempi a lot via the “human programming” input - manually tapping the rate I want each output to be. This feels much more natural and musical to me, and often faster as I’m not always sure what the corresponding div/multiplication is to what I hear in my head. ALSO, it’s much easier to set the Phase this way. If you, for example, just tap two quarter notes on the upbeat, you just inverted the phase from another quarter note output. No key combo necessary. Granted, this will take some practice, start with 100% human quantization and try and quickly move towards 50%. 50% isn’t that hard unless you have virtually zero rhythm. 25% gets tricky, but, there’s a lot more happy accidents in there too. Especially if you’re…
Using the MOD input. Ironically, while the mode that bears my name was designed to make it easier to sync to a DAW (or some other MIDI/Clock source), I’ve only ever used it that way when testing the idea and beta testing firmware updates. I do often use it, however, as a way to manually reset and run the clocks. For example, using Pressure Points or some other manual gate into the MOD input. I find having most of the Tempi outputs programmed to multiples more fun, and usually I find it more interesting if Tempi is not chasing some external clock - I’ve set the tempo and I’m just playing along. If having to constantly play the MOD input manually is not your thing, add a Gate sequencer in place of the manual gate. Note that this mode is not that interesting if you self-patch a Tempi output to the mod in.
The Rob AA Lowe mode is also very fun as you are essentially switching outputs on/off with an external gate/clock. This can create a lot of interesting interactions that, on the surface sound like clock divisions, but depending on how you clock the mod input, can become rhythms and not just steady divisions. This is mode IS very interesting if you self patch a Tempi output to the MOD input.
I find I don’t use STATES that often. When I do use them, it’s usually in a live performance context where I know I need to have specific states I need to return to reliably. One where all the channels are muted for example (quick stop), or maybe 2 or more channels are clocked at a steady 4 on the floor for a driving beat. With Rene2 however, having the Rene states locked with the Tempi states is incredibly cool and useful. And MESH / MESH Paste on Tempi is amazing (discussion on MESH probably too out long/out of scope for this thread).
In general though, I would suggest that the best way to explore Tempi is similar to most modules. The two most import key points, imo:
- Sit down without a goal in mind. Just sit down to explore.
- Start immediately with a setting/function/feature that you have never used before.
#1 is always the most important. If you’re sitting down to write a hot new track and you want to experiment with some new functionality - that’s a recipe for frustration in my opinion. Or, you end up doing the opposite of #2, you eventually just turn back to what you already know or are comfortable with.
I guess #3 might be “don’t worry if you don’t understand what’s going on”. Sometimes you just need to remember the steps you took to get something interesting so that you can follow those steps again. Over time and repetition, you’ll come to fully understand what’s happening. For example, there’s lots of fun with, say, Tempi clocking Rene2, and using the C gate output to drive the MOD input. But it may not be obvious which controls will have what outcome, you just have to experiment and see what is the most interesting to you.
