05:31 <azonenberg> So every time i port a feature over to ngscopeclient, I look for obvious way to improve

05:31 <azonenberg> one thing i want to fix is FFT peak labeling

05:33 <azonenberg> in glscopeclient, every frame we calculate position of the labels anew, statelessly

05:33 <azonenberg> s/frame/refresh/

05:33 <azonenberg> anyway, point is this model does not allow manual dragging to move the labels, and it also can lead to disorienting jittering of the labels from trigger to trigger

05:36 <azonenberg> So i want to make a new system that is stateful

05:36 <azonenberg> where a) you can drag the labels and they nicely collision-detect when you bump them into each other

05:36 <azonenberg> and b) if peaks move slightly, the labels stay matched to them and don't move much if at all

06:28 <azonenberg> Not going to finish tonight but should make good progress tomorrow

20:33 <azonenberg> Also i think i'm going to write a "generic i2c device" decoder

20:33 <azonenberg> basically a generic-ified version of the i2c eeprom decoder

20:33 <azonenberg> that takes in raw i2c read/write transactions plus a parameter for device address and number of register address bytes

20:34 <azonenberg> and spits out a bunch of read/write transactions to device registers

23:31 <d1b2> <david.rysk> Yeah that’s one of the nice parts about the Saleae software — a bunch of generic decoders like that