I found myself alway using the airspeed instrument for takeoff and landing, but not in flight.
(as in CDI, GS, TAS, OAT, etc.) without the need for a second display, or dependence on the Garmin CAN buss and/or G3X feed.
Agreed about round gauges for takeoff and landing. Here's my take on steam vs glass:
* If the parameters are changing slowly, and you know in advance what they should be (like when you're flying an approach, heads down), it's a wash;
* If you have no idea what the parameters are and you have to read them instantaneously because your attention is required outside the cockpit, steam wins big time;
* If the parameters are changing rapidly, steam wins.
I chose the G3X over the Dynon because at the time, G3X had decent steam gauges to display and the Dynon had an extremely clunky display. (Also, I'd talked to the chief engineer at Dynon at one of the shows and developed an extreme dislike for him. He's long gone, now. And there were other reasons.)
More thoughts:
* On tapes, having the current value overlay the tape is stupid. What this means is that you can't use the tape per se for fine adjustment because the value you want is under the digital readout, and you have to read the digits for fine resolution. Maybe this was a requirement when screens were narrow and there wasn't room, but these days, there's room to move the digital readout off the tape;
* Altimeter tapes are usually +/- 225 feet, but when a "thousand feet to go" chime goes off, that means "sometime in the future, but not now, start looking for the altitude bug." It would be so much smarter to have full scale +/- 100 feet, and reduced scale outside of that so that you can see the altitude bug appear when the chime goes off;
* RockwellCollins had a truly brain dead display in which the vertical speed was a short white needle on light blue tick marks. This was extraordinarily hard to read because there was so little contrast, but the human factors books say don't use blue anyway. (There was so much more to distrust about RC when I was there...);
Second point: the G3X now has a direct OAT input, so it can do all the air data stuff. And the instrument can be toggled between HSI and attitude indicator. Also, if it is hooked up to the CANbus, it serves as a backup ADHRS in case the ADHRS goes poop. In my system, if the ADHRS goes poop, I lose OAT (hence, TAS and wind, no big loss as the system reverts to groundspeed, which is what you really care about), AOA (no big loss there! only the aural stall warning is of interest) and magnetic heading (no loss there as I still have ground track from the GPS. And 99% of the reason for having heading is so that you can iteratively get the ground track you want, but GPS gives that to you directly.)
I've never seen a refereed article comparing readability of tapes to glass... but lots of opinions!
Last point. You know when the first vertical tapes flew? On the Boeing 757/767? C141? Nope, a little bit earlier, like 1925. Sperry made them, but I've not found any pictures on the web.