My MT12 arrived yesterday. Haven't had time to play with it much since I'm heading overseas today, but my initial impression of the menus is that it definitely looks like it was designed by a bunch of engineers with little UX consideration. Lots of options but not particularly intuitive. The ergonomics feel a bit odd, might take some getting used to. I've got fairly big hands but the trigger throw felt like a bit of a stretch. The calibration on the throttle seems a bit concerning, it'll only hit 98% when braking normally. I need to press forward and up simultaneously to hit 100%. The left dial also sits at around 1-2 points to the left when you hit the neutral point. Turning the radio on is also a bit weird, you have to hold the button while the screen does the animation. If you release it before the animation completes the power cuts out immediately. There's a setting in the menu to change it but even when I set the delay to 0 it still requires me to hold it for at least a second or 2.

If wheel, throttle or dials are a bit off, run calibration. That should sort it out.
Calibration is in the SYS menu, screen 6/7 by default.

I didn't like the feel of the MT12's steering, so I modified it.

1) I removed the burr in the hinge pin hole of the lever so that the lever could move smoothly.
https://cimg8.ibsrv.net/gimg/www.rct...fe03e3b2d3.jpg

2) I removed the looseness of the bearing (850ZZ) on the back side of the steering shaft.

3) I removed the initial tension of the return spring and increased the spring rate.
I prefer a stiffer spring, so I set the number of turns to 11 turns.
https://cimg5.ibsrv.net/gimg/www.rct...fb663c7138.jpg

The free length of the spring was set to a little over 15mm so that the pretension could be adjusted to 0.
https://cimg5.ibsrv.net/gimg/www.rct...271e7a667f.jpg

With the above modifications, the steering feel of MT12 has been greatly improved.

I would argue strongly against this - these hobbyist test methodologies are not standardised, neither by yourself nor other testers - the best thing to do would be compare hardware performance using your methodology, test equipment and environment as constants.

My radio should arrive today - do you feel this is the first thing I should do?

I'm confident I'm going to be able to set up endpoints (weight?) but not sure about how to do neg expo on throttle and steering.

When you receive your radio, the first thing you should do is calibration. It may not be necessary, but there's no harm in doing it.

Initial setup (neutral, endpoint, etc.) is usually done on the "OUTPUTS" screen (page 6/11). EXPO will be held on the “CURVES” screen (page 7/11).

Yes, I think you are absolutely right. It is natural that the test results will be different if the environment is different. If I have enough time, I will do a latency test of MT12 (elrs).

I said this.
"I will only get the same results as other people's tests"
Yes, this is just my "guess" and I have no proof. It is natural to have doubts about the results of my "SANWA test".

However, I think you can trust my "SANWA test" to a certain extent. This is because I have conducted other latency tests in the past and have confirmed that I have obtained almost the same results as other testers. I posted the REDPINE test results because I wanted to convey that.

My test results may not be sufficient for quantitative consideration, but I think they can be used for qualitative comparison. Please take it as such.

OK I am going to grab this

https://www.amainhobbies.com/radioma....5a2s/p1533868

as a change of radio this year, I chose this over the 4n1 for the super fast response. Are there different types of receivers? I need one for my 5th scale, 8th scale, 1/14 and 1/12th scale and trying to get the proper units. any help is appreciated

Basically, I think ER3C-i is sufficient. If you need 4 or more channels, ER5C-i is available.

thank you

Some accessories have just come through on email. Most look like spares, trigger assy, wrist strap etc, but there's what looks like a right to left hand enclosure option, which I believe is new.

https://www.radiomasterrc.com/produc...48375063019751

Something seems to be off either with your methodology of testing or the one condac used in This Thread (Radio Benchmark program and results).
It shows hugely different results than your findings. It shows an average latency of 4.1 ms with a standard deviation of 0.730 ms. And this results are from the Sanwa M12 NOT M12S or even the Sanwa M17.

You don't need to configure expo in the curves screen. You can configure a simple percent expo right in the inputs screen. Curves is more for if you need to configure a many point curve. You can also configure expo in the mixes screen.

General rule of thumb for EdgeTX is to use the inputs screen for your rates and expo. That's a carry over from planes where you might have multiple outputs tied to the same input by different mixes. The most common one is you have a servo per aileron. So you need two aileron mixes. To avoid having to apply your dual rates or expo config to both mixes you can instead apply it to the single Aileron input, which saves some work and room for error like apply an expo change to just one aileron. With a car where you just have throttle and steering it doesn't really matter of course.

There's a lot of different ways to do the same thing in EdgeTX, not really a wrong way.

That is correct. I overlooked it.

Certainly, the results are completely different. I'm also interested, so I'd like to check out condac's arduino code.