Got birds? Click Here!

CSM vs. RFD

There are a number of RC helicopter simulators on the market today, each of which have their own strengths and weaknesses. I have personal experience with only two of them so far: the CSM 3-in-1 simulator (hereafter, CSM) and RealFlight Deluxe (hereafter, RFD).

Simulator	Pros	Cons
CSM	Simulation The CSM simulator emulates the flying characteristics of a radio controlled helicopter very, very well. It taught me everything from my first hover to forward flight to pirouetting circuits and rolling circles. The flight characteristics can be adjusted to match the characteristics of your own helicopters, an exercise that will also teach you a thing or two about helicopter tuning.	Graphics The CSM's graphics were merely adequare in version 8.3. Better than the Dave Brown sim, but only incrementally. Version 10 makes a substantial improvement, but still won't take advantage of the spiffy features of your Voudou Banshee Viper 6000 SUX Nitro Ultra video card. Of course, it's rare to hear of people having video driver problems with CSM!
RFD	Graphics RFD's graphics are pretty much state of the art for personal computers. Because of the way RFD uses DirectX 3D software, its graphics and may well improve every time you upgrade your video hardware and video drivers, even in the absense of new simulator releases from Knife Edge (makers of RFD).	Simulation RFD's simulation is reasonably believable in a hover, but quickly degenerates as you move into forward flight. Reputable sources tell me that the airplane simulation is excellent, but unfortunately the same cannot be said of the helicopter simulation. More on this later.

All things considered, I'm sticking with CSM, at least until RFD's simulation catches up. The prettiest graphics in the world are worth nothing to me if they don't behave like the real thing. I'm more than a little bit disappointed in the quality of the simulation. It doesn't live up to its own press. People commenting on RFD on the h-list, IRC, or newsgroups mostly fall into one of two categories: experienced pilots who can't get RFD to fly like it should, people who claim it flies perfectly but who turn out to be relatively inexperienced pilots ("I'm just getting into forward flight, and...").

Specifically, RFD falls short in these areas:

1. The cyclic response varies widely with speed. With my RFD models, and also with two custom models that two people have sent me, if you fly way out to the edge of the field, do an eye-level flyby at full throttle, pull back and try to do a circular loop, it's obviously bad. The cyclic response changes so the first part of the loop almost becomes square, before response tapers off and the second half of the loop comes around very slowly - you have to pull back HARD to get upright again.

2. The cyclic trim also changes with speed, and not in the way that a real model helicopter's trim changes with speed. When I try to do a flyby in RFD at full throttle, I have to hold the cyclic stick back back quite a bit to maintain level flight. It hovers just fine though. Something changes the fore-aft trim A LOT when the airspeed builds up.

   I first suspected that chassis drag might be at fault here, but tests with upright and inverted flight show that the problem is the same in either orientation - the model wants to dive as speed increases.

   One fellow sent me a ".heli" file he'd made that doesn't have a fore-aft trim problem at speed, but it rolls left instead. I think that's doubly odd for clockwise rotor rotation, where differential lift might explain a tendency to roll right (not left) in fast forward flight.

   I've had minor pitch-UP problems with both of my real helis, as a result of my experimenting with flybar & paddle setup - in fact my Concept SRX still does this a little bit - but to this day I've only heard from one person who has ever had a real model that pitches down in forward flight.

3. The auto performance is very "generous" with everything I've tried so far. Turning the blade weight down to 100 grams left way too much hang time at the end, and even a mere 50 grams still produced longer autos than I get out of my (real) Futura SE with its 185 gram blades. In the real world, autorotations become challenging even with 75 gram blades on a 30-class helicopter.

4. The tail rotor response is very "soft" and cannot be made crisp enough. The yaw rate accelerates too slowly. Note that the problem is not that you cannot get fast enough pirouettes - you can - but the pirouette rate builds rather slowly. With a well-tuned helicopter, pushing the rudder stick to full deflection gets the helicopter spinning like a top in a heartbeat. With RFD, the yaw rate builds more slowly. Experimenting with the CSM gyro's 'yaw acceleration' parameter suggests that my Futura SE accelerates at a rate of roughly 4000 degrees per second per second. Turning the yaw rate to the maximum allowed by RFD (9000 d/s/s) still yields poor acceleration. The effect is most noticable when practicing knife-edge 360s and 540 stall turns. 9000 degrees per second per second should be extremely twitchy, but it's not - in fact it's considerably less responsive than my Futura when the gyro is programmed for 4000 degrees per second per second.