Fantastic setup! Kinda (read:absolutely) regretting my commercial rig purchase.
My DIY is limited to the panel. I only fly a Cessna so I’ve built mine specifically for it.
I’m building another panel to match a Bonanza though. A couple of Arduinos, a handful of switches & a piece of plywood and I get a second cockpit for about $100. It should take about 10 minutes to swap them out.
Not sure whether this is useful - I recently read on discord that the DOF motion rigs are compatible with FlyPT Mover. Not sure how much work is involved as I haven’t bought mine yet, but theoretically if it is compatible and could be setup correctly then a DOF rig may be capable of supporting the taxi movements and random turbulence similar to Romans setup.
Cheers,
Thank you for the heads-up! I’ll look into it.
I’m running it on the same computer, forcing it and other software with Process Lasso to only use CCD1 (logical cores 8-15) while MSFS is forced to cores 0-7 (the ones with 3D cache on 7950X3D). it doesn’t interfere with MSFS at all.
Very nice! My setup is universal though, a different approach, as in VR I can’t see the setup, so it should be somewhere easy to reach by feel alone. If I’d fly something reguralrly IRL a better approach would be the exact model, so the muscle memory is trained, but just for swimming, this works great, as I can control all important systems in any plane without thinking, and they are located in roughly the right directions, enough to not break immersion too much - and way better than any mouse/VR controllers, both functionally and immersion-wise.
Also don’t forget the vibration feedback - it adds a lot of immersion when coupled with movement. In taxiing you feel the seams on the runway going by faster and faster when speed builds up, and gear/flaps/turbulence and subtle engine vibrations work amazingly well with motion, creatine a wonderfully real illusion of flight. When something goes wrong and SimShaker doesn’t connect, of I forget to turn on the amp, it feels very thin of a sudden, very noticeable. But in terms of motion, subtle random noise on taxi and take-off roll works great! I also added a motion “bump” (via pitch axis) on the first instance of air-to-ground transfer, which mimicks the touchdown perfectly, I can really feel it quite a lot now (just the vibration touchdown was too weak, now it only felt on nose gear touchdown which is OK). It also affects the take-off. So together with noise, I get a noisy and bumpy take-off roll, then a moment of take-off and then smoothness of the flight is a stark contrast with the take-off roll, really accentuated the soaring. Then I feel the buzz of gears going up, then flaps. Feels very real.
This is true! I have a Buttkicker and the sim feels lifeless without it! I would almost give up motion, before giving up vibration.
That sounds like a solution I might try. I run my Mover on an oldish but fairly decent laptop and it consumes about 50% of the CPU so it’s not a negligible load. I think I’ve worked out a way to install Simconnect on a second PC and network it. I’ll report back if it works.
This thread, and your other one on this forum which is unfortunately dead and locked now, are the only places so far that I’ve seen the subject of how to implement flightsim motion on a motion rig. Very interesting and helpful, so thanks for your insight. I’m some way behind you but catching up (hopefully).
Glad to share! I’m so happy I took the plunge and built it, it’s a completely new level in enjoying a flightsim.
I also used it with Elite Dangerous until I got bored with the game. In space I was not using gravity obviously, but the momentum of rotations and sideway/up/down translation thrusters. Amazng fun!
And for kids I have “Rpic Roller Coasters” which I enjoy too. It can make me sick in about 6 rides. Girls only make it 2 before they have to stop.
OK, so I tried running mover on the main game PC and Task Manager reports it only taking 2-3% of CPU so perhaps my old laptop is rather older than I thought. Now that Mover is on the main PC, the Simconnect source installed correctly and I’m up and running in MSFS as well as DCS.
I had hoped that my DCS settings might just transfer over but that’s a no go. The roll axis is backwards for a start and the sim variable outputs are different, so it’s a setup per sim, and probably per plane too once we get into fine tuning.
Now that I know what I’m doing (a little) with Mover and what I’m looking to tune, it only took about half an hour to get setup in MSFS and it’s working great.
After about half an hour of flying around and being totally wowed by how beautiful MSFS looks, my enjoyment was cut short, as it so often is, by technical problems. This time it’s the huge “internet connection lost” window which keeps pointlessly and infuriatingly popping up, usually just as I’m on final to land, and locks out the controls causing a crash. A quick look on this forum reveals people have been complaining about this for three years… Maybe my connection is a bit dodgy but I can multiplayer dogfight on other sims without a glitch. So once again I did what I normally do when MSFS goes into diva mode, which is to give up and go back to flying in DCS. Maybe if I sprinkle perfume scented rose petals into the MSFS folder and whisper sweet nothings in its ear it might decide to come out and be beautiful again in a few days time.
In the meantime I’ve got some new position sensors for the motion rig to try and improve the resolution. At the moment I can feel the rig stepping when moving slowly so hopefully these new sensors will fix that.
My setup works great for all planes in MSFS. What helps is that I added an optional Potentiometer to the Arduino that scales down everything. I placed it within reach, so basically, I can dial the motion up or down if I feel that it’s too strong for a particular plane. The only problem I’ve seen is helicopters - I get the shakes on the ground. Something is fishy with the data MSFS sends, or my noise function for the touchdown or ground bumps is acting strange, but only in helis.
I used extremely cheap and frictionless hall sensors and I get plenty of precision, no stepping feel ever, it’s completely smooth. Though with my setup they can only sense 180 degrees of rotation, then they just go back to 0. I only need about 90 degrees of the arm movement though, plus some safety margins, but because a runaway would be catastrophic (trying to do a full rotation through the frame etc.) I added hardware stop switches that cut power in the event of a runaway, on top of configuring soft stop switches in SMC3 Tools that should cut power well before the hall sensor signal reaches it’s plateau. It’s all on my website including the 3D models for 3D-printing the housings for the hall sensors etc.
Make sure you test any new sensors with arms detached, or this could break the rig (or you).
Edit: I just realized there is a “Home Cockpit Builders” category, so I’m moving the post here. I think it’s the best category for it, as the concept is more than just a motion simulator, it’s the entire VR cockpit with DIY controls.
Since switching PTMover onto my game PC I’ve been getting stutters in VR. I thought this was just MSFS being a diva but it’s there in DCS too and it disappeared when I switched Mover back to a second PC. So there seems to be some sort of resource conflict going on. I’ll have to investigate this Process Lassoo thingy. Another dollop of learning for me on this project.
Also I fitted the new position sensors but they made no difference to the steppy motion, which is sort-of gratifying as it means my home made jobs were working just as well as these rather expensive new ones, but it still leaves me with the problem of the steppy motion. After more diagnostics I’m now pretty sure it’s the motors that are the problem. The stepping I’m feeling is the cogging effect of the armature moving too slowly past the magnets. Presumably I need a higher revving motor with a taller gear ratio to fix this but I’m not sure I can stand the pain of finding and fitting new motors at this stage. Perhaps if I fit some sort of Buttkicker-like sound/vibration system it might mask the stepping or maybe some sort of springy connection the motors will help.
So I’m a bit hacked off with it all at the moment but will have to find some more enthusiasm and push on to victory. Once you go down the road of motion simulation and get a glimpse what can be, it’s very hard to go back to a static chair. It all feels very dead and unsatisfying without the movement. A bit like sampling VR and then going back to a screen.
i’ve noticed i have to hide every ptmover panel while msfs vr is running, otherwise it gets very “steppy” like you say. with just the gray module panel it’s fine. i run it at 2ms cycle with high precision turned on.
And the prize for Hot Tip of the Day goes to… @smitty9792
I tried again and hid all the Mover windows and the VR stuttering went away. Thanks for that. 
There is your problem! Don’t use high precision, there’s absolutely no need. Developer himself told that it increases CPU usage immensely, and without it all motion is extremely smooth and precise. AFAIK, Arduino and its interface run at much lower frequency anyway and doesn’t need this precision at all. There’s just no benefit, and it’s wasting CPU cycles. Without it you can have whatever windows open that you like, too.
Also, while VR is never perfect and I do have some occasional stutters unfortunately, they happen with or without FlyPT and they are not regular. What kind of stutters are you having?
Generally, there is no need for Process Lasso even, if you launch it via BAT file or from command line. Here’s how i do it - I use a custom BAT files to load all the software I use in proper order and timing, and I even have Y/N menus so I can select what to load for a particular flight, i.e. if I want ATC or charts etc.
start "" /affinity 0x00000000FFFF0000 "d:\MSFS Projects\Motion Simulator\FlyPT Mover\Roman Rig MSFS.Mover"
This forces FlyPT to stick to CCD1 cores only - logical cores 16 to 31. Google “CPU affinity calculator” to figure our the correct code for your CPU and desired core combination.
I wasn’t even aware there is a high precision mode. Found it now in the options list. Seems to default to off and 2ms update rate. The output timing for SMC3 is 10ms so I guess anything faster is a waste of resources.
Still, as long as the Mover windows are hidden I’m not seeing any ill effects now. I will Google up on CPU affinity though. Thanks.
Totally righteous!
I wish I had the SPACE for something like that. Not to mention the money! 
Well, it takes almost the same space as a large office chair with its legroom. So not much space at all. And it cost me under $500. So not much money either 
One thing I wanted to mention that it’s not limited to MSFS. FlyPT can get data from many games. Elite Dangerous is amazing - no gravity in spaceflight, so it’s all motion cues from acceleration and it feels great when thrusters fire, including lateral etc.
I’m looking forward to Starfield. I will try to get it to VR (with Depth3D or VorpX) and I will use joystick movements feed to guess accelerations. I did this with Elite Dangerous before I found a working data feed from the game. It works quite well. I’d get acceleration cues from joystick pitch and roll, and from the delta of HOTAS thrust adjustments - smooth acceleration fed to pitch, based on how quickly and how far I move the throttle. It was quite convincing. Let’s see how well I can make it work with Starfield. It all depends on how well it can work with those VR mods.
I have managed to reconfigure my VR motion rig and control for playing Starfield. Considering that the game doesn’t support joysticks at all, not has a data output for the motion, nor does it support VR - it was quite a task. If anyone wants to see a video on that - let me know…
Count me in! This’ll be in IMAX, right? 
Seriously, go ahead and post the link.