MWO Simulator work log
#141
Posted 05 February 2013 - 12:19 PM
#142
Posted 05 February 2013 - 12:31 PM
#143
Posted 05 February 2013 - 10:34 PM
#144
Posted 12 February 2013 - 07:10 AM
Why didn't someone tell me that the game looked this good on maxed settings....wow.
I got around to connecting all three of my displays up to the new rig. Nvidia's surround setup is, well painless. Hook up the monitors, check the box for surround and make sure the screen identifiers match the positions you have them in. Then you do the bezel adjustment. It tells you what rez it is pumping out and your done.
Playing on three displays:
It is pretty cool. Even with the bezels that get in the way of the chat it really draws you into the game. I have yet to try changing the FOV, but I may do that soon. I find myself actually looking left and right to see out of the "windows". Awesome.
Screen alterations:
I am seriously considering tearing the screens apart to get the LCDs closer to one another. What I am not sure about is if I can remove the plastic bezels and maintain the vesa mounts or not.
The mech pit:
Work continues on the pit, even if slowly. I have some pretty lofty ideas for the screens that are in the "dashboard" of the catapult that I will get into if I can get any traction with them. Otherwise one will act as a web browser and the other maybe a rear view camera. Got to know when I'm getting snuck up on by the wife or kids.
#145
Posted 14 February 2013 - 01:58 PM
This is what I started with.
I had the displays on the side set slightly ahead of the center and slid in to try to minimize the bezel width. Somewhere in the 1.5 inch range of visible bezel.
Front bezel off with the front controls removed
And the neked screen, front and back
Then I stripped the other two screens, covered the silver around the panels with good ol electrical tape and mounted them. This time I decided to set the center screen forward to minimize the visible bezel. Now I'm at something like .5" visible.
I am considering lining them up edge to edge. I'll play like this for a bit and see how it feels.
I do have a gripe. Using surround I end up with one of the bezels through the center of the chat window. The starting screen is split by the bezel as well but that is not nearly as annoying.
Much happier with the idea of this arrangement in the mechpit than the single. There would have been no way I could have drove the 3 displays from my older rig however.
Next task is the seat, strakes and ejection system.
#146
Posted 14 February 2013 - 03:50 PM
I wouldn't THINK Nvidia would handle the display any different. . .
Edited by Sen, 14 February 2013 - 03:50 PM.
#147
Posted 15 February 2013 - 11:09 AM
So I did some reading on Nvidia surround and from what I gather it gets pissy if the monitors do not match resolution, refresh and sync polarity. (Don't hold me to that last one) I'm at work so I can't check things out but one of my monitors is not the same model so it is possible that something there is off, even if ever so slightly.
If I can't get surround to play nice, I will try SoftTH. It is just not as friendly to use in terms of bezel correction.
I will update after I have had time to play with things. Thanks for the thought Sen.
#148
Posted 15 February 2013 - 03:39 PM
If Surround really is having fits due to one monitor being a bit different (seems odd... why would it care if they are all the same resolution?)... you could try just taking the bezel correction out before messing around with SofTH.... after all, your have reduced your bezels to fairly minimal.
Another possibility, if you just can't stand to not have the bezel correction... you could physically mount your monitors with more physical overlap, to make the images line up the way they would by using bezel correction. (Basically, just slide your side monitors in a little more behind the outside edges of your center monitor). [Side benefit, to help sell yourself on it... your video card will have fewer pixels to render.]
I know you are building a nice spiffy box to play in, but sometimes you just gotta think outside the box!
#149
Posted 18 February 2013 - 07:24 AM
Surround was not used.
My center monitor shows that it runs at the same resolution and refresh as the other two, the fact is that it is really running about 5 hz faster than the other two. This completely blows surrounds mind. Cant handle it. An additional factor is that the center monitor is running DVI-D and the other two are DVI-A, this makes surround bring down the resolution of the center monitor to what the outside monitors can handle at 75hz.
I gave up on surround and went to SoftTH. The first several attempts (until the face palm solution) resulted in me playing through those matches in a resolution that looked like the full 5020x1080 was stuffed into the center monitor. The issue was that I had been running MWO in full windowed mode. Once I switched it to full screen I got the SoftTH logo on the side screens and a full display.
The rub there is that SoftTH takes my frame rates down to the mid 20s low 30s on ultra high settings. Once I dropped down to high I am running in the mid 30s
#150
Posted 19 February 2013 - 08:06 AM
Good build so far. Can't believe I missed this originally. If you ever want inspiration go look at the Vulture pod. That dude's wood skills are insane.
My pod will be complete soon, in the memory of the Virtual World VR2.5/3.0 pods. Should be fun.
#151
Posted 03 March 2013 - 12:23 PM
Foust, on 01 August 2012 - 05:44 PM, said:
This number bothered me for two reasons. First, because I do not believe that I have all of the available commands that will be in the game I am fairly confident I am missing some. That means I cant layout all my panels, or order all the hardware I'll need etc etc.
Second, windows only detects 8 axis, 32 buttons + 4 way hat on the joystick. 32 + 4 =/= 39.
Crap. So now what? Another encoder? A factory Hotas? Hack a keyboard?
I decided not to panic and wait until I could work with the GenericHID again and see what I could work out.
I stripped most of the buttons out and replaced them with a 8x8 matrix. Filled the rest of the spots with buttons. Programmed the Teensy++.
In theory this should produce 8 axis, 75 buttons. Windows shows 8 axis, only 32 buttons.
By the grace of God, the internet and google I found the solution in a program called SVMapper. This particular application was written to handle a device called MJoy16, another home brew HID interface. The trick to this is though that it will recognize over 100 keys. The caveat is that only the first 32 can be detected as simultaneous presses. I can work with that. SVMapper also will handle the toggle to momentary problem that I was addressing with xpadder, which means I do not have to add a third supporting application.
So for those following along at home:
Teensy++ Loaded with GenericHID
SVMapper to assign buttons 33-75 to keystrokes and handle the toggle to momentary issue.
AutoHotKey to play sounds on key press.
MODS: If the teensy++ link is a no-no, strip it.
So i have my Teensy 3.0 and got it up and running with the default shipped program. You mention loading it with GenericHID. So lay out the buttons like you want them.
So following the steps
Open GenericHID
Build your keys
Microcontroller > Program
At which point i don't get G-HID to see my T3.
Or did you export it using G-HID's default ***.mcd and then use Teensey loader 1.06? T3 uses the 1.07
I'm trying small steps. Like renaming the device. then having only 1 button active. That one button triggers the autokey. I have the autokey part done.
Any help from any Teensy owner works too
Any pit updates from you Foust? Last i heard in a different threat you tore yours down for upgrades?
#152
Posted 04 March 2013 - 06:07 AM
I would have to look up the teensy 3.0...
As to the pit, I did tear it down. It is undergoing iteration 2. Ive built what I call the "cheap aces" seat by taking the AB seat and part of the base and building a 2x4 frames for it to sit on. Add some casters and seat is functional.
I will try to get some pictures of its current state up tonight if I remember.
I did make the decision to remove the foot rests from the rudder petals and make them into separate push buttons. This allows me the ability to move the petals farther apart (important to putting a center pillar in for the dashboard) and it will allow me to change the angle that they are mounted so that the seat can move closer to the dash without bringing your knees to your chest.
Project slowed by sickness in the family, playing MWO and FTL.
Edited by Foust, 04 March 2013 - 06:18 AM.
#153
Posted 04 March 2013 - 06:21 AM
Foust, on 14 February 2013 - 01:58 PM, said:
This is what I started with.
I had the displays on the side set slightly ahead of the center and slid in to try to minimize the bezel width. Somewhere in the 1.5 inch range of visible bezel.
Front bezel off with the front controls removed
And the neked screen, front and back
Then I stripped the other two screens, covered the silver around the panels with good ol electrical tape and mounted them. This time I decided to set the center screen forward to minimize the visible bezel. Now I'm at something like .5" visible.
I am considering lining them up edge to edge. I'll play like this for a bit and see how it feels.
I do have a gripe. Using surround I end up with one of the bezels through the center of the chat window. The starting screen is split by the bezel as well but that is not nearly as annoying.
Much happier with the idea of this arrangement in the mechpit than the single. There would have been no way I could have drove the 3 displays from my older rig however.
Next task is the seat, strakes and ejection system.
I did three monitors and it looked like trash, even with adjusting the FOV. The menu system being super skewed is a huge hindrance and completely breaks the immersion. I chose then to go back to a 30" screen with higher resolution and a second screen where I can pipe out the MAP at a later date (if they add the feature). Once you run span it's tough as nails to get your forth monitor to break it.
#154
Posted 04 March 2013 - 06:24 AM
#155
Posted 04 March 2013 - 10:54 AM
Quote
Good call. Not sure how much you have played with pedals yet, but so long as you are using central pivot airplanecentric pedals, you are definitely in 'hard mode' due to the on/off nature of MWO's current turning scheme. Once you separate the pedals and can get them to independently generate 'a' and 'd', steering mechs with your feet is a loooooooooot easier. After playing many rounds with normal pedals, it was clear that airplane pedals were less than ideal.
Reason being, with the *on/off nature of turning, one must often do a rapid series of small inputs. With normal pedals, this req you to move both feet and shift your entire body in the seat, and leads to either undercontrolling the mech due to lazyness/discomfort, or serious back/body pain and driving is a lot clumsier than it need be. Painful as it was to abandon the pedals I was making for MWO (they will still be used on another pit of course), it was definitely the way to go. Mine were destined to be 16" on center, which would have left plenty room to get my legs around the column, but there was no getting around the sub-optimal arrangement of using airplane pedal arrangement to drive a tank. Here's what they look like on their test platform.
*MWO turning is on/off, but with TARGET I have programmed a scheme that allows a small degree of variation to turning speed, by varying the length and frequency of the 'a' and 'd' keystrokes being generated by my pedals. I began to outline it in this post, with more to follow, as well as a post about my new (2nd for those following) mechpit!
Edited by Loc Nar, 04 March 2013 - 10:56 AM.
#156
Posted 04 March 2013 - 11:26 AM
For me the teensy build was part to resurrect the suncom stick (that I have since abandoned) and part for the additional input options. Then a very generous individual gifted me the TM hardware I have wired up to the teensy now. Other things...
It really is tempting to pick up one of those T16000's, but I am a tightwad and already have WAY to many joysticks in the house not to mention the hours invested in the rebuild of the F-15 stick. Where you peaked my interest was a comment that you made about rotational dampers.... Tell me more...
#158
Posted 04 March 2013 - 01:14 PM
Quote
A while ago I was looking through some backposts on simhq, paying particular attention to any custom/modifies/hacked sticks and throttles I could find. There is a user there, V010Meister, that makes what are serious contenders as the premier homegown controls that he CAD designs and cnc cuts, using nothing but the finest parts. Reading this thread of his (fabricated gimble for WW2 fighter), I learned about these great little rotational dampers, that seemed to be exactly what I what I didn't know was missing to bring my own stick up to the next level.
My stick is not lacking any function, but the gimble I made relies on greased plastic rubs with adjustable spring (since changed to rubber cushions) loaded tension on the axes, allowing the stick to be left in whatever position it is in when my hand lets go of it as well as providing some drag and damping force. Overall this system works well, and the stick moves oh-so-smooth, but has it's limitation. While the stick is incredibly precise and easy to use in the rough sense, it's more challenging than need be to make very small fine tuned *aim adjustments, say like when you are sniping at 1000m with advanced zoom, owing to a minute ratcheting tendency. It was mild enough to ignore before I learned about these dampers, but knowing the possibility of using these got me searching the ebay, and luckily I found some pretty quickly.
Out of the 7 I got in a lot, only a few are useable though, since some have different forces for the different directions of travel, and some also have a 'soft start' spring recoil mechanism in them as well. I have not seen any more available since, but am continually searching. Rather than destroy my current gimble to accommodate the new parts, I am just making a whole new one. I wish I could afford another ratted out Cougar, but in the meantime I will use the Cougar I have, just mated to an entirely new gimble.
I have made a completely new second pit, so at some point want a second stick anyhow. I am also looking into rotational dampers made by ACE Controls, which provide a much more economical solution (Sesco dampers are around $600 apiece!) that would be more easily repeatable, by others owing to availability. I have been doing the trig calculations trying to establish the correct parameters for a ground-up stick design based on new parts, an actual engineered solution. I'm going to call ACE in the next few days and go over my calculations and application with them, and I'll be posting about it here at some point but can pm you sooner if you are interested and I turn up anything worthwhile.
The new gimble design is proving quite challenging, owing to the restrictions in design that I am imposing on it in order to make it look and incorporate the same exact way as it's progenitor. If I compromise any one of 4 parameters however, it will be much easier to make and only at small visual expense, but I'm generally stubborn in my design efforts so the restrictions remain and I use my current setup as-is for the time being. I am approaching the 2000 match mark on this hardware, and it's showing no signs of fatigue. There has been some revisions and minor repairs along the way, but each time addressed a known weak point and there is little left on that list now.
*my stick has a rough and fine digital aimer, using the china hat and d-pad respectively. These work really well in this situation, and I also now use a control scheme that automatically dynamically scales the stick sensitivity to cycle along with zoom functions, which was actually pretty easy to do in TARGET, who really ought to be paying me at this point ;)
#159
Posted 13 March 2013 - 03:10 PM
I have a different project going (ICP F-16 Falcon cockpit module) but am having trouble programming my Teensies (I purchase three more ++2.0 boards). I cannot get them to load into the programmer in GenericHID. Nothing has changed since the last time used it but no go. I also installed the program (and the library) on the Windows XP computer running my CNC but it too cannot see any of the Teensy boards (even the original one?!).
Any ideas?
#160
Posted 13 March 2013 - 04:07 PM
7 user(s) are reading this topic
0 members, 7 guests, 0 anonymous users