29 replies to this topic

[ArchMage Sparrowhawk]

Chuck YeaGurr, on 14 December 2015 - 12:09 AM, said:

If it was possible and could easily make money most companies would be glad to implement it. But implementing technology across a wide spectrum is never simple. After a couple of years any system may have a multitude of hidden patches/fixes that have to be applied because of the interface of users/hardware/software they can not control but are required to make work. The folks who take the abuse from the public when things suddenly become glitchy are the same folks who take the abuse because some new cutting edge tech is not being implemented.

Well being an easy money maker is one hing. But simple is a relative term. If they looked and found that all they would need to implement is a basic receiver widget or do some straightforward method like the Elite: Dangerous folks figured out how to do somehow, it might be worthwhile. If it's straightforward, if not necessarily "simple and easy" but in the realm of possibility, why wouldn't they do it? I'm not suggestion some perfect solution that covers all hardware devices, hell we don't have that with joysticks and mice. But, to open the door to possibility, and such possibility at that? I think they could find the time. Some time. 10 minutes.

[Chuck Jager]

ArchMage Sparrowhawk, on 14 December 2015 - 12:29 AM, said:

Well being an easy money maker is one hing. But simple is a relative term. If they looked and found that all they would need to implement is a basic receiver widget or do some straightforward method like the Elite: Dangerous folks figured out how to do somehow, it might be worthwhile. If it's straightforward, if not necessarily "simple and easy" but in the realm of possibility, why wouldn't they do it? I'm not suggestion some perfect solution that covers all hardware devices, hell we don't have that with joysticks and mice. But, to open the door to possibility, and such possibility at that? I think they could find the time. Some time. 10 minutes.

It takes more than 10 minutes to set up a test server.
I have had to work directly with Adobe implementing a flash server on a university website. everything worked at multiple levels and we were told by high level support that there should be zero problems. Except no - the federally mandated security protocols for student data that were being used on our servers blocked the streams. Long story short there are too many customized issues that do not allow simple installs of new technologies that service a broad base. Also it is never in any company's best interest to say yes or no to a new technology until it is mandatory or broken. It either becomes a "promise" or a "negative". I do not like this but no information has a harder time of biting you or being misconstrued.

If it really was easy they would do it. If it is not possible it is not in their interest to discuss it until it has a felt effect on their ability to operate/profit.

[ArchMage Sparrowhawk]

It takes 10 minutes to set up this vr to see what's what and to go from there.

Test server isn't even really necessary such an early point. Just from a set up the hardware and try it for yourself standpoint, not trying to support every ios and android device out there all at once.

Ok, from a bird's eye view, this is not a worm, fine. But they'd still be smart to usher this in.

Edited by ArchMage Sparrowhawk, 14 December 2015 - 02:33 AM.

[Kissamies]

ArchMage Sparrowhawk, on 12 December 2015 - 01:56 AM, said:

So, using FreeTrack and a webcam + LED clip I was able to get head tracking to work!

So I can now move the mouse just by moving my head, and it's fairly accurate.

Now....the problem. Mouselook and torso twist all seem to be bound together.

I can't seem to control only the freelook with the mouse or bind it only to a joystick, because the moment I activate freelook, the joystick's torso twist is also bound to freelook. '

Seems like a support limitation, but maybe someone knows how to futz around with the bindings and mouselook controls in an ini?

Can't you use the game's innate TrackIR support with Freetrack for freelook? I do this for my EDTracker with Opentrack and it works just fine. Granted, EDTracker is just a head tracking device, not VR, but the same principle should apply.

[Mycrus]

Troutmonkey, on 19 November 2015 - 03:04 PM, said:

Short answer - No. MWO has to plans to support Google Cardboard VR.

As a dev, I can tell you that it would be hard enough getting one VR system to work, let alone trying to be compatible with 100's of different Android phones with different resolutions, drivers, and gyros. Heck, PGI can't even scale the UI to work on a triple monitor setup yet. It's not going to happen.

Don't even talk about surround... simple 3d vision is borked (HUD)

[ArchMage Sparrowhawk]

Kissamies, on 14 December 2015 - 02:58 AM, said:

Can't you use the game's innate TrackIR support with Freetrack for freelook? I do this for my EDTracker with Opentrack and it works just fine. Granted, EDTracker is just a head tracking device, not VR, but the same principle should apply.

I was trying to get Freetrack to work with MWO, but everyone elsewhere is saying that the TrackIR folks changed their module so it only recognizes their signature. I wasn't able to get MWO to recognize Freetrack or any of the TrackIR type support, since I don't have any official NaturalPoint™ gear or software.

There is a program than can be downloaded that supposedly changes the MWO files and fixes this, temporarily, but I'm a little wary, so of course would like any kind of more permanent solution.

[Kissamies]

Well, the Freetrack ("freetrack 2.0 Enchanced") protocol that comes with Opentrack works fine for me. Worked with 2.2. With 2.3 it didn't initially work, but it started to work a few release canditates in. Currently I have 2.3 rC21p4. The previous one I used was 2.3 rc12p1 and that worked too. I don't remember what one before that was, probably RC3 or something, but that didn't work. I've never done anything to MWO to get it work and I don't have any NaturalPoint software either.

I think they have kept tinkering with Freetrack to get past those NaturalPoint blocks.

Edited by Kissamies, 14 December 2015 - 12:59 PM.

[ArchMage Sparrowhawk]

Success.

Thank you Kissamies, otherwise I would have never stumbled on to this. So, get this. Opentrack works with MWO using the Freetrack enhanced. Freetrack 2.2 last updated in 2008 doesn't work with MWO, but thats the only tracker that currently works with only a single point LED model (the one I have). Otherwise if I had a 3 or 4 point LED model (4 lights in a square or triangular array on the HMD/headset), it would work off the bat with Opentrack. (If there is another that works with a single LED besides Freetrack, I'd like to know. )

But, I was clicking around and I had both running at the same time, and I discovered if I start MWO and it detects the tracking from Opentrack, I can close OT and open Freetrack, and it will pick up the tracking when I tab back into MWO. Nice. so. *******. nice.

Ok kiddos. So I have ~50fps video to my VR headset, and now I have head tracking, independent of mouse. And it looks divine. MWO, I love you.

[Kissamies]

Excellent! Maybe one point model gets added to Opentrack some day. There seems already be some discussion about it, but sthalik sees a bit dismissive about it. "Use aruco" and "I'll reconsider if someone writes the tracker", he says.

Any chance of any sort of video demonstrating this? I'm sick of those videos of Oculus Rift guys using their heads to aim. One with freelook would be better, I believe. Hmm, actually there is this recent video:

Edited by Kissamies, 15 December 2015 - 01:35 AM.

[ArchMage Sparrowhawk]

I'll consider it, once I get everything perfected, and get another camera since all of mine are engaged in the enterprise.

But yeah that video is pretty close to what I get with freetrack though it looks like he's using the accelerometer motion method of tracking, which is a little laggier, and I am using a webcam + headtracker program.