Jump to content

[1.6.x] RasterPropMonitor - Development Stopped (v0.30.6, 29 December 2018)


MOARdV

Recommended Posts

About the new HUD implementation ... would I be correct in saying that rendering the HUD over an external camera view is no longer supported? The reason I ask is that B9's MFDs used to do this but the cameraTransform field appears to be gone now.

Link to comment
Share on other sites

Thinking about it, I'm not really sure what it could be on my side.. Maybe an NVIDEA control panel setting isn't optimized? Iam running the settings that Nvidea recommend through a KSP profile from the nvidea auto update.. I'm willing to try anything if you have a couple ideas..

I've got an NVIDIA 670, and I'm not seeing the issue. The 970 shouldn't put any more demand on the system, since I'm running with all the graphics settings maxed, as long as you're not running with v-sync turned off (and really, there's not a reason to turn off v-sync in this game). Are you on Windows or Linux? If Windows, the default DX9 mode, or are you using --force-opengl or --force-dx11?

About the new HUD implementation ... would I be correct in saying that rendering the HUD over an external camera view is no longer supported? The reason I ask is that B9's MFDs used to do this but the cameraTransform field appears to be gone now.

I didn't realize anyone actually used the HUD like that. I did that for development, and I left it in. I can add that feature back in fairly easily.

Link to comment
Share on other sites

Hi MOARdV.

I've discovered a bug in RPM with regard to Transparent Pods. I am planning an upgrade release to DeepFreeze Continued... to implement a new transparent freezer part and plan to use JSITransparentPods.

During my testing I discovered a bug.

JSI Transparent pods work fine. The issue is related to JSINonTransparentPod. Or more specifically. When you leave flight scene or recover a vessel with a transparent pod the Culling masks of CAMERA 00 are not having layer 16 and 20 turned off correctly. Then if you launch a new vessel with no transparent pods on it the internals are showing in flight view (picture).

This is supposed to not happen through the use of JSINonTransparentPod module being added to all parts that don't have a transparent pod. BUT - as I mentioned, the culling masks are being left on from a previous transparent vessel. So the next vessel that doesn't have any transparent pods is NOT executing OnGUI or LateUpdate correctly because in both of these it is checking a Variable JUtil.CameraMaskShowsIVA which is actually out of sync with the actual camera mask.

I thought this should be easy to fix. I noticed in JSITransparentPod tries to set the culling mask off in OnDestroy and OnPartPack. But I picked up this error message:-

Could not find camera "Camera 00" to change it's culling mask, check your code.

So I dug further, seems the Camera 00 is being disabled BEFORE OnDestroy fires, and this happens only SOME of the time. In my testing though it was happening about 8 times out of 10.

So I tried adding a onGameSceneSwitchRequested GameEvent (in my own dev copy). But like OnDestroy this does not seem to ALWAYS happen before Camera 00 is disabled.

So then I thought remove the JUtil.CameraMaskShowsIVA variable and actually check the culling mask , but that is a bit of a performance hit which is probably why it is a variable.

In the end I added an actual check of the CAMERA 00 Culling Mask in JSINonTransparentPod.OnStart and resetting the JUtil.CameraMaskShowsIVA var. It doesn't seem to have any ill effect on non transparent or transparent pods and it only Fires ONCE, so no performance issue.

so, do you want me to raise a github bug report with all this in it? do you want me to push the change to you in github?

Edited by JPLRepo
Link to comment
Share on other sites

(snip excellent bug report)

so, do you want me to raise a github bug report with all this in it? do you want me to push the change to you in github?

Thank you for a clear and detailed bug report. :)

If you've already found the fix, please feel free to push a change to the Mihara/RPM repo. If you prefer, you can open an issue with a link back to this post, and I'll take care of it when I get to bug fixes in the next week or two. I've got a big mess of code that I'm tidying up right now, so I can't look at it just yet.

Is there a way to change the prograde marker textures on JSIHeadsUpDisplay, for heading bar and horizon ladder separately?

Not yet. Would you want to supply a custom texture in this case, or use one of the other stock markers?

Link to comment
Share on other sites

I didn't realize anyone actually used the HUD like that. I did that for development, and I left it in. I can add that feature back in fairly easily.

It would be useful for the B9 stuff at least. At least one of the cockpits has awful visibility so it would be a huge help in that case. If it's too much trouble I can probably figure out how to add it back in myself...

Link to comment
Share on other sites

Not yet. Would you want to supply a custom texture in this case, or use one of the other stock markers?

I would like to supply my own custom textures... :3

PS: also, can you add "menuTitleFormatString" option to JSISASMenu?

I also found that only AND, OR operators are working with Custom Variables... NAND, NOR, XOR aren't...

Edited by luizopiloto
Link to comment
Share on other sites

Thank you for a clear and detailed bug report. :)

If you've already found the fix, please feel free to push a change to the Mihara/RPM repo. If you prefer, you can open an issue with a link back to this post, and I'll take care of it when I get to bug fixes in the next week or two. I've got a big mess of code that I'm tidying up right now, so I can't look at it just yet.

NP. I've created a Pull request.

https://github.com/Mihara/RasterPropMonitor/pull/323

Link to comment
Share on other sites

Updated on DropBox. This doesn't include the JSINonTransparentPodFix, and it doesn't have much in the way of new features. The main change with this dev release is internal, where the main computational module was moved from an InternalModule to a VesselModule (if you don't know what that means, don't worry about it). I would greatly appreciate additional eyes testing it - I've looked at it with my own Mk1-2 custom cockpit, and I've looked at it with a fairly brief test in the ALCOR lander pod, but that hardly covers the gamut of RPM uses.

What I'm specifically looking for:

  • Any difference in performance (laggier / less lag / memory hog / whatever)? My PC is fast enough that I don't hit a lag problem like some people have reported, and there seem to be plenty of memory leaks in KSP (at least my installation), so I haven't noticed a difference, although that also wasn't my focus.
  • Anything broken? Parts not working, NullReferenceException in the log related to JSI / RPM modules, plugins that work with v0.21.2 stop working, that sort of thing.

Any feedback would be appreciated - I'm holding off pushing this change to GitHub until people have had a chance to try it, since it's easier to fix things now or abandon the change if it makes things drastically worse.

Link to comment
Share on other sites

I dont think this is a bug but when I go EVA, from the Alcor, I get this banner in yellow letters about checking my configuration... It wont go away unless I go to the space center and back...

Link to comment
Share on other sites

hey MoardDV... Is it possible to add horizontal bars to JSIHeadsUpDisplay? :P

Not yet (other than the heading bar, which is hard-coded behavior right now). You're the second person to request it, so maybe it's time for me to implement it. :)

Link to comment
Share on other sites

I have a problem with grey screens: http://postimg.org/image/wdhbnbzpj/ any ideas?

Are you using texture compression / changing mods? That's usually the culprit, if they're trying to change the textures used internally by RPM. Next possibility is incorrect installation. Without more information (such as log files), I can't be more specific.

Link to comment
Share on other sites

I also found that only AND, OR operators are working with Custom Variables... NAND, NOR, XOR aren't...

I forgot to hook up NAND and NOR in the constructor - they evaluate correctly otherwise. That'll be fixed in the next update. XOR works for me - how is your custom XOR variable configured?

Link to comment
Share on other sites

well... to be sincere, I didn't tested XOR... but NAND and NOR I tested and it showed the var name instead of any value... :P

PS: I have found another minnor bug, JSISASMenu text strings for Radial In / Radial Out are inverted... In points Out, Out points In...

Edited by luizopiloto
Link to comment
Share on other sites

So, how this page looks?

It's taking more time than expected to finish... but I'm putting all my heart in this... :P

It looks okay, I guess. It's really hard to tell from just a screen shot. It'd be easier to tell if you posted a download link so I could look at it in game. :D

Seriously, I really like the "hint" icons next to the navball for finding the cardinal orbital directions. That is a really clever touch. Now I wish I could find the time to work on my own MFDs and IVA redesigns... Maybe after the next RPM release, I'll take a vacation from working on it (other than bug fixes) for a few weeks so I can use the mod instead of updating it. :)

Link to comment
Share on other sites

It looks okay, I guess. It's really hard to tell from just a screen shot. It'd be easier to tell if you posted a download link so I could look at it in game. :D ...

this is far from be completed... but I have finished the NAV section... :P

https://www.dropbox.com/s/8v3bq4gxpk2ysj8/VexcomMFD_0.01.zip?dl=0

Edited by luizopiloto
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...