Shadow blinks

L

lavi

New member
when landing a helicopter (i tried few) the shadow on the table blinks.
in the same time there is a constant shadow on the ground (and there shouldnt be).
I have all graphics set to maximum.
 

Attachments

  • ScreenShot1180779550.jpg
    ScreenShot1180779550.jpg
    294.4 KB · Views: 82
  • ScreenShot1180779551.jpg
    ScreenShot1180779551.jpg
    294.2 KB · Views: 84
At least the shadow on the ground is a common problem. G3 has some issues with shadow projection. The fuselage's shadow is also visible on the rotor blades for helicopters.
For me, the shadow is sometimes on the object and sometimes not when the heli's over an object. It's not actually flickering or blinking though. But maybe if you manage to be on the "border of visibility", it might as well appear to blink.
 
Last edited:
I just noticed last night that I have shadow blinks and the landing gear on my Nextar blink. I don't often use my real PC, as I'm not living at home now. Anyone wanna buy a house in Hopedale, OH??? I updated the video driver, with no improvement. Searching around other threads shows that Vista and Directx may be a problem. I also saw where the latest drivers may not be the greatest drivers. And to try other non-oem drivers. This problem wasn't there before I upgraded to Vista. I think I'll start an 'I hate Vista' thread.
 
The problem has always been there...

There is no culling of shadow projections and as a result, shadows project THROUGH solid objects!

G3's always done this from the start.

Eliminating this would kill CPU power... imagine what those leaves on the trees would cost in terms of CPU utilization....

( And yeah I know you CAN clip this, but you would have to do so at a distance... ).
 
Can't stop FLICKERING !!!

I can't stop the FLICKERING on the shallows either!! I'm running Nivida GT7600 on XP Media with the latest drivers and it still flickers. I have done everything I know to do, but with no luck. FLICKERING CONTINUES ......
 
My shadows are flickering too and the textures are garbled on the planes and other objects. When I disable shadows the flickering disappears mostly but there is still some textural abnormality occuring. :confused:

My graphics card is an ATI Radeon X1900 PCI-E with 512 MB and I'm running G3.50.060 on Vista Ultimate x64 with ASUS A8R-MVP motherboard, 2 GB RAM.
 
What was the point of the last question?

I answered it in the prior posts.

You can't do anything about this. At best you could play with driver settings to help minimize the flashing effect, but you will not be rid of it.
 
This flickering issue is quite annoying. Does anyone knows if this should be solved in the next Real Flight release, or if developers are working on a patch ?
 
What I see is that when the heli moves over the edge of the table there´s some point where the ´table-shaddow´ instantly turns on and vice versa in the opposite direction. I can´t call this flickering.
With a dominion 90 it is no problem at all. with the trex 450 based on Mx400 often you never reach `the point`. So the shadow stays off even if the heli has landed on the table.

Sitting motionless on the table I see no flickering.

I´m sure this behaviour could be done better but for me it´s ok and my bigger sorrow is the performance of my 5 year old machine here... :D
 
Unlikely, as light culling will probably require a good bit of rewritting of the entire engine...
 
I´m often reading in the forum of my space-simulator celestia and these problems are quite common there. (moon shadows on planets, satelites flying through planet shadows, satelites flying through moon and planet shadows, parts of satelites or planets eg.saturn rings covering each other... )
It is possible without doing everthing from the start, I observed the development process ( only as spectator ) but you can be shure you´ll have to pay with performance.
 
opjose said:
Unlikely, as light culling will probably require a good bit of rewritting of the entire engine...
Click to expand...
I don't think so. Indeed it's quite surprising that KE is not able or willing to update the shadow code. Even 5 years ago, there were articles and free examples how to implement solid shadows based on stencil shadow volumes.

Either they are using very simple projected shadows which were used in games 5 to 6 years ago or there's a bug in the shadow implementation. Indeed it could even be a problem of using unclosed surfaces. There is a similar bug in the transparency code which even extends to the collision system and which IMHO is caused by unclosed surfaces.

Then again, I can live with these bugs and would prefer KE to finally fix the major bugs in calculating the effectiveness of control surfaces.
 
0xdeadbeef said:
Either they are using very simple projected shadows which were used in games 5 to 6 years ago or there's a bug in the shadow implementation.
Click to expand...

From the artifacting I see, it's the former, hence my assertion.

0xdeadbeef said:
Then again, I can live with these bugs and would prefer KE to finally fix the major bugs in calculating the effectiveness of control surfaces.
Click to expand...

"Major" implies that control surfaces do not work as expected at all.

Isn't it more a question of fine tuning.

What are you referring to? Stalls?
 
opjose said:
From the artifacting I see, it's the former, hence my assertion.
Click to expand...
At least some of the bigger birds show self shadowing. I'm not sure if it's generally not visible on the smaller planes or if this is just coincidence. Maybe it's also somehow related to reflection. Anyway, self shadowing is rarely seen with simple projected shadows. That made be think it's maybe a really, really buggy volume shadow code.



"Major" implies that control surfaces do not work as expected at all.
Isn't it more a question of fine tuning.
What are you referring to? Stalls?
Click to expand...

"Major" means that the effectiveness calculation is completely off in some cases. E.g. the rudder of small planes is usually much too effective compared to the rudder size. This is partly due to the fact that the whole vertical wing surface seems to be taken into account. But even if you replace the default vertical tail with generic fixed and movable wings, the result of the latter is much less effective. This is a clear indication that there is a bug in the control surface calculations.
The same effect is visible on all the other control surfaces, but less pronounced. For bigger planes the error seems to get smaller. Probably since control surfaces stall easier for heavy planes which might partly hide the boosted efficiency. Maybe there are other factors.
Anyway, it's very irritating and frustrating that modeling control surface as close to the 3D model as possible usually results in completely weird behaviour for small and light planes.
 
SOOOO COOOOOOL. Since upgrade 3.50.065, no more shadow flickering :)
Knife Edge developers rulez :cool:

David
 
.065 didnt fix shadows

updating to .065 did nothing to fix the shadow problem. Seems to be worse on some airport than others... Ugg...pretty poor programing in todays age....
 
You must log in or register to reply here.
Back
Top