Alex, for some odd reason, I am now able to render. The steps I took were increasing my memory limit to 3000MB in render options and also I tried software and open GL rendering briefly before switching back to DIrect 3D. After switching back to Direct 3D (still Direct x9), I deleted and re-created my Daylight system. Now it works?!

Also, Ive found when locking the FG map and reusing tends to produce problems if the geometry or the camera moves. Its like it creates the map for that particular angle and if you move the camera or geometry, there are voids where there is no FG calculation.

regarding fg maps:

FG is view-dependent. locking them does exactly that - they are “fixed” from the camera angle/frame you rendered originally and saved. so don’t lock it. just render every frame to the same map and it will build a large FGmap that will work across all frames. then lock it once the whole sequence is saved.

regarding your fix:

i’m pretty sure increasing the mem limit was your solution, then, not the others things you did. you should always set the mem limit to something realistic (i.e. not more than you have but always more than you need). setting it too low is pretty much guaranteed to crash. placeholder objects and conserve memory help occasionally too.

I have never heard of MR storing the FG map across several frames. I didnt think MR was THAT smart! Thanks for the info.

alexyork - 10 October 2008 04:06 PM

regarding fg maps:

FG is view-dependent. locking them does exactly that - they are “fixed” from the camera angle/frame you rendered originally and saved. so don’t lock it. just render every frame to the same map and it will build a large FGmap that will work across all frames. then lock it once the whole sequence is saved.

regarding your fix:

i’m pretty sure increasing the mem limit was your solution, then, not the others things you did. you should always set the mem limit to something realistic (i.e. not more than you have but always more than you need). setting it too low is pretty much guaranteed to crash. placeholder objects and conserve memory help occasionally too.

I’m having the same problem as Syndicate5. The weird thing is DBR was working just fine before I reinstalled Max ( and reinstalled the hot fixes and SP1 (does the order matter I tried installing everything from old to new)). I remember reading in the help concerning DBR to check stuff in the Translator Options. Before net render kept having errors wouldn’t let me render MR on my two slaves, now its ok but I can’t DBR (since I last reinstalled while keeping all my old settings and profile).

I think it is the memory, but after reading this thread and giving it another try with these recommendations it doesnt seem to work.

MSG 0.0 error 011228: mi_msg_connect: failed to create active socket
MSG 0.0 error 011706: cannot add host 192.168.1.165:7509
MSG 0.0 error 011228: mi_msg_connect: failed to create active socket
MSG 0.0 error 011706: cannot add host 192.168.1.165:7001
PHEN 0.3 warn : mia_material(_x): Refractive falloff is used, but the shadow mode is not set to ‘Segment’.
PHEN 0.3 warn : Due to this, the falloff will be fixed, and not be influenced by distance.
PHEN 0.3 warn : This can look as good, but is not as physically accurate.
PHEN 0.3 warn : If distance-dependent falloff is desired, the ‘Segment’ mode must be used.
PHEN 5.3 warn : mia_material(_x): Refractive falloff is used, but the shadow mode is not set to ‘Segment’.
PHEN 5.3 warn : Due to this, the falloff will be fixed, and not be influenced by distance.
PHEN 5.3 warn : This can look as good, but is not as physically accurate.
PHEN 5.3 warn : If distance-dependent falloff is desired, the ‘Segment’ mode must be used.
PHEN 0.3 warn : mia_material(_x): Refractive falloff is used, but the shadow mode is not set to ‘Segment’.
PHEN 0.3 warn : Due to this, the falloff will be fixed, and not be influenced by distance.
PHEN 0.3 warn : This can look as good, but is not as physically accurate.
PHEN 0.3 warn : If distance-dependent falloff is desired, the ‘Segment’ mode must be used.
PHEN 0.3 warn : mia_material(_x): Refractive falloff is used, but the shadow mode is not set to ‘Segment’.
PHEN 0.3 warn : Due to this, the falloff will be fixed, and not be influenced by distance.
PHEN 0.3 warn : This can look as good, but is not as physically accurate.
PHEN 0.3 warn : If distance-dependent falloff is desired, the ‘Segment’ mode must be used.

Oops Updated: I mistyped my IP address. False alarm but may still be useful. I have a feeling this could happen to anyone since there are a ton of stuff to double check.

ive had this many times. It has always been one object, with multi subobject material where one of the ID’s is missing. like .. 1,2,3,5,6,7, <---- that would cause this error.
adding 4 into the mix there would fix it. maybe this is your problem.

although i have never managed to produce this error manually.

if you have this problem but only if you use DBR. that is weird. I’ve only had it when using Ambient occlusion light.

try turning off DBR and adding an occlusion light to see if it produces the same error.