distributed rendring - different brigtness of buckets on master and slaves

[fisla2]

hi, when i render one scene with dr it makes a bucket grid. master buckets are lighter than slave buckets. it makes just on some parts of image, there is no visible border between parts with same buckets and different one.
i tried to use just one material, and it still makes this checker, so it cannot be some missing map.
finally i found out that it is gi problem. (it was allright without gi).
when i use irr map with lightcache it works fine. but when i use brute force with light cache or brute force with brute force it makes different buckets. but when i use brute force and no secondary bounces it is fine again.

when i switch master with slave it works absolutely the same, master is still lighter... so on the same machine is different result if it works like master or slave.

did anyone have this problem, is there solution? any tips?
thx. fisla.

[vlado]

Do you have the GI send/receive multipliers changed for any objects in your scene (through the V-Ray object settings dialog)?

Best regards,
Vlado

[fisla2]

yes i had different generate GI multiplier on some object/objects. when i set up it back to one it works fine. thx a lot.

[vlado]

It will also work if you set your regional settings to use "." as a decimal symbol, instead of ","

Best regards,
Vlado

[kellyctv]

I'm having the same problem, but I have not messed with the GI send/receive multipliers. My regional settings to use "." Any other reasons that this is happening?

[vlado]

The other reasons are missing texture maps or discrepancies in the gamma values.

Best regards,
Vlado

[guapador]

distributed render bucket variation

It will also work if you set your regional settings to use "." as a decimal symbol, instead of ","

Best regards,
Vlado

I have the same issue. I check off Generate/Recieve GI and Visible to GI for only my glass objects in the vray properties dialogue box. When the image renders, the master machine produces buckets that ignore this setting while the slaves produce an accurate result.

All machines are set to gamma 2.2 and use a "." as a decimal point in the windows regional settings panel. Is there another fix?

Many Thanks,
Yan