I’ve just updated to 3.6.05 from 3.2 (I know …it’s a bit late) …but I’m having issue with the vrfb
When I load a previous rendered image from vrfb history and make a region to go on fine tuning details then it clears the buffer (it turns to black);
it’s seems a nonsense …is it expected to work this way?
no ideas, suggestions or whatever?
I’ve tested with a new simple scene and it works without refreshing vrfb otherwise with my last three production scenes it always do a full refresh when using region over a preloaded image from history.
I’ve also tried to merge the scene in a new file then I have copied render settings one by one …well, it works untill I add all previous used render elements (as new ones, not merge) …after doing that the vrfb start refreshing.
So, the last difference among scene is now that history image has lens effects on (then I’ve effects results, bloom and glare elements as well) while the new scene don’t and it makes no difference if I switch on/off lens effect before region.
As side note it seems that “clear image” button dosn’t fully work: it clears the vrfb but element names are still visible in the dropdown list
…so I suppose there is something broken inside my scene but I’ve no more ideas where I’ve to look.
Thank in advance
Hello,
Sorry for the delayed reply.
Do you have render elements with names longer than 64 symbols? There’s a bug that we haven’t fixed when render elements names are longer than 64 symbols the frame buffer is always cleared.
Best regard,
Yavor
I’m trying it with the same elements and things seem correct here. Can you send the scene (you could delete most of the geometry).
here it is, the file is a stripped one from a problematic scene
I insert just a dome, teapot camera and plane but the behaviour is the same
repro steps:
- render
- save image to history
- clear buffer
- reload image from history
- press region and select an area
- render
vrfb closes and open as empty rendering only the selected region on black
test vrfb.zip (86.5 KB)
Thanks for the scene. With it it happens here too.
We have this bug already logged in our system and it is already fixed for V-Ray Next. Not sure when a fix can be merged back to 3.6
do 3.6 nightlies have already this fix?
is it already present in 3.6.04 (instead of 3.6.05) or not?
Unfortunately no the fix is not in any branch of 3.6
Well, let me sat this it’s not a so minor issue for me.
History WAS extremly useful to continue fine tuning details day after day: you were able to render full res once in your spare time then reload the vrimg everytime you think of something to fix or improve some part of the image; actually a lot of time it’s wasted just to rerender the image in order to restore the starting point …this way history is almost useless.
So if you can bring up priority fot this fix …please do it.
Thanks
PS
Release notes for the nightlies are not updated anymore …so we don’t know what’s fixed and what is not
hi guys,
I’m sorry to bother you but, as release notes are broken, I’ve no other ways to be updated
any news about those two issues?
thanks
Unfortunately there are no changes to those.
forgive this bump
hi yavor, any news about this?
Hi,
Sorry for the delayed reply.
I hope that in a day or two there will be some fixes in the 3.6 stable nightlies related to the buffer clearing.
Will post again when ready.
Best regards,
Yavor
good news!
thanks a lot
The fix should be in today’s 3.6 stable nightlies (28979)
super!
I’ll test it asap and let you know
well, the main issue seems to be gone …thanks
…but now I’ve a new one:
is there any know issue with load/compare VRay-settings? …I mean the one that pops up right clicking history images
It seems sometimes settings are saved correctly and sometime they are not (ex it says they are equal when I’m sure they are not)
I’ve noticed that reloading some vrimg rendered with external render farm but it seems to work when if images are rendered locally
I don’t think there’s any known issue similar to what you are describing. Could you share some reproduction steps or a vrimg file that gives you wrong results?
got it, so I’ll try to investigate more closely this issue in the next few days and keep you posted