after the transition to 1.11 there were some problems finally resolved by an updated sdk version from otoy. but i'm fighting a memory leak for several days now, which eats several gb ram after a few hundred frames of animation rendering - what is of course not exactly good.
apart from that, the 1.11 update is promising; using my usual test scene - a stonemason city ruins building, which is of medium size and uses quite a bunch of individual objects and lots of mats, the update from 1.02 gained 10-20% rendering performance, when using individual meshes (which allow fast scene updates, as you may be aware of). while hunting the memory leak i was also forced to implement an alternate scene building method, which builds all scene elements into one single mesh. this is now user-selectable for "final" renderings, and is always used for animation rendering, since it adds another 5-15% in speed (depending on how much different objects are in a scene). and depending on how the memory leak will be resolved, animation rendering will possibly have a pre-build mode which builds the next frame geometry while the current frame is rendering, actually cutting inbetween-frame voxelizing time to zero.
these speed improvements actually mean, that the plugin is able to render as fast as the standalone, where you usually have just a single mesh. until now the plugin would have rendered always notably slower because of splitting a scene into as much different meshes as possible, to allow individual updates - since this is now selectable, one can use the appropriate mode to get fast updates while tweaking a scene or fast speed for doing a final rendering (or animations).
another improvement is of course multithreaded voxelizing within octane (and is also causing my troubles btw). a genesis for example is built in 0.1-0.2 seconds using the base mesh resolution, on subdivision level 1 it lasts 0.3-0.5 seconds.
other additions like dirt texture, matte switch for diffuse mats, and other things, that came after octane 1.02 was released, are of course supported.
and a note about the manual; future octane/plugin manuals will be available online, and this is already also the route for the daz plugin manual. apart from that otoy allows to release a "work in progress" manual, of course to allow an "earlier" release. my experiences with the testers from the closed beta is, that there is not much explanation needed to get into work with the plugin, also there is a bunch of tooltips available, and having some standalone knowledge, this wip manual should not be a problem for early adopters. on the other hand i will learn from your feedback (as soon as you can get your hands on), what parts and concepts will need in depth explanation.
finally: i mentioned the GTC next week... otoy likes to have as many plugins ready for the next week as possible - no more to say

@ericg: there were a lot of iterations already - just not in public

my slogan was and is: if the final product makes me happy in first place, it'll make everyone else happy too. still believing that, even if it means i missed release promises not for the first time. that the amount of troubles achieving that goal reached unthinkable heights wasn't also really planned, and not all causing this was in my hands

@itou31: a few pages back you posted about the poser plug and my work - well, paul did imo a great job with the poser plugin, is a nice and very helpful guy btw, has done 2 plugins in a time span i haven't even completed one, and apart from all that, his poser plugin has one major advantage: it is available for a bunch of months already

@all: the sun is rising already (outside my window, not just in the float3daylightsystem node)... so, good night!
