blendigo SW 1.1.10 (Beta)

Announcements, requests and support regarding the Blender Indigo export script
pxl666
Posts: 308
Joined: Wed Oct 10, 2007 10:10 pm

Post by pxl666 » Sat Nov 01, 2008 7:49 pm

SmartDen wrote:soon. maybe tomorrow
uhuh - CANT WAIT!! will there be layers and .igm support???

pxl666
Posts: 308
Joined: Wed Oct 10, 2007 10:10 pm

Post by pxl666 » Sat Nov 01, 2008 7:54 pm

Big Fan wrote:BTW another aspect to this is that Blender 2.50 work has begun.
This will be a sustantial overhaul that could be very helpful for a better Indigo/Blender workflow.
There will probably be a number of UI improvements intertwined with python and it seems likely a renderer API will happen too so in fact it may be a good time to stop this particular line of script development or general exporter approach.
I would not actually want to spend too much time moving forward into the the 1.2.x series without being aware of where Blender is going or indeed where Indigo is likely to go.
It may be a good time to completely overhaul many aspects of the script and interface or even abandon a script in favour of a plug-in.
Already I know someone is looking at easy incorporation of icons in scripts although I am not sure where this fits in with the 2.50 initiative.
I have mentioned to the person about a native support for zoom and scroll for the scripts window as well but I really dont know what is being considered in Blender 'head office' for the 2.50 refactor to say how things might pan out for a Blendigo in the future
Python 3.0 is going to be a little different too but as far as I can tell the impact on blender Python and scripts is not too great.
A wait and see approach may be best at this time but all said and done I wont be doing any more updates. For me it is a convenient place to leave it.
regards
BF

EDIT: you out there @JeffR?
I was wondering what the state of the SW API exporter for Indigo is?
hi bigfan - could u briefly explain what are the differences between export script and a plugin??? ppl talk about this issue but so far i see that scripts are very good idea since (for example) indigo is being updated almost every week and what we need is just to download a new version of the script and we are good to go, if we have exporter implemented within blender itself we would have to download appropriate build of blender...what is pointless in my opinion...am i missing somenthing? would u make thing clear pls??

thank u

Big Fan
Posts: 745
Joined: Tue Oct 17, 2006 9:37 am
Location: Nelson NZ

Post by Big Fan » Sat Nov 01, 2008 9:01 pm

yeah well it is true Indigo has been in constant development and it has been updated on average probably every 3 weeks or so however ATM except for people having my script version as a stand in the official blendigo hasnt been updated for a long time.

in fact as things have panned out it probably could have been available only with blender versions 2.47 and 2.48 and the situation would have not been any different re waiting than we have had recently.

I really dont know how useful the renderer API will be to blender/indigo.
I guess it will allow the integration of an exporter not unlike yafray was done in the past however the potential is that it would be better handled inside Blender with the new UI treatment.
Of course it would remain optional because the API would be just as available to Lux as to Indigo or any other renderer.

I see no reason why you couldnt download a plug-in for indigo 1.x that will work with blender x or y - I dont forsee there to be that much change post 2.50 that you might need to have a specific version made available with each blender release. Even so it probably wouldnt be any more inconvenient than having to update your blendigo/indigo versions and with the ancilliary conf file , previews, IES, camera responses etc etc
People do update python itself and blender fairly often.
Again I cant say -its a matter of seeing what possibilities emerge with the release of 2.50 -perhaps a script would be best after all.
I didnt say it should be done - just noted the possibility or potential.
I wouldnt get too upset about it.

I guess a plugin means it being done in C or C++ and presumably it would incorporate ono's condensed file formats etc. as well so it would be very much faster.
I would like to think various tools could be rounded up and integrated as well if that is the case but of course someone has to do the work.
I am sure people are aware Lux for instance is bringing in the material preview function to the blender UI
I am sure there are other possibilities for integration that would interface with the home application as well - IES lights in Blender are talked about and there are all sorts of render goodies creeping into blender including atmosphere and volumetrics with a mostly physical implementation
Of course being in C means finding people with different coding skills but you might think of the python scripts as a development mode for C integration - a low cost opportunity to sort out and trial how things should be implimented and arranged in 'real' code.
There have been a few scripts that have moved on from python to C

I am not sure that indigo will always have the same development speed as now. It seems to be mid life now as far as useful features go.
It may well be that an update every 3-4 months that could coincide approximately with a new Blender release would work quite well.
It isnt absolutely necessary to publish a script update for every indigo version except by doing so we do ono a favour by finding his bugs quickly.

It also should be remembered that the people who do updates have real lives and that you cant expect them to be continuously on the job every 3 weeks and doing substantial re-enginneering of code at the same time if necessary.
I imagine smartden is a very busy fellow ATM to make up as much ground as approx 10 test releases in one hit.
Its all very well to say 'thanks' for these things in a forum post but believe me there are a lot of hours sunk into what you download in a few seconds.
I would hope that we can actually have some stability of direction so that exporter developers dont have to do extensive rewrites at this stage of the game and into the future.

I am far from an expert in coding but IMHO the size and complexity of script we have presently is about as much as you would really want to make it I think. I dont think it is really intended for the heavy duty use we are extracting from it. At one stage I went looking for a LAA build of python because it was exceeding 1.7 gb of ram use for ex SW import/export. This is serious space use for a tack on faciity - sorta like the tail wagging the dog.

If ono gets seriously into shaders I dont think that python has tools to handle it - a bit like the tabulated spectrum stuff we had to pass on. There are some things the python API isnt suited to.
In my script version I have hardcoded some shaders but I wouldnt want to do a whole lot of them

The UI capabilities of python are also somewhat limiting when it comes to making a decent front end to things.
I am not really sure how what I have done is perceived as far as presentation or practicality but it isnt that easy to work with inside the script and it isnt all that adaptable to change.
People complain if it doesnt fit their window or its too big or small and yet script writers have to squeeze in new things all the time and keep it tidy and logical.
It would be nice to have more tools in regard of those that are more closely linked to blender itself.

In any case where blender/indigo goes is not my concern.
I am happy to use indigo at version 1.1
For my needs that is enough.
Dont worry about it in the meantime.
Go try to make some nice renders
People dont pay attention to what I say, think or want anyway :lol:
HTH

Post Reply
78 posts

Who is online

Users browsing this forum: No registered users and 8 guests