Food for thought...

(dmoc) #1

While awaiting some positive news about Blender’s future I have been reviewing other modelling packages and utilities. Despite the many and powerful features Blender has I can’t help arriving at the conclusion that what has seriously let it down is the lack of inbuilt import/ export options. Even if a comprehensive set of Python scripts were supplied as standard I suspect that the interest in both the free version (possibly even with a low cost) and especially Publisher would have been much greater. Although not a fair comparison once you realise Blenders potential, even MilkShape3D and 3DCanvas seem to be thriving despite having a fraction of Blender’s functionality. And they cost! What do you think?

Another issue is possibly using Python as the scripting language. I would NOT like to see Python removed because although I am not a Python expert (novice, if that) it has definate advantages for advanced user’s. However, I came across Lua (, a free and simple scripting library, and think that if this was provided in addition to Python then beginner’s/ intermediate’s would benefit greately. Lua is very similar to VBA. What do you think?

Finally, although I was one of the few who paid for Publisher (at the time simply to support NAN but also with aspirations) I am in limbo regarding putting it to any serious use. My problem is that not all the object hierachy appears to be visible/available from Python so that I for instance could write my own import/export scripts. Am I wrong? If anyone can clarify then please do! Maybe the Dynamica team know something?

At the very least I hope some of this post creates the sort of debate we use to have on the old forums and provides some feedback to Ton that might help in determining Blenders future.


(BgDM) #2

I can’t help arriving at the conclusion that what has seriously let it down is the lack of inbuilt import/ export options.

I agree a litlle bit here. I think however, that the biggest drawback has been the lack of a built in raytracer. This would then negate the exporting to other programs to do better renders. I would gladly pay a small cost to get a better blender program. THis has been discussed before and I think Ton should consider it, if and when, blender comes back.

I am not a Python expert (novice, if that)

I am no Python expert at all. I totally depend on the group to provide the scripts, etc. to make things easier/better in blender. I am truly indebted to all you Python guru’s out there :smiley: .

My problem is that not all the object hierachy appears to be visible/available from Python so that I for instance could write my own import/export scripts. Am I wrong?

I beleive you are totally correct here. You can not export some of the material and texture settings with Python, becuase of the invisibility of the hierachy.

At the very least I hope some of this post creates the sort of debate we use to have on the old forums and provides some feedback to Ton that might help in determining Blenders future.

As long as this doesn’t become a pissing macth, like some of the old forum debates became. It’s nice to stir the pot, but don’t piss in it.

(SimonHK) #3

I guess all this comes down to what is on the “big list of stuff to do” for whoever picks up the development of B. There has always been “debate” on what features should and shouldn’t be added to Blender.
I seem to remember someone had a webpage that summarised all the feature requests, anybody know where that is??

How about we have one here at Elysiun?

I agree that the Python API should be well up on the priority list, all the other 3D apps have an exhaustive API and it shows in the amount of plugins etc available…

anyway, still waiting for a bandsaw/knife tool type thing… :-?

regards to all


(acasto) #4

I agree that one of the problems is lack of import/export, but I also think a built in raytracer would be nice. I believe that what it boils down to, when you look at Blender it has the capabilities to compete with some of the bigger packages, but the bigger packages are more wideley accepted across the industry. Take maya for example, because it is the industry standard, there is tons of plugins and add on programs for it. I think that if Blender had some way to work (export) flawlessly with more ‘industry standard’ programs, it would help it.

When I was thinking about setting up a graphics studio, I would have liked to used Blender on most for modleing and such, and Maya on one or two.Then design on Blender and finish the more advanced features on Maya. But I have to wonder about how Blender would perform in the export. One thing I noticed is flip flopped axis compared to the other program. Other used and X,Z with and up Y.

Now if Blender could export easily to Maya or 3DSM, you could still have the fast modeling capabilities of Blender, yet be able to take advantage of the hudge benefits of all the third party programs such as ZBrush, AfterEffects, etc…

(Timothy) #5

I’ve been thinking about this yeah,…

Basicly the design I had in mind would be something like this:

  • anyone could post a feature request.
  • anyone could vote on this feature.
  • anyone could place comments regarding this feature.

A feature request would have to be fairly detailed in description, perhaps with the ability to add images to the request etc. Also all feature requests would be aproved by a moderator first (to see if there isn’t alrteady such a feature request,… or a actual implemented feature).

All of this greatly depends on what NaN will do though, but if such a page would be formed I think it would be great for any developpers of Blender.


(overextrude) #6

A missing set of comprehensive import/export functions is no doubt a shortcoming, but it is only one of several. After working with a model, vertex groups, and armatures for a bit, it has become painfully apparent that this functionally was intended only for the simplest of animation/production use. Unless I’ve completely overlooked something, there’s no way to make iterative changes to a model that has been rigged with armatures and vertex groups without a great deal of redundant work and hassle. If it’s in Blender’s future to become a commercial product, MUCH more attention needs to be paid to workflow/productivity issues.

(acasto) #7

Even if it’s not in it’s future to become a serious commercial product, it is a wonderful modeler. We just need it to be able to integrate into other programs easier that have the better animation tools and such. Right now the way it goes, is if you can’t do it on blender (besides rendering) your pretty much stuck. If it would play nicely with other stuff, there might be a lot more acceptance.

(steve343) #8

whats wrong with dxf !!!

no ofence i am new to this but i thought that dfx was a good universal export!

(acasto) #9

Isn’t it basic mesh only though?
Also, in most programs the axis is different from Blender, which would mean you would have to either create it different or roate it once in the other program.
The more processing you have to do to get something in another program is more chance for serious error to occur. We could probably write a python script to translate everything to the right axis on export though, I’ll look into it.

Blender can work in a production type environment, however it takes quite a bit of work with getting the scripts and everything layed out, then if something changes, you have version skew all over the place.

I think for a product to be sucessfull commercialy, people just want to install it and use it. Not search google all night for an ancient sript, then debug it to get it to work, then if they’re lucky get what they are hoping. The commercial electronic product TiVO has/had a 50% return rate which isn’t good. People said they want a packaged deal, that they just have to plug up and use. They didn’t want to have to learn how to then program everything themselves.

That is something I am trying to get with the Blender package. A set system that would allow you to interoperate easier than you currently can.

(steve343) #10

ok cool.

i find other programs a pain in the arse to moddel in so i dont use them mabey this is the answer.

(dmoc) #11

See below… Twice! Short time-out on login!?

I just spent 20 minutes writing a reply, pressed submit, was prompted for my password and then presented with a blank reply form again! What’s going on? Lost the whole the whole thing so excuse the brevity…

Inbuilt Raytracer: based on the number of past requests and comments I agree.

Import/ Export plus full hierarchy/object visibility: I believe top priorities as they allow many other requirements to be addressed.

Focus: NAN may have been spreading their resources too thinly and possible mis-directed in some cases (see next comment). Ideal seems to have been “Blender does everything”. In reality there is always another app/utility that does some things better.

Web-based Content: Good idea but slow uptake. Long runner Blaxxun went belly-up around the same time as NAN. I looked at VRML years ago for this sort of thing but it’s been slow to evolve and latest outlook indicates a new spec may replace it.

Ease of Use: Blender is a steep learning curve for beginners (I know!). Detailed locally installed on-line help and reference manual required.

Scripting: At least an option of Python and a VBA-like scripting language (eg Lua) is required. Personally, while I appreciate what Dynamica are attempting, so far it’s been vapour-ware and I don’t like having to wait on others. I also suspect the Dynamica team have inside info/help allowing them better insight when designing solutions. Apologies if I’m wrong!

Cost: Blender should never have been free. There is a discontinuity in peoples perceptions between low cost and free. A nominal fee may also have avoided the current situation.

Open Source: The longer the current situation continues the more I come around to thinking that would be the best route.

Publisher: Considering the sudden loss of support for those who, like myself, bought Publisher I think existing Publisher users should be granted an unrestrictive licence. Plans were made and have to go on hold until some concrete news arrives. If it takes too long most may migrate to other apps.

Community: In terms of determining Blenders future I think the community have so far been very passive. I think a features list and development road map should be created by Blender users and presented to represent the people who, outside NAN, have the most vested interest in Blender’s future.


(steve343) #12

ur problem of loging in is cos ur conection to the internet stoped thus logging u out of the forem. if u have a dial up modem like me always copy long replies befor u post. so if u have to log in agian all u have to do is paste.

(acasto) #13

I agree that one of the problems could’ve been the cost. A lot of people have a weird mindset…if it’s free then it has to suck!!! I have seen this many times with Linux. When it comes to certain things, people will pay just for the heck of it. Look at Microsoft, people pay for each copy of an operating system that continually locks up, just because it’s the thing.

Sometime charging more bring more assumed credability, people think if it’s expensive it must be good. I know this is stupid, but it is hard to change their minds. I have met many Microsoft and networking consultants and they all have this mindset. One Microsoft consultant once told me that “Linux made a terrible server solution and that it was rarely used in the industry for servers”…it was one of the most arrogant comments I’ve ever heard.

A good example is Exluna, the company that make BMRT. BMRT is free, but the commercial version, entropy, is around $1500. The biggest difference is the price and name (although entropy is faster and comes with support and a few other things). I think this protects it from the “free” image and gives it more commercial credibility.

Maybe if publisher was made to just ‘seem’ drastically better than the regular blender. Give it an image of being a great new thing out on the market, and maybe charge a little more.

I don’t know, just my 2 cents