From Nicholas Bishop, one of the sculpting participants of the Google Summer of Code:
"Here's a rough guide to the effect your base mesh has on the speed of sculpting:
1. When you sculpt, it optimizes by ignoring parts of the mesh that aren't near the brush. There are two resource-intensive things that sculpting does: the actual brush calculations needed to modify the surface, and redrawing the areas that have been changed. For this optimization to work, it needs to break the mesh up into pieces. That way, just one small piece can be sculpted or redrawn, rather than millions of polygons in a detailed mesh. So the question is how is mesh split up?
2. When you multires a mesh, it gets split up into grids. You know how when you apply one level of subsurf to a model, each quad turns into four smaller quads, and each triangle turns into three smaller quads? Each of those quads counts as one grid in multires. So if your input mesh is 6 quads, like the default cube, you'll have 24 grids, whereas a monkey (500 faces), produces 1968 grids.
3. One grid is the smallest chunk of the mesh that can be sculpted on and redrawn independently. So no matter how far you subdivide, you have the same number of grids, and you have the same limit on the number of chunks the mesh can be split into for optimization.
Note that only the topology of the base mesh matters, not the shape.
As I said, this is a rough guide -- it doesn't contain any numbers! So you still have to experiment a bit to see what level of detail in the base mesh works best on your system.
I've just been figuring out how to use Blender for the first time. Learning how to rig, took me a while to find the information collate all the relevant information, and then get a basic one to work. It wasnt as complicated as I thought it, pretty staightforward. Which was contrary to all the tutorials I saw which spent ages going through a rote specific context rig. When all I wanted was the basic raw concept, of how to manipulate and create the rig, and then how to apply that rig to a mesh.
But I know its a beta.
I'd like to see more integrated help files within Blender. Like the Ribbon tooltips in 3DSmax or 3DCoat. Where you press a button in the preferences to say you want help tooltips. Then when you hold your mouse over any of the buttons, you get a pop with really detailed info on what it does, and an brief example of it.
I also wish the help was not so reliant on remembering keystrokes I know my brain does not remember them as well as I do a button that is located somewhere logically in the ui.
Well, a big bunch of people pay, get a say in what features they would want, these developers who otherwise wouldn't exist without the money add these features.
And everybody, even nonsubscribers would get these features due to the very nature of GPL.
Yea thats true, but not everyones chosen features can be done all at once and that is bound to cause some friction.
Im all for Blender improvements, but here are a few problems that i could see happening
If they pay £100/$150 per year subscription to fund a feature they want and it gets put to the back of the que or gets released behind someone elses chosen feature, people are going to get mad and i can see the whole "what did i pay for" argument popping up.
People who subscribe would be placed upon a pedestal above those who do not.
It is possible that feature requests for people who do not subscribe will be ignored or Devs will simply move along to the subscription wish list
I think a fairer funding model would be that the Devs would post a list of projects that there is interest in developing simultaneously and users donate to fund that feature or users could top up the wages of a Dev who is working on a feature currently for the foundation or under supervision by the foundation to ensure completion.
*EDIT*
I just read the Sunday Dev meeting notes summary on BlenderNation and it says that funding is not as much of a problem as actually finding people to do the work.
Meeting shortly discussed Tom Musgroves questionaire, and the idea to hire another developer based on people giving subscription based donations. Ton replied that theres budget already for 1 or 2 positions, based on current shop revenues and sponsoring. Its currently harder to find and manage people than to find money.
since I didn't upgrade to the yearly account you for surveys I closed the survey earlier to make sure I didn't get overage charges.
Regarding Tons statement about difficulty finding developers - I tend to disagree. He wants coders who are brilliant self managers and who work for peanuts and have a deep knowledge of every aspect of Blender. However, if you have a good manager - then the self management issue is less important. If you are willing to allow the developer to focus on their core competencies and grow into other areas - then you lose the need for having an uber developer that knows ever Blender nook and crannie. If you get adequate funding together, then the work for peanuts is on longer an issue. I have 10 or so developers willing to work for me, all of them with Blender commit rights, and could double or triple that amount if funding were in place. Most of these are Blender developers who have had to move on to other things because they having a life and working full time for free are not very compatible. Others are Blender devs who are students who are looking to enter the work force in the near future.
Metaliandi,
As to how it will work out - I don't think the risk of disillusionment is that high. The point of having the user identify multiple areas they would like worked on is to increase the odds of their desires corresponding with a large enough group that it increases their odds of getting something they want.
Based on the results on my survey the top wish list items cluster heavily based on what industry the individual works in.
Also for these subscriptions they would be on the order of 120$ a year.
Also there are items like - save confirmation for quiting in all contexts; or selective bevel - that a huge number of users have wanted forever but that are not priorities with any particular developer.
I think if I show major progress towards core issues that have been a priority for many users 'forever', as well as showing rapid progress on areas that will make a major difference on workflow and productivity, that the majority of users will feel that their money is well spent.
I think a fairer funding model would be that the Devs would post a list of projects that there is interest in developing simultaneously and users donate to fund that feature or users could top up the wages of a Dev who is working on a feature currently for the foundation or under supervision by the foundation to ensure completion.
The bounty model works absolutely horrendously. It has a number of substantial flaws that make it unworkable. 1) The typical artist for whatever reason places the value of any particular feature at between 5 and 20$ or rarely for something that is a years worth of developer time they might go as high as 100$. The actual dev cost of any particular feature is usually 1000$ for a simple feature to 10,000$ for a moderately complex feature. 2) The 'cost' for both the developer and artist is high in terms of amount of time invested versus features gained 3) the developer essentially has to complete the feature - then hope that the funding will come through. Funders tend to be fairly unreliable for post hock payment (lost interest, misestimated cash flow, never intended to pay, or a whole host of other reasons). 4) It encourages developers to game the system - withholding features and improvements since they are a potential revenue stream.
As to 'contributors being put on a pedestal' they already are. They get mentioned on web pages as donors; they get their names in movie credits, etc.
Also for those who aren't interested in subscribing there is always art for code. Of course good feedback on tools or clever ideas also get implemented. Artists who show great artwork done with the tools and point out limitations will also get listened to. Those who contribute in some way - either feedback or funding are listened to.
If an artist doesn't provide either I guess I'm not sure why they would expect to be listened to.
...If an artist doesn't provide either I guess I'm not sure why they would expect to be listened to.
Because that's the history of Blender, if you use it you have a say, the new shift in focus (towards a more professional application and userbase) conflicts with that and it can't be had both ways.
Ton (as he's the project director?) needs to re-address exactly what he wants from Blender, if a more professional direction is wanted then the entire development and business model has to change to cater directly to that; trying to do both just doesn't work because it alienates both sides of the fence for the same reasons, people don't feel they're getting their moneys worth.
I have to agree with Kat in that using the program should be enough to hold weight with its development, otherwise eventually everyone will feel like they are on the wrong side of the fence and those who do not have time or money to spare may very well fell bad and a lesser part of the community for not being able to give anything.
There is quite a difference, in my mind at least, between buying a DVD every 18 months for £30 and paying £120 for the same period and expecting actual code and features.
People buy the DVDs to support development on vaguely specified new features that may or may not actually happen...there is no guarantee...but if all else fails, you get a DVD at the end.
When people start paying lots of money towards an end, they are going to expect feature X to be in the next release and if its not then people will get mad.
I dont mean to be pessimistic but that's just the way people are.
Take the ZB4 release yesterday, the day wasnt even over and people were already complaining that it wasn't out yet and that was for a free upgrade.
the bounty model worked very well for the dynamic loading/freeing pledge in the bge, i should know since i actually donated money for that .. i think $50 and while that is a pittance for any paid coder, it was the volume of people who donated that made it work.
as for blender needing to decide on if it should go a more professional business route i am slightly confused. could you explain how this direction differs from the one its already going in?
It doesn't really work to have one group of users, usually in the minority (Professionals, subscribers, customers that buy into the brand) paying for another group who get exactly the same thing for free. There has to been some benefit for paying beyond being altruistic otherwise they don't see any return on their investment.
Professionals and corporations don't necessarily invest for altruistic reasons either, that's something hobbyists can afford to do as there's no inherent return required for that investment other than the kudos. It's not beside the point either that from a corporate point of view - where most 'professional' users will be from - it's arguable that "future development" isn't a tax deductible because it's not a received 'service' (although I'm sure they'll be some way to do it).
That's kind of what I mean about making the shift to targeting professional uptake, they have completely different motivation for using something that's tied to work and income. Hobbyists and amateurs don't.
Because that's the history of Blender, if you use it you have a say, the new shift in focus (towards a more professional application and userbase) conflicts with that and it can't be had both ways.
Actually it has never been the case of 'just using it having a say'. Users who contribute in terms of good feedback 'have a say'.
Ton (as he's the project director?) needs to re-address exactly what he wants from Blender, if a more professional direction is wanted then the entire development and business model has to change to cater directly to that; trying to do both just doesn't work because it alienates both sides of the fence for the same reasons, people don't feel they're getting their moneys worth.
Ton is the founder of the Blender Institute and Foundation. He doesn't have much interest in the pro market. Your argument doesn't make much sense. Have you refused to use software because it doesn't have enough copy protection? If not then the pirates are freeloading off of you. Do you avoid software that charges a different price for different segments? Does piracy in China keep you up at night?
There has to been some benefit for paying beyond being altruistic otherwise they don't see any return on their investment.
Funny I thought they were getting to direct development towards the features that they find most useful and getting those features sooner than they otherwise would have.
I find your argument rather bizarre.
Professionals and corporations don't necessarily invest for altruistic reasons either
It isn't altruism, it is enlightened self interest. Use a bit of game theory - we have 4 options 1 don't contribute others contribute; 2 don't pay others contribute; 3 contribute others don't; 4 contribute others contribute.
Option 1 - ultimately a loss since development is at a standstill
Option 2 - a net win for you, and a win for your opponent, but since this game is multiple round - tit for tat might result in option 1 and thus ulitmately a loss. Also even if the freeload strategy pays off is it the best possible outcome
Option 3 - a net win for you, but less than option 2.
Option 4 - the maximum win for both you and your opponent.
Of course we can expand the game beyond two player. Instead of pure tit for tat the strategy with non contributers is more complicated.
Note that this exact same scenario exists for pirates of software - as long as the threat of punishment for piracy is nil (which for all intents and purposes it is for the vast majority of users) then they play the freeloader roll.
Since we know that the majority of users of most software use pirated versions have you decided to switch to being a pirate too? Are you only dissuaded from piracy for fear of getting caught? If the answer is no - then obviously getting increased value out of the software is of more strategic significance to you than your worries about freeloaders.
Strategies for winning in tit for tat involve various combinations of 1) getting enough cooperating players that you find the benefit more worthwhile than your concern for freeloaders 2) punishing freeloaders in some way - you could act condescending towards them if it would make you feel better . You could make freeloaders sit at a different table during lunch. You could not do business with them. etc 3) getting a bonus benefit for contributors. - I'm willing to offer email support and some bug fixing prioritization. There is also the 'useless and cheap but nice looking bauble' - a special tshirt; a limited edition stuffed animal, a coffee mug. A DVD of a movie that the exact same thing can be downloaded off the internet Note that many of the useless but nice looking baubles often have a goal of accomplishing 2.
That's kind of what I mean about making the shift to targeting professional uptake, they have completely different motivation for using something that's tied to work and income. Hobbyists and amateurs don't.
the ROI is new features. Within 3 months or so I could get Blender developed to have the best workflow on the market for sculpting and retopology and texturing.
To me a faster workflow is worth paying for - being able to freeload at the cost of slower development is not a trade off that I find appealing.
If you bribe the guard it reduces the time all prisoners spend in jail for each prisoner that bribes the guard. Paying the guard more reduces the sentence more, and the guard treats bribes from wealthy prisoners even more favorably. Some prisoners are millionaires, others are impoverished. The guard is willing to accept any bribe amount. You are a middle income individual. Do you decide to avoid bribing the guard because your impoverished fellow prisoners will benefit as well? What if you found out a wealthy or middle income prisoner refused to pay - will that affect your pay/don't pay strategy? If your paying/not paying could likely induce others to pay/not pay does that affect your strategy?
Since your goal is 'get out of prison as soon as possible' - your optimal strategy is to always pay. If your paying influences the paying of others - it is your optimal strategy to pay and try to influence others to pay as well.
Paying for features is fine, but you just have to be careful how it is implemented so that it doesn't cause divisions in the community.
Maybe it was a bad choice of words, but things like branding people as free-loaders if they don't contribute unevens the playing field even before any suggested subscription is in place.
@ LetterRip: I'm not (I don't think anyone is) trying to undermine your efforts here, I just don't think that 'pay-to-play', in the literal sense, is the right way to develop the application (see metalliandy's replies.. I too would pay for feature inclusion/development). If you have a tonne of ideas already, you should have the freedom to implement those internally without straining the financial resources available. If that's not the case then there's something else going on here.
Just so you know, I'm not one these freeloaders we're alluded to here, nor am I a pirate, or someone that's never given back to the Blender community, so my comments shouldn't be construed as "***k that, I'm not paying for noffin".
I think a subscription type thing might be less controversial if it was run by the developer himself, rather that the Blender Foundation. Of course that would be more of an extended donation system, which encourages exciting prototypes for advanced features and not bug fixing/optimizing/small but essential additions...
In any case it sounds like the Blender Foundation has enough cash on hand already, they just need a larger developer pool to hire from. Maybe that will improve once a stable version of 2.5 is released?
metaliandy, changed to contributor and non contributor, wasn't trying to be pejorative, meant to say 'free rider' which is the term in game theory but misremembered.
kat,
I wasn't trying to say that you were I was just doing game theory of why the game theoretical optimal strategy is cooperation. I realize that folks do have some psychological blocks that make it easier to pay if they 'get something', even if that something is not of great intrinsic value itself.
Also a note, the game theoretical optimal result - while cooperation is pretty much best, the cost of contributing can be a factor. Ie if your contribution represents a significant cost to you then it is different than if it represents little or nothing. So the 'give up a coffee once a month' folks the cooperative strategy is most viable and the behaviour of non contributors can be completely ignored as irrelevant; for those where cooperating represents a significant risk (ie somewhere where annual income is on the order of 1000$ a year) then not cooperating is a valid strategy since maintaining the status quo only represents an opportunity cost, and not maintaining the status quo could be costlier than any marginal utility.
jrs100000,
that has been tried and is working kind of poorly due to lack of good management skills. The difficulty is that most people are really bad at self manamagement.
In any case it sounds like the Blender Foundation has enough cash on hand already, they just need a larger developer pool to hire from. Maybe that will improve once a stable version of 2.5 is released?
As I said I have a number highly talented developers that have Blender experience interested in working on Blender. At least two are 'brecht caliber'.
Yea i thought it was just a bad choice of words..no worries
What are the chances of these "brecht" calibre devs joining the BI seeing as Ton has funds for a couple of positions?
That would be awesome seeing as he is joining the Octane team.
Another option is for a foundation-run bidding website, where users can pledge a certain amount for a feature. This brings up the issue of lost interest, but one way is to have people actually pay this money to the foundation in order to receive 'points' they can spend on projects. This way, if a feature-request doesn't get picked up by any programmers, users can get those points back to spend on other projects. Once a project has started, those points can't be taken out.
Something like that sounds like a feasible way to run this, wouldn't you agree? This also enables many more people to get involved, as students or hobbyists will be much less able to spend a set amount every month, not having steady income. This way they can spend a small amount every now and again, when they can afford to.
Off-topic, but Blender related. Not sure how many people were aware of or used http://blenderjobs.com previously but the site was shut down recently. Long-story-short, I've taken it on-board so for those looking for or wanting to post opportunities specific to Blender 3D users (artists, engine, python etc.) the new link is http://jobs.katsbits.com.
Off-topic, but Blender related. Not sure how many people were aware of or used http://blenderjobs.com previously but the site was shut down recently. Long-story-short, I've taken it on-board so for those looking for or wanting to post opportunities specific to Blender 3D users (artists, engine, python etc.) the new link is http://jobs.katsbits.com.
I had no problems following the tutorials and getting things into Sio2. Going from Maya to Blender is no problem either. Want to go from Blender fluids to Lightwave later.
Ok Max is retarded, Its only sorta usable after 9000 addons and modifications... for modeling and uv mapping Modo is the best in my experience. Its cool to see Blender become more viable and I wish them further luck developing it!
Well I'm not going to vouch for Blender as a modeling app because the lack of n-gon support kills that part of it for me.
Its been a while sense I last used Modo, but I did not find it intuitive at all (look at Silo for an example of an intuitive modeling app). And when it comes to it vs Max, other then being able to work directly on insoline surfaces I really don't remember seeing any truly amazing modeling features that you'd think makes it so much better? Though I do like Modos UV tools more then maxs (even though they do need some addons as well such as senecas tools).
Well I'm not going to vouch for Blender as a modeling app because the lack of n-gon support kills that part of it for me.
Its been a while sense I last used Modo, but I did not find it intuitive at all (look at Silo for an example of an intuitive modeling app). And when it comes to it vs Max, other then being able to work directly on insoline surfaces I really don't remember seeing any truly amazing modeling features that you'd think makes it so much better? Though I do like Modos UV tools more then maxs (even though they do need some addons as well such as senecas tools).
I actually feel the opposite way when it comes to ngons in Blender.
Though BMesh will sort out many of the issues that the archaic mesh system currently has (enabling nice bevels etc.), i would rather not use ngons as i prefer to edit meshes into quads by hand.
May i ask why you find ngons preferable?
I like using them in some instances when doing high poly work, it can save time getting particular shapes and smoothed areas. I don't use them for finished lowpoly pieces (never do that) but while working find them faster then dealing with holes or excess tris. They're obviously not fundamentally critical in the modeling process, but I feel that the extra time that is spent working around such a limitation is simply not worth it when virtually every other comparable app is not limited by it.
Just been getting back into modeling in Blender myself. Got most of my custom setup done which I'm pretty happy with (SO much nicer than working in Maya). But yeah, it's really needing to ngon support. I don't use ngons myself in the final model either, but when you're going through cutting geometry it's a pain when the program just can't do it like normal because it can't do ngons. Instead you end up having to delete some geometry and rebuild faces that would just cut through.
If you haven't seen this video I highly recommend it. It shows how to create custom keys. The system is currently unpolished as it's more akward setting things up than it should (and I'm sure...will) be, but this tutorial goes more in depth than just 'press the key to set it as the hotkey for "x"'.
Even if subd modeling sucks in its current state, the modifiers and deformers bring it nearly up to par with max's modeling features. I just supplement it with Silo if I need to do any fine modeling.
You can do stuff like this a lot more easily than you could in Maya even. The curve beziers can be transformed, scaled, and rotated using the standard manipulators. No need to hop into a graph to modify twist:
The grease pencil is another nice addition. I use it a lot to sketch blueprints into the viewport, draw topology onto a mesh, mark up camera views to fix composition, etc.
This is a real workflow enhancement, not just another "let's copy this feature"
Yeah there's a bloke at my work who has been talking up the Grease pencil feature since it started getting shown. The demo videos i've seen have some impressive stuff in them and it looks like it could be a really useful retopo tool.
Have any of you tried the sculpting yet? Not had a go myself yet as I've been focusing on my normal modeling workflow with the new Blender. Seems like it might be pretty good and if they add in the dynamic tesselation and ptex...pretty nifty indeed.
@ Womball: with 2.5 there's an increasing interest in it for 'pro' work, still not as much as with other apps (obviously) but it's a growing market, its one of the reasons I took on blenderjobs.com so that aspect of Blenders growth is maintained.
I like using them in some instances when doing high poly work, it can save time getting particular shapes and smoothed areas. I don't use them for finished lowpoly pieces (never do that) but while working find them faster then dealing with holes or excess tris. They're obviously not fundamentally critical in the modeling process, but I feel that the extra time that is spent working around such a limitation is simply not worth it when virtually every other comparable app is not limited by it.
yeah, thats one thing that is bothering me too
i would really like to be ABLE to use ngons. maybe its just me, but it would make some things alot easier
but aside that, iam really comfortable with blender. imho its more intuitive than max
I always found that blender was just a huge pain in the ass to use. I'm all for open source programs but I think they should take the proven designs of other popular software and build upon that instead of being different just because they can. I'm sorry but why do I need to hit a hotkey every time I want a selection box? Is selecting an edge > control E > 7 > DEL > click "yes" really the most efficient way to delete an edgeloop? It always seemed a bit convoluted to me.
I do however give them major props for the redesign. Having downloaded version 2.55 and seeing that the little interaction drop box had "Maya" just made my day. What do you know...I can actually maneuver around the 3D space now. And they have buttons! honest to God buttons so that I dont have to learn some crazy control + shift + something else hotkey to access the tools I want. That, and the fact that you can remap your hotkeys and have a [hopefully soon] customizable toolbar and I might just actually start working this into my workflow. :thumbup:
I always found that blender was just a huge pain in the ass to use. I'm all for open source programs but I think they should take the proven designs of other popular software and build upon that instead of being different just because they can. I'm sorry but why do I need to hit a hotkey every time I want a selection box? Is selecting an edge > control E > 7 > DEL > click "yes" really the most efficient way to delete an edgeloop? It always seemed a bit convoluted to me...
I think box selection is a hotkey because it shares functionality with paint selection, one B to go into box selection, B - B to go into paint selection.
also:
alt+select to select loops, x->edgeloop to delete it.
the ALT key combined with selection functions the same way no matter what selection mode you are in, and you can even use it to select loops and edges in the UV editor.
Replies
Awesome! Cheers
"Here's a rough guide to the effect your base mesh has on the speed of sculpting:
1. When you sculpt, it optimizes by ignoring parts of the mesh that aren't near the brush. There are two resource-intensive things that sculpting does: the actual brush calculations needed to modify the surface, and redrawing the areas that have been changed. For this optimization to work, it needs to break the mesh up into pieces. That way, just one small piece can be sculpted or redrawn, rather than millions of polygons in a detailed mesh. So the question is how is mesh split up?
2. When you multires a mesh, it gets split up into grids. You know how when you apply one level of subsurf to a model, each quad turns into four smaller quads, and each triangle turns into three smaller quads? Each of those quads counts as one grid in multires. So if your input mesh is 6 quads, like the default cube, you'll have 24 grids, whereas a monkey (500 faces), produces 1968 grids.
3. One grid is the smallest chunk of the mesh that can be sculpted on and redrawn independently. So no matter how far you subdivide, you have the same number of grids, and you have the same limit on the number of chunks the mesh can be split into for optimization.
Note that only the topology of the base mesh matters, not the shape.
As I said, this is a rough guide -- it doesn't contain any numbers! So you still have to experiment a bit to see what level of detail in the base mesh works best on your system.
-Nicholas"
I'm doing a survey for where folks would like Blender to go
http://www.polycount.com/forum/showthread.php?t=75677
Please have a look and fill it out, also feel free to ask questions in the thread to keep it from sinking into obscurity,
Tom M.
LetterRip
But I know its a beta.
I'd like to see more integrated help files within Blender. Like the Ribbon tooltips in 3DSmax or 3DCoat. Where you press a button in the preferences to say you want help tooltips. Then when you hold your mouse over any of the buttons, you get a pop with really detailed info on what it does, and an brief example of it.
I also wish the help was not so reliant on remembering keystrokes I know my brain does not remember them as well as I do a button that is located somewhere logically in the ui.
Im not too sure how the subscription suggestion will pan out through...I can see that causing major problems
And everybody, even nonsubscribers would get these features due to the very nature of GPL.
Im all for Blender improvements, but here are a few problems that i could see happening
I think a fairer funding model would be that the Devs would post a list of projects that there is interest in developing simultaneously and users donate to fund that feature or users could top up the wages of a Dev who is working on a feature currently for the foundation or under supervision by the foundation to ensure completion.
*EDIT*
I just read the Sunday Dev meeting notes summary on BlenderNation and it says that funding is not as much of a problem as actually finding people to do the work.
since I didn't upgrade to the yearly account you for surveys I closed the survey earlier to make sure I didn't get overage charges.
Regarding Tons statement about difficulty finding developers - I tend to disagree. He wants coders who are brilliant self managers and who work for peanuts and have a deep knowledge of every aspect of Blender. However, if you have a good manager - then the self management issue is less important. If you are willing to allow the developer to focus on their core competencies and grow into other areas - then you lose the need for having an uber developer that knows ever Blender nook and crannie. If you get adequate funding together, then the work for peanuts is on longer an issue. I have 10 or so developers willing to work for me, all of them with Blender commit rights, and could double or triple that amount if funding were in place. Most of these are Blender developers who have had to move on to other things because they having a life and working full time for free are not very compatible. Others are Blender devs who are students who are looking to enter the work force in the near future.
Metaliandi,
As to how it will work out - I don't think the risk of disillusionment is that high. The point of having the user identify multiple areas they would like worked on is to increase the odds of their desires corresponding with a large enough group that it increases their odds of getting something they want.
Based on the results on my survey the top wish list items cluster heavily based on what industry the individual works in.
Also for these subscriptions they would be on the order of 120$ a year.
Also there are items like - save confirmation for quiting in all contexts; or selective bevel - that a huge number of users have wanted forever but that are not priorities with any particular developer.
I think if I show major progress towards core issues that have been a priority for many users 'forever', as well as showing rapid progress on areas that will make a major difference on workflow and productivity, that the majority of users will feel that their money is well spent.
The bounty model works absolutely horrendously. It has a number of substantial flaws that make it unworkable. 1) The typical artist for whatever reason places the value of any particular feature at between 5 and 20$ or rarely for something that is a years worth of developer time they might go as high as 100$. The actual dev cost of any particular feature is usually 1000$ for a simple feature to 10,000$ for a moderately complex feature. 2) The 'cost' for both the developer and artist is high in terms of amount of time invested versus features gained 3) the developer essentially has to complete the feature - then hope that the funding will come through. Funders tend to be fairly unreliable for post hock payment (lost interest, misestimated cash flow, never intended to pay, or a whole host of other reasons). 4) It encourages developers to game the system - withholding features and improvements since they are a potential revenue stream.
As to 'contributors being put on a pedestal' they already are. They get mentioned on web pages as donors; they get their names in movie credits, etc.
Also for those who aren't interested in subscribing there is always art for code. Of course good feedback on tools or clever ideas also get implemented. Artists who show great artwork done with the tools and point out limitations will also get listened to. Those who contribute in some way - either feedback or funding are listened to.
If an artist doesn't provide either I guess I'm not sure why they would expect to be listened to.
Ton (as he's the project director?) needs to re-address exactly what he wants from Blender, if a more professional direction is wanted then the entire development and business model has to change to cater directly to that; trying to do both just doesn't work because it alienates both sides of the fence for the same reasons, people don't feel they're getting their moneys worth.
There is quite a difference, in my mind at least, between buying a DVD every 18 months for £30 and paying £120 for the same period and expecting actual code and features.
People buy the DVDs to support development on vaguely specified new features that may or may not actually happen...there is no guarantee...but if all else fails, you get a DVD at the end.
When people start paying lots of money towards an end, they are going to expect feature X to be in the next release and if its not then people will get mad.
I dont mean to be pessimistic but that's just the way people are.
Take the ZB4 release yesterday, the day wasnt even over and people were already complaining that it wasn't out yet and that was for a free upgrade.
as for blender needing to decide on if it should go a more professional business route i am slightly confused. could you explain how this direction differs from the one its already going in?
Professionals and corporations don't necessarily invest for altruistic reasons either, that's something hobbyists can afford to do as there's no inherent return required for that investment other than the kudos. It's not beside the point either that from a corporate point of view - where most 'professional' users will be from - it's arguable that "future development" isn't a tax deductible because it's not a received 'service' (although I'm sure they'll be some way to do it).
That's kind of what I mean about making the shift to targeting professional uptake, they have completely different motivation for using something that's tied to work and income. Hobbyists and amateurs don't.
Actually it has never been the case of 'just using it having a say'. Users who contribute in terms of good feedback 'have a say'.
Ton is the founder of the Blender Institute and Foundation. He doesn't have much interest in the pro market. Your argument doesn't make much sense. Have you refused to use software because it doesn't have enough copy protection? If not then the pirates are freeloading off of you. Do you avoid software that charges a different price for different segments? Does piracy in China keep you up at night?
Funny I thought they were getting to direct development towards the features that they find most useful and getting those features sooner than they otherwise would have.
I find your argument rather bizarre.
It isn't altruism, it is enlightened self interest. Use a bit of game theory - we have 4 options 1 don't contribute others contribute; 2 don't pay others contribute; 3 contribute others don't; 4 contribute others contribute.
Option 1 - ultimately a loss since development is at a standstill
Option 2 - a net win for you, and a win for your opponent, but since this game is multiple round - tit for tat might result in option 1 and thus ulitmately a loss. Also even if the freeload strategy pays off is it the best possible outcome
Option 3 - a net win for you, but less than option 2.
Option 4 - the maximum win for both you and your opponent.
Of course we can expand the game beyond two player. Instead of pure tit for tat the strategy with non contributers is more complicated.
Note that this exact same scenario exists for pirates of software - as long as the threat of punishment for piracy is nil (which for all intents and purposes it is for the vast majority of users) then they play the freeloader roll.
Since we know that the majority of users of most software use pirated versions have you decided to switch to being a pirate too? Are you only dissuaded from piracy for fear of getting caught? If the answer is no - then obviously getting increased value out of the software is of more strategic significance to you than your worries about freeloaders.
Strategies for winning in tit for tat involve various combinations of 1) getting enough cooperating players that you find the benefit more worthwhile than your concern for freeloaders 2) punishing freeloaders in some way - you could act condescending towards them if it would make you feel better
the ROI is new features. Within 3 months or so I could get Blender developed to have the best workflow on the market for sculpting and retopology and texturing.
To me a faster workflow is worth paying for - being able to freeload at the cost of slower development is not a trade off that I find appealing.
If you bribe the guard it reduces the time all prisoners spend in jail for each prisoner that bribes the guard. Paying the guard more reduces the sentence more, and the guard treats bribes from wealthy prisoners even more favorably. Some prisoners are millionaires, others are impoverished. The guard is willing to accept any bribe amount. You are a middle income individual. Do you decide to avoid bribing the guard because your impoverished fellow prisoners will benefit as well? What if you found out a wealthy or middle income prisoner refused to pay - will that affect your pay/don't pay strategy? If your paying/not paying could likely induce others to pay/not pay does that affect your strategy?
Since your goal is 'get out of prison as soon as possible' - your optimal strategy is to always pay. If your paying influences the paying of others - it is your optimal strategy to pay and try to influence others to pay as well.
Maybe it was a bad choice of words, but things like branding people as free-loaders if they don't contribute unevens the playing field even before any suggested subscription is in place.
Just so you know, I'm not one these freeloaders we're alluded to here, nor am I a pirate, or someone that's never given back to the Blender community, so my comments shouldn't be construed as "***k that, I'm not paying for noffin".
In any case it sounds like the Blender Foundation has enough cash on hand already, they just need a larger developer pool to hire from. Maybe that will improve once a stable version of 2.5 is released?
kat,
I wasn't trying to say that you were
Also a note, the game theoretical optimal result - while cooperation is pretty much best, the cost of contributing can be a factor. Ie if your contribution represents a significant cost to you then it is different than if it represents little or nothing. So the 'give up a coffee once a month' folks the cooperative strategy is most viable and the behaviour of non contributors can be completely ignored as irrelevant; for those where cooperating represents a significant risk (ie somewhere where annual income is on the order of 1000$ a year) then not cooperating is a valid strategy since maintaining the status quo only represents an opportunity cost, and not maintaining the status quo could be costlier than any marginal utility.
jrs100000,
that has been tried and is working kind of poorly due to lack of good management skills. The difficulty is that most people are really bad at self manamagement.
As I said I have a number highly talented developers that have Blender experience interested in working on Blender. At least two are 'brecht caliber'.
What are the chances of these "brecht" calibre devs joining the BI seeing as Ton has funds for a couple of positions?
That would be awesome seeing as he is joining the Octane team.
Something like that sounds like a feasible way to run this, wouldn't you agree? This also enables many more people to get involved, as students or hobbyists will be much less able to spend a set amount every month, not having steady income. This way they can spend a small amount every now and again, when they can afford to.
there is planned to have an option of one time payments.
Nice one mate!
Then again, I was trying to import cinema4d models into blender. "Improbability drive" import settings.
Lamont, why Sio2 over Unity 3D? This is the first time I hear of Sio2.
Wait, what? You don't need a modifier to edit components, just convert your primitive to an editable poly and away you go.
Modo is neat, but its got alot of problems too. As a whole, It by no means 'blows max out of the water'.
Its been a while sense I last used Modo, but I did not find it intuitive at all (look at Silo for an example of an intuitive modeling app). And when it comes to it vs Max, other then being able to work directly on insoline surfaces I really don't remember seeing any truly amazing modeling features that you'd think makes it so much better? Though I do like Modos UV tools more then maxs (even though they do need some addons as well such as senecas tools).
I actually feel the opposite way when it comes to ngons in Blender.
Though BMesh will sort out many of the issues that the archaic mesh system currently has (enabling nice bevels etc.), i would rather not use ngons as i prefer to edit meshes into quads by hand.
May i ask why you find ngons preferable?
I guess for me its a case of never having them and thus not seeing why some people miss them.
Thanks
If you haven't seen this video I highly recommend it. It shows how to create custom keys. The system is currently unpolished as it's more akward setting things up than it should (and I'm sure...will) be, but this tutorial goes more in depth than just 'press the key to set it as the hotkey for "x"'.
http://www.blendercookie.com/2010/05/17/custom-hot-keys/
http://www.graphicall.org/builds/builds/showbuild.php?action=show&id=1576
Im not sure how stable they are though.
You can do stuff like this a lot more easily than you could in Maya even. The curve beziers can be transformed, scaled, and rotated using the standard manipulators. No need to hop into a graph to modify twist:
The grease pencil is another nice addition. I use it a lot to sketch blueprints into the viewport, draw topology onto a mesh, mark up camera views to fix composition, etc.
This is a real workflow enhancement, not just another "let's copy this feature"
yeah, thats one thing that is bothering me too
i would really like to be ABLE to use ngons. maybe its just me, but it would make some things alot easier
but aside that, iam really comfortable with blender. imho its more intuitive than max
http://www.blender.org/development/release-logs/blender-256-beta/
I do however give them major props for the redesign. Having downloaded version 2.55 and seeing that the little interaction drop box had "Maya" just made my day. What do you know...I can actually maneuver around the 3D space now. And they have buttons! honest to God buttons so that I dont have to learn some crazy control + shift + something else hotkey to access the tools I want. That, and the fact that you can remap your hotkeys and have a [hopefully soon] customizable toolbar and I might just actually start working this into my workflow. :thumbup:
I think box selection is a hotkey because it shares functionality with paint selection, one B to go into box selection, B - B to go into paint selection.
also:
alt+select to select loops, x->edgeloop to delete it.
the ALT key combined with selection functions the same way no matter what selection mode you are in, and you can even use it to select loops and edges in the UV editor.