Jump to content
dave2017

3D graphics - is it possible?

Recommended Posts

While trying to get to grips with isometric drawings and modelling in AD I wondered if it would be possible to extend AD to permit 3D modelling. Perhaps not - as the package may not have been designed with 3D in mind, but it just struck me that if an object model is going to be built up - possibly fairly tedious by the designer/artist - that it might not be much harder to generate a 3D model, then anyone who only wanted 2D could render the model in 2D. That way both current and future users would be catered for.

Share this post


Link to post
Share on other sites

Hi dave2017,

No, it's not feasible. Affinity Designer is strictly a 2D design/illustration software. Isometric drawings are simply 2d representations of 3d objects in a 2d plane, there's no true 3D depth/volume (or z axis) involved. A 3d modelling software implies a different architecture/heavy changes to the current app at several levels. There's no plans to move in that direction.

Share this post


Link to post
Share on other sites

Hi MEB,

I think that's the state of affairs now, but would it not be possible for the systems implementers to put in some 3D modelling? Much of the User Interface components are already there, and there are clearly mathematical models behind the scenes. Some transformations in 3D are not a great deal harder than doing them in 3D - "just" needs a 3D representation  for points defining - er points, and edges and you're away. I know it gets a bit more complicated than that if one starts putting in light, reflectance, shading etc., but I just wondered if there's any chance of your implementers doing some of this. Otherwise users might start out with a tool like AD, then switch to a differernt 3D tool if they discover that (a) 2D is limited, and (b) that they can do what they want more easily in a 3D package.

Share this post


Link to post
Share on other sites
6 hours ago, dave2017 said:

[…] I just wondered if there's any chance of your implementers doing some of this. Otherwise users might start out with a tool like AD, then switch to a differernt 3D tool if they discover that (a) 2D is limited, and (b) that they can do what they want more easily in a 3D package.

Wouldn’t implementing some halfhearted 3D functionality cause exactly that problem? People would start with AD thinking they can do 3D modelling and then discover that a lot of necessary things are missing and would be more disappointed than if they had known beforehand that AD is not a 3D tool.

Share this post


Link to post
Share on other sites
8 hours ago, VIPStephan said:

Wouldn’t implementing some halfhearted 3D functionality cause exactly that problem? People would start with AD thinking they can do 3D modelling and then discover that a lot of necessary things are missing and would be more disappointed than if they had known beforehand that AD is not a 3D tool.

I wasn't going to suggest that "halfhearted 3D functionality" would be the outcome, but rather sufficiently good 3D functionality that it would be easy enough to implement. For example, somewhere there is a demonstration of making a can using AD - takes about 30 minutes. In 3D some of the steps would be similar, while some could actually be quicker. Make the outline of the can - then use a surface of revolution  (https://en.wikipedia.org/wiki/Surface_of_revolution) about the central line to produce the cylinder, perhaps including the base. The top of the can could be made from a similar approach to generate a circular top - and a ring pull could be added - something not shown in a 2D representation. These two objects could then be joined by a algebraic approach. Then the branding material could be added as a rectangular strip which would be wrapped round the cylinder.

I think from the designer's point of view this wouldn't be a lot harder than using AD or similar tools, and might actually be a lot easier. Once the basic shape/volume has been set up, then it could be rotated into different positions, and lighting effects added.

What would make things a bit harder would be if the ring pull on the can were pulled open or off, and then the interior of the can needed to be modelled for lighting - and also if there were any liquid still in the can. Let's not make things too complicated, though!

Would it be difficult or impossible to implement? Well - I don't think it would be impossible, as there are already tools which do this. It might be a bit harder than implementing a 2D package, but it might not as it would be a generalisation of 2D - or rather think of a 2D package as being a constrained version of a 3D one. A 3D package should be more versatile.

Would it be computationally expensive - and therefore present performance limitations which would inhibit the end user? Maybe it would be slightly more computationally expensive than a 2D system, but I suspect that it wouldn't be greatly so. As modern computers are becoming more powerful year on year I don't think that need be a huge barrier. If 3D performance scaled up by a significant factor compared with 2D performance, then this would be an issue - but I'm hoping and expecting that this would not be the case.

To some extent I'm flying a kite here - and it may be that full 3D functionality is not a realistic prospect right now in a piece of mass consumer oriented software. To get a better view of this would probably require input from designers who have used 3D and 2D packages for similar projects. There may not be so many people who have the experience to give a reliable opinion on both approaches. The number of systems implementors who have done this kind of project or could tackle such projects might be rather small. That could present difficulties and barriers for future development by some companies.

Perhaps I'd better stop, before I mention 4D and the introduction of a time axis!

 

Share this post


Link to post
Share on other sites
12 minutes ago, Fatih19 said:

Cant you just use Blender and call it a day?

Your response came in while I was writing my reply to earlier posts above. Do you have significant experience of Blender? Can it do most of what is required already - and sufficiently easily for the user? If you take my can example one aspect which might be better handled right now in AD is the design of the rectangular branding "paper" strip to wrap round the can. Here I am ignorant - I don't know if it's possible to generate such a rectangular 2D object and input it into Blender, and then perform the wrap round the can, or alternatively whether it would be easy to do that "paper" strip in Blender. A "fun" thing might be to put a realistic representation of a 3D scene on a flat 2D paper "object", then wrap it round the can object. I wonder what it would look like. Then toss the can up into the air (4D - space+time) and see what comes out!

It may be that using existing tools would be more useful than trying to make new ones - but then designers and end users would need to be aware of the merits of each and would need expertise in several - which probably isn't much different from what some users do right now.

So in answer to your question I'd suggest that using Blender (or similar) might well be a way to go - but the amount of learning users would have to do to learn one or more systems may be a significant barrier at least in non commercial or small scale environments. In cooperative team environments the expertise could be divided/shared between different users.

I have only used Blender for some very small actions - it is part of my tool kit, but I've not tried anything significant using it. I used to know people who worked on the implementation and use of 3D design tools. I could ask some of them about this area - but things may have moved on. OTOH, since a lot of the work was theoretical, it could simply be that feasible and practical software is only now beginning to catch up with the theory.

 

Share this post


Link to post
Share on other sites

Do you really want to add 3D functionality to a 2D illustration software that is still in version 1.6 and developed by a small team? We doesn't even have warp/distort, pure vector brush, and a knife tool. Do you really want that?

Share this post


Link to post
Share on other sites
27 minutes ago, Fatih19 said:

Do you really want to add 3D functionality to a 2D illustration software that is still in version 1.6 and developed by a small team? We doesn't even have warp/distort, pure vector brush, and a knife tool. Do you really want that?

Fair comment. Are you part of that team or close to it?

It could well be that developing such an enhanced tool is outside the realms of possiblity for a small development and implementation team.

I was just flying a kite, as I mentioned earlier, and it did occur to me that some features of AD which are already implemented could be used in a 3D package. Of course it may also be that the representations of software components within a 2D package are not suitable for use in a 3D one - for example the representation of objects which can be combined and interact as 2D layers in a 2D package - might just not be compatible. I'm thinking of the operations like Boolean Add, Subtract etc. - which might have to be implemented using 3D "objects" from the ground up.

The benefits of working in 3D from a user's point of view - assuming the object representations are easy enough to use - is that the effort of positioning for a 2D rendering are done in the software package, and can be done quickly. Otherwise designers can (or have to) imagine what the 3D representation would like, or even take a representation such as a photo, and they then have to do the work to get that representation into a 2D design which is useful to them, or in a commercial world, to their clients.

The software development aspects are not inconsiderable - and you are right to query adding in features, or my suggestion of a significant strategic change of direction.

Share this post


Link to post
Share on other sites
Just now, dave2017 said:

Fair comment. Are you part of that team or close to it?

It could well be that developing such an enhanced tool is outside the realms of possiblity for a small development and implementation team.

I was just flying a kite, as I mentioned earlier, and it did occur to me that some features of AD which are already implemented could be used in a 3D package. Of course it may also be that the representations of software components within a 2D package are not suitable for use in a 3D one - for example the representation of objects which can be combined and interact as 2D layers in a 2D package - might just not be compatible. I'm thinking of the operations like Boolean Add, Subtract etc. - which might have to be implemented using 3D "objects" from the ground up.

The benefits of working in 3D from a user's point of view - assuming the object representations are easy enough to use - is that the effort of positioning for a 2D rendering are done in the software package, and can be done quickly. Otherwise designers can (or have to) imagine what the 3D representation would like, or even take a representation such as a photo, and they then have to do the work to get that representation into a 2D design which is usefult to them, or in a commercial world, to their clients.

The software development aspects are not inconsiderable - and you are right to query adding in features, or my suggestion of a significant strategic change of direction.

3D functionality is fine, but for AD to implement it? A long long time. And no, i'm not a part of the development team.

Share this post


Link to post
Share on other sites
4 minutes ago, Fatih19 said:

3D functionality is fine, but for AD to implement it? A long long time. And no, i'm not a part of the development team.

Maybe the "solution" is to use AD in conjunction with other tools then. Export and some integration with other (perhaps open source) packages may be a way to go. Interface formats between different packages might be important.

I was surprised yesterday to discover some different packages which would take SVG input generated by AD - and to realise that most modern browsers can render SVG files. It may be that data interchange between 2D (e.g AD) software and 3D software - could be 2 way - may be the way to go.

Share this post


Link to post
Share on other sites

Since we are not talking about animation but still images it would make more sense to go the other direction. Create your scene in AD and export it into your 3D program as a background image and place your 3D object(s). Once you have it looking the way you want, render the still image.

Share this post


Link to post
Share on other sites
2 hours ago, dave2017 said:

I wasn't going to suggest that "halfhearted 3D functionality" would be the outcome, but rather sufficiently good 3D functionality that it would be easy enough to implement.

But what is “sufficiently good” and when is it reached? My point is that either you implement it properly with all the consequences or not at all. If you start with simple 3D functionality but leave out related things (e. g. lighting) then people are gonna be disappointed and say AD is just half done.

I think 3D is something for a separate application (as there are already) – it could, of course, interact with AD’s current functionality (much like AD can interact with AP already, and both of them will be able to interact with APub in future) but should probably not be implemented as part of AD because that would open a whole nother can of worms for the developers (I’ve said the same thing about HTML/CSS export and prototyping tools).

Share this post


Link to post
Share on other sites
On 18/04/2018 at 5:54 AM, Fatih19 said:

Cant you just use Blender and call it a day?

I have looked into this, and it now appears to me that it makes much more sense to use Blender, which is now at least relatively mature, and does handle the 3D side of things quite well. It also has had the benefit of many years of work. There would be merits in exporting some 2D files to Blender, then doing some work in 3D within that, and then if needed exporting 2D renderings of work back to either APhoto or ADesigner for futher work.

 For animation work there may be other usable possibilities - maybe Motion but it's not the only tool.

Some people might want to just work in Blender, while others might be happy use several tools and have files exchanged between them to get a (possibly) better result.

Share this post


Link to post
Share on other sites

Hey Dave, you can export svgs out of Blender into Designer if you're looking for a 3d to 2d pipeline. I've used it with illustrator and it works pretty well. You do have to know how to get around in Blender though. Here's a couple of links. The second link is a step by step tutorial.

 

https://inkscapetutorials.wordpress.com/2015/01/07/blender-adds-freestyle-svg-export-to-create-inkscape-editable-3d-lineart/

http://goinkscape.com/use-blender-freestyle-to-export-svg-artwork/

Share this post


Link to post
Share on other sites
2 hours ago, retrograde said:

Hey Dave, you can export svgs out of Blender into Designer if you're looking for a 3d to 2d pipeline. I've used it with illustrator and it works pretty well. You do have to know how to get around in Blender though. Here's a couple of links. The second link is a step by step tutorial.

 

https://inkscapetutorials.wordpress.com/2015/01/07/blender-adds-freestyle-svg-export-to-create-inkscape-editable-3d-lineart/

http://goinkscape.com/use-blender-freestyle-to-export-svg-artwork/

Thanks for that. That'll work for some images, but it doesn't look as though it'll work for all. I was thinking of generating models in Blender, then putting texture on them - possibly photos (AP or other) or 2D from Affinity Designer and then being able to manipulate the 3D textured object in Blender, then to output a 2D version which could be tidied up in another 2D package. If the output from Blender could be scalable (SVG) and textured that'd be great, otherwise I guess that raster images would have to be used.

Depends what one wants to do I suppose. It might be possible to do almost everything within Blender, but personally I'm still finding that rather a hard system to get to grips with. A great thing about the Affinity packages, particularly Photo, is that there are so many tutorials (video or text) to show how to get things done.  There are some for Blender, and the ones I've seen are good, but I'm not sure - having only sampled a few - whether they are comprehensive enough to cover most ot the things which are possible, or which one might want to try to do. As you say you "have to know how to get around in Blender".

Share this post


Link to post
Share on other sites
On 18/04/2018 at 7:21 AM, InfoCentral said:

Since we are not talking about animation but still images it would make more sense to go the other direction. Create your scene in AD and export it into your 3D program as a background image and place your 3D object(s). Once you have it looking the way you want, render the still image.

I'm not sure that animation is ruled out. How would you do that? For a one off still image the approach you suggested would probably work, though there might be iterative steps if things didn't work out the first time - i.e. backtracking and repeating the export-import step until things looked OK.

I think you are suggesting that "static" scenes would be best/easiest to do in a 2D package, then 3D objects could move relative to that scene/background - and to do that in a 3D program. That'd work as long as lighting and viewpoints didn't change.

For a cartoon like character "object" - person or animal, wouldn't that be better done in the 3D package? One problem I foresee is that texture might not be handled so well in a 3D package, so exporting back to 2D packages might permit improvements there. For video (animation) one might hope that the use of keyframes would reduce the amount of work needed to get acceptable effects.

Share this post


Link to post
Share on other sites
1 hour ago, dave2017 said:

There are some for Blender, and the ones I've seen are good, but I'm not sure - having only sampled a few - whether they are comprehensive enough to cover most ot the things which are possible, or which one might want to try to do. As you say you "have to know how to get around in Blender".

"Some for Blender"? That is an understatement if I ever read one! The number of good Blender tutorials number in the zillions by now. The amount of Affinity tutorials pale in comparison.

Trouble is, Blender is far broader and deeper in scope than both Designer and Photo combined. So you will have to pick your battles. Start out with a basic course, and then choose a path: character modeling and rigging, rendering, product modeling and rendering, lighting, animation, visual effects, architectural modeling and rendering, texturing, particle effects, etcetera, etcetera, and so on.

Share this post


Link to post
Share on other sites
25 minutes ago, Medical Officer Bones said:

"Some for Blender"? That is an understatement if I ever read one! The number of good Blender tutorials number in the zillions by now. The amount of Affinity tutorials pale in comparison.

Trouble is, Blender is far broader and deeper in scope than both Designer and Photo combined. So you will have to pick your battles. Start out with a basic course, and then choose a path: character modeling and rigging, rendering, product modeling and rendering, lighting, animation, visual effects, architectural modeling and rendering, texturing, particle effects, etcetera, etcetera, and so on.

 

True that, Blender by far dominates the youtubes. More added daily, probably hourly.

Share this post


Link to post
Share on other sites
1 hour ago, dave2017 said:

Thanks for that. That'll work for some images, but it doesn't look as though it'll work for all. I was thinking of generating models in Blender, then putting texture on them - possibly photos (AP or other) or 2D from Affinity Designer and then being able to manipulate the 3D textured object in Blender, then to output a 2D version which could be tidied up in another 2D package. If the output from Blender could be scalable (SVG) and textured that'd be great, otherwise I guess that raster images would have to be used.

Depends what one wants to do I suppose. It might be possible to do almost everything within Blender, but personally I'm still finding that rather a hard system to get to grips with. A great thing about the Affinity packages, particularly Photo, is that there are so many tutorials (video or text) to show how to get things done.  There are some for Blender, and the ones I've seen are good, but I'm not sure - having only sampled a few - whether they are comprehensive enough to cover most ot the things which are possible, or which one might want to try to do. As you say you "have to know how to get around in Blender".

 

Do you have an example of what your wanting to do?

 

Although the export from Blender could be scalable svgs, I'm guessing any bitmapped texture applied in Blender would not export as svg, if at all... and if it did the result would probably be less than ideal. Again, it would be good to see an example of what you're trying to create. Some people here might have a suggestion that may prove to be very do-able or easier than you think.

Share this post


Link to post
Share on other sites

First:

As some have already said, I absolutely agree that it makes no sense to add 3D-capabilities to Affinity Designer. I mean it would be different If we would talk about the ability in AD to have possibilities to draw complex and CORRECT(!!!) 3D-Perspective. Someone in this Forum mentioned a perspective drawing technique called "Circle of View". If there would be a way Affinity Designer could add Tools for that kind of correct 3D-perspective drawing, then I'm totally into it. What I mean is that people who are really interested in 3D-perspective drawing will find out sooner or later, that the common knowledge of perspective drawing has some major faults in it.

 

Secondly:

If I want to make 3D-Object/Modeling, then I'm definitely using Blender! If I have a 2D-Text made in Affinity Designer and I want to make it 3D, then I'm just exporting the Text in svg for Blender. If there would be a clean and better way to export and import Scalable vector graphics from Affinity Designer to Blender and back, that would be something I and many others would totally love. This will be in future anyway better because the Blender Foundation is currently working on Blender 2.8. I think there is a good reason Serif is keeping Affinity Designer mainly as a 2D-Tool and I'm happy they stick to it, because they know it would be a huge waste of time and money to add 3D-functionalities. We should also keep in mind that the creative industry is also working on all fronts to develop kind of new standard graphics formats (e.g. overhaul for formats like Raw, Vectors, jpeg and so on), which is a win-win for everyone! It would not be smart to add something to AD that might be anyway obsolete and useless in a few years, because other softwares (like Blender) are superior in what they can offer. Affinity Designer would lose its identity.

 

You see, It is not only the subscription model why people are canceling Adobe's creative cloud. It is also because more and more people do not recognize Photoshop (as one example) anymore. 5-10 Years ago, creative people would know what Photoshops Identity as a product was. Today Photoshop is more than Photoshop. It is so jam-full of functionalities (it can also do 3D, which by the way do not use it! It is just a wast of time to use Photoshops 3D-tools) that even professionals today only know a fraction of what Photoshop is capable of. And that is for me the reason Adobe products are not attractive to me anymore. Although I still have to use InDesign, Illustrator and Photoshop at work (unfortunately), which is a pain. Over the past years Adobe just lost its identity hard!

 

That is something I hope Serif will never do. Until now Serif has proven this to me very well. They show me products with a clear identity. I know what it can do and what it can not do. It does not brag about capabilities other softwares can do better. It does show their strong capabilities so I know the main purpose for it. That is why I'm more and more using Serif products.

 

So please Serif, do not add something that would be waste of resource^^ 

Share this post


Link to post
Share on other sites
10 hours ago, Medical Officer Bones said:

"Some for Blender"? That is an understatement if I ever read one! The number of good Blender tutorials number in the zillions by now. The amount of Affinity tutorials pale in comparison.

Trouble is, Blender is far broader and deeper in scope than both Designer and Photo combined. So you will have to pick your battles. Start out with a basic course, and then choose a path: character modeling and rigging, rendering, product modeling and rendering, lighting, animation, visual effects, architectural modeling and rendering, texturing, particle effects, etcetera, etcetera, and so on.

It's interesting to read about Blender here. I didn't realise there were so many tutorials, and indeed I haven't found too many yet – but maybe I've not been looking in the right places. I do, however, now agree that there's no point in Affinity Designer trying to do 3D in competition with an established program with many years of development.

I only bought AD because I thought it would be good, and also useful, having bought AP and also the iPad version of AP. I have used it, but really only to annotate photos, rather than anything tricky or creative. The notion of the 3D came after I tried to make an example using isometric layouts.

Obviously it's possible to do isometric models in Blender, but with the merit of being able to rotate and manipulate the models in more sophisticated ways. However Blender can be (IMO) a bit confusing at first, and the interface and features take some while to get to know. I haven't managed that yet!

I am probably not alone in coming to AD via this route, though obviously others come from a creative arts background, and have experience in graphics art or animation. I have many years of experience of computing and software, but virtually no experience of graphics, illustration or animation. I did get involved in video recording and editing a few years ago – more or less by accident, as I was asked to record a concert. I was happy to record audio, but whilst negotiating I was asked “how will you do the video?” - which rather threw me. In the end I did that, though it wasn't perfect, but I did produce a DVD as well as audio CDs. In doing that I ended up learning how to use video tools, including Quicktime, iMovie, Premiere Elements, followed by FCPX which is now (sometimes) my preferred tool. I also checked out a few others, such as Lightworks, Da Vinci Resolve and Blender. I found Blender was good for specific tasks such as splitting and joining sequences of images – and that has been useful for one of my cameras which generates MP4 files in some modes, rather than groups of jpegs or raw files.

I suspect that there will be quite a market for AD for users like me, who have limited advanced experience of graphics art and tools. One helpful tutorial/video I found used AD with Motion and FCPX. If users realise that it's helpful to use a toolbox of different tools, and exchange data between them to get a desired effect, that could be very helpful, rather than expecting to do all the work in AD. I think there are a number of other helpful videos showing how to use AD with other tools, including Blender.

Share this post


Link to post
Share on other sites
On 2018-04-20 at 10:57 AM, dave2017 said:

It's interesting to read about Blender here. I didn't realise there were so many tutorials, and indeed I haven't found too many yet – but maybe I've not been looking in the right places. I do, however, now agree that there's no point in Affinity Designer trying to do 3D in competition with an established program with many years of development.

I only bought AD because I thought it would be good, and also useful, having bought AP and also the iPad version of AP. I have used it, but really only to annotate photos, rather than anything tricky or creative. The notion of the 3D came after I tried to make an example using isometric layouts.

Obviously it's possible to do isometric models in Blender, but with the merit of being able to rotate and manipulate the models in more sophisticated ways. However Blender can be (IMO) a bit confusing at first, and the interface and features take some while to get to know. I haven't managed that yet!

I am probably not alone in coming to AD via this route, though obviously others come from a creative arts background, and have experience in graphics art or animation. I have many years of experience of computing and software, but virtually no experience of graphics, illustration or animation. I did get involved in video recording and editing a few years ago – more or less by accident, as I was asked to record a concert. I was happy to record audio, but whilst negotiating I was asked “how will you do the video?” - which rather threw me. In the end I did that, though it wasn't perfect, but I did produce a DVD as well as audio CDs. In doing that I ended up learning how to use video tools, including Quicktime, iMovie, Premiere Elements, followed by FCPX which is now (sometimes) my preferred tool. I also checked out a few others, such as Lightworks, Da Vinci Resolve and Blender. I found Blender was good for specific tasks such as splitting and joining sequences of images – and that has been useful for one of my cameras which generates MP4 files in some modes, rather than groups of jpegs or raw files.

I suspect that there will be quite a market for AD for users like me, who have limited advanced experience of graphics art and tools. One helpful tutorial/video I found used AD with Motion and FCPX. If users realise that it's helpful to use a toolbox of different tools, and exchange data between them to get a desired effect, that could be very helpful, rather than expecting to do all the work in AD. I think there are a number of other helpful videos showing how to use AD with other tools, including Blender.

 

 

Hey @dave2017 here's a link to some beginner Blender sites. Lot's of good stuff on all of these sites for anyone looking to getting into this powerful 3d app.

 

https://opensource.com/article/18/4/5-best-blender-video-tutorials-beginners

Share this post


Link to post
Share on other sites

For basic 3D modelling I recommend vectary.com - for quick doodling is perfect, very easy to get grasp of it and with bit of patience you can create pretty advanced designs. And best of all it imoprts svg's which then can be modeled in 3D. Really great tool . But it's online only.

 

Definitely I would pay money for something like this, but as desktop app with few more features (lightning control, some texturing, etc.) 

 

So, Affinity, maybe there's something to do here? You guys have done great job with simplyfying advanced drawing/photo software, so why not 3D next?

 

 

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×

Important Information

These are the Terms of Use you will be asked to agree to if you join the forum. | Privacy Policy | Guidelines | We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.