Hello everyone!

I want to ask the community about shader editor, that is now on the roadmap.
I believe that its cost is near like "Hardware instancing" request, which is too expensive for one person to fund.
Here are a lot of binary and source license owners on the forum, do you need that feature? I think that having that feature can attract a lot of new members and customers, which is good both for community and for Greg.

So what i want to say in this topic? I want to make some kind of croudfunding for shader editor (50-100$ per person), so i want to find people who will participate If we find 15-20 people who agrees to donate such sum, we will get approx. 2000$. As for me i'm ready to pay 100$.

So what do you think?

Good idea. I'm in.

It's already near the top of the roadmap.

I don't think we need to fund it, I think we just need to convince Esenthel that the shader editor has to stay at the top (and not be moved down for other, smaller features). I'm pretty sure it will become a priority for him after he finished his current work.

Because, honestly: Yes, I would really like to have a shader editor!
Speaking of it, I am confused by the explanation on the roadmap. Esenthel: Will you just allow us to have more control over the current shaders, or is it to create shaders easily from scratch? I really hope it's the last one.

(03-05-2015 10:15 AM)Tottel Wrote:  It's already near the top of the roadmap.

It is said that:
"There is no timeframe or order of development provided."

So no matter if it's in top or in the bottom. That's why i created this post

Actually, the features closest to the top get developed sooner. He just means that things can be added as top priority all the time (which also happens).

I believe It would be more advantageous to integrate something like substance designer into EE - I can't imagine a custom shader editor offering the same kind of range for procedural materials/textures and also be flexible enough for all kinds of crazy shaders people are making, unless you put an awful lot of thought and effort into it, otherwise many would be limited by the nodes/blocks offered and would need to develop shaders manually anyway. If we are talking about a node/blocks kind of editor or even a custom language, then this feature is potentially a much bigger time/effort sinkhole than any other on the roadmap.

I personally would lean more towards just being able to edit shader sources in EEs code editor, with compile errors and line numbers shown. Something like this http://glslsandbox.com/e#23080.2 or https://www.shadertoy.com/view/lsj3zy would be perfect for me and should be relatively easy to implement (and the 'gallery' button could be 'shader store'). And as is evident on those two example sites, we would all benefit if we all shared our shaders more freely. We could come to a point where the majority of users who has the need for a custom shader wouldn't even need to make one because it would already be bundled or available for download/buy.

Hello!

I'd like to do a Shader Editor too, however that's a very big task to do, and would take at least a month, maybe 2, would require lot of effort, during the development, and after, to make sure it works correctly on both DirectX HLSL and GLSL (especially since Android OpenGL shader compilers are very buggy).
I already have great plans for the Editor, I have a system designed in my head, that would allow for great flexibility.
For example you could have an existing shader, but no knowledge about it, and do just:
1. write one function like
void ModifyTexUV()
{
TexUV+=xx..;
}
2. and then in a shader/effect editor just stack this 'ModifyTexUV' function together with existing functions.

What I'm saying is you'd be able to write just tiny "modifier" functions, and then mix them together for complex effects, without having to write everything from scratch.
Minimum shader knowledge required, but great results would be possible.

(03-08-2015 11:24 PM)Esenthel Wrote:  [...] and would need at least 3,000$ to cover the development cost. [...]

I was under the impression that we subscribed for a monthly/yearly fee to cover your development costs.

(03-09-2015 04:57 AM)aceio76 Wrote:  

(03-08-2015 11:24 PM)Esenthel Wrote:  However my concern right now is I'm not sure 3,000 would be able to be gathered, since HW Instancing is still in progress.

I'm not sure if there will be more contributors to HW Instancing. There was a small group that once self-initiated a monthly contribution movement but that stopped when that group was *ahem* "insulted" for their efforts. Plus, it just doesn't seem right that the same users that already subscribe are going to pay extra 'contributions' to further the engine that they are already paying subscription to. It's cyclic redundancy, robbing peter to pay paul, so to speak, and when more new users jump on-board with esenthel because of the new feature that was made possible because of the contributions, it's not the contributing users that truly benefit twice-fold (since they are paying twice).

One fair way I can think of is if current users, particularly paying users, be allowed to 'invest' that amount you seek, and when the same amount is reached through new subscribing users, the money is returned to all investing users. This way, you reduce your risk of taking the time to build something that might not be inviting enough to new users and at the same time make paying users request only credible/worthwhile feature requests that may help grow the engine's user base. But such system is difficult to build and/or track, but you can see the intent in this poor attempt at an offer of a solution.

or a % of the contribution could be an idea, but again tracking this would be a pain in the a** considering you would atleast yearly have to download transaction history and filter for contribution parts only.

Personally the only thing I don't really like about it is that I've contributed a huge amount of the total for HWI and it saddens me to see it has so little traction from when I stopped. Thats to say I still donate money towards it when I can afford to separate some money beyond my living costs, because amongst HWI and advanced occlusion culling and shader editor they are the features I really want in the engine at this moment, and yes it could be done by us source licensees however I feel the "feeling" of sharing that implementation might not be all that big, but in another perspective it could be considered a type of donation on its own, in terms of labour contribution, then again I think most if not all source licensees only have source because they need something specifically for their game(note, I am not saying all as definitive conclusion).

I guess I've covered the pessimist, the cynical and the optimist point of views in my post. ; )

(03-09-2015 08:15 AM)Tottel Wrote:  

(03-08-2015 11:24 PM)Esenthel Wrote:  [...] and would need at least 3,000$ to cover the development cost. [...]

I was under the impression that we subscribed for a monthly/yearly fee to cover your development costs.

I think many sort of did, but the contribution list itself was to add it as highest priority, when it reached max it would be put to highest priority from what I remember Esenthel said.

(03-08-2015 11:24 PM)Esenthel Wrote:  I already have great plans for the Editor, I have a system designed in my head, that would allow for great flexibility.
For example you could have an existing shader, but no knowledge about it, and do just:
1. write one function like
void ModifyTexUV()
{
TexUV+=xx..;
}
2. and then in a shader/effect editor just stack this 'ModifyTexUV' function together with existing functions.

What I'm saying is you'd be able to write just tiny "modifier" functions, and then mix them together for complex effects, without having to write everything from scratch.
Minimum shader knowledge required, but great results would be possible.

- It's a shame you have the system designed in your head so we can't read through the details and perhaps get more excited about it...



I'm not against such an editor, I see many benefits for many people in it, the problem for me is that EE1.0 had a system in place (was ok but could be streamlined more) which the 2.0 update left in a kind of limbo state - it wasn't removed, but it didn't exactly work either (without workarounds, correct me if I'm wrong here please, and if anything changed since then). I'm OK with the shader editor being considered a luxury feature, so donations based. But things such as importing the shader in a project and later selecting it in the materials drop down without much effort, I'd consider basic functionality and would put it a bit higher in the priority list..

(03-09-2015 08:15 AM)Tottel Wrote:  

(03-08-2015 11:24 PM)Esenthel Wrote:  [...] and would need at least 3,000$ to cover the development cost. [...]

I was under the impression that we subscribed for a monthly/yearly fee to cover your development costs.

I feel sure he has other costs to cover besides development costs...such as food and shelter

I think the main issue is that the licensing has changed dramatically (pay once for forever -> subscription), but the contribution system is still the same.

If we pay once, it makes a lot of sense that we can further contribute to features that are important to us.
But now with the subscription system, isn't that what we are doing? We are already paying every month/year to get engine developments. What will you develop if no one contributes extra money? Features that aren't so important? I hope you can see that doesn't make any sense.

I don't see how the argument of "covering development costs" applies here: You now have a stable income, and all you have to do keep (or even expand) your paying-user-base is add important features that benefit many people.
In my opinion, adding custom shaders should be top priority right now, just like Para said. It's unheard of that an engine doesn't allow you to add custom shaders (for binary licenses).
Even worse: It used to allow custom shaders, but it was removed in a big engine update.

I don't know your current situation, I don't know if you are making enough money to live comfortably. But if you are, then the subscription system seems like the perfect tool to take more risks in development.. You can afford to spend time on big features because you have a steady income, and when you release the update, you can get even more subscribers.

(03-09-2015 04:18 PM)Zervox Wrote:  because amongst HWI and advanced occlusion culling and shader editor they are the features I really want in the engine at this moment

As for me such features are:

• integrated shader and special effects editor
• hardware instancing
• physical cloth editing and tutorial
• destructible objects creation
• advanced occlusion culling
• storm/rain effects
• plugin system
  

For the last item - i don't know how to implement it, but it would be cool to have an ability to create your own plugins (not like editor network interface) like plugins to wordpress, netbeans etc, when you place a file (like .dll) in "Plugins" folder and you get new items in your interface.

Just a little example - you create a plugin "Mirror an image", that adds two items to context menu in file list to files of "Image" type with the name "Mirror UP-DOWN" and "Mirror LEFT-RIGHT", and when you press it, the image is changed. And there is no reason to "export -> edit in editor -> reload" or use editor network interface.

Hello!

I think there is a bit of confusion about what the contribution system is about.
Normally I've always been working on the things in the engine according to my own plan of development.
The contribution system is about making some features in first priority.

Please let's not turn this thread into an argument between people