Def
Fishaholic
- Joined
- Sep 5, 2004
- Messages
- 679
- Reaction score
- 0
Yeah, it's the modelling that we need. The code is going to be relatively straightforward, I'll code it as an MFC SDI app shell for the GUI with the view handled by DX.
As far as I can see, we can't start without a modeller otherwise all the work will be for nothing. If you want to be the modeller, you'll need to be prepared to render all the scenery (rocks, bogwood etc) and fish meshes to .X files (start with .X imports and we can move to .3DS, MD3 or a custom format later). The plants etc don't need to be rendered, we just need high quality pictures of them to use as billboards.
I just want to stress this for people that want to get involved but don't have coding experience - WE NEED HIGH QUALITY PHOTOS OF INDIVIDUAL OBJECTS. For example, let's say we wanted to use this plant and I've just taken a photo of it in my tank:
Ignore the quality issue for a moment, the actual photo itself is unnacceptable for a program like this. More than just the plant is visible - you can see substrate, another plant etc etc. This is actually what we need:
It's just a representation, but hopwefully it gets the point across. Every single plant and photo we use MUST BE CROPPED PRECISELY. The reason for this is so we can use it in the app - the plant must be on a uniform background so that it can be made transparent and placed in the virtual aquarium.
That is *a hell of a lot of work*. Be under no illusion. If I'm coding this and you send me an image that isn't good enough, I'll send it back to you because "nearly right" looks VERY WRONG on screen. If you don't think you can deal with this precision, then please don't volunteer.
Here's what we need:
- A modeller, who knows their stuff and can use 3ds/maya and export .X files
- Many photographers, who can produce high quality photos of plants, rocks, substrate, bogwood etc for use as billboards and textures
- Many art people, who will spend their time enhancing, cropping and editing the images for the coders to use.
- Data people like dirtydogg who will create XML data for the properties of each and every object; plants, fish, water parameters, wood, substrate etc
On the code side, it sounds like jflowers has the necessary knowledge so far. I propose the base system is based on a Visual C++ 6 workspace, as that can be read in VC6, VC.Net and VC2003. DirectX 9 will be used for the 3D view, and MFC will be used for the interface. It is coded to accept plugin api DLL's (through late binding or similar) so we can easily extend the functionality and code 2 things at the same time.
Up front, I'll say that this is going to be a LOT of work, no matter what anyone thinks. I've coded commercial DirectX apps and trust me - it's not easy. I'm happy to even do this project for free (ie: give it away to board members etc), but only as long as we have a bunch of people who are DEDICATED to getting the job done. I don't want to get halfway through this and have everyone bugger off because they got bored. That will be a waste of everyone's time, and I know from experience that managing virtual teams is not easy.
If enough people commit, we'll get the mods to create a new forum where we can post updates and keep in touch. Meanwhile, post to this thread if you'd like to work on it.
As far as I can see, we can't start without a modeller otherwise all the work will be for nothing. If you want to be the modeller, you'll need to be prepared to render all the scenery (rocks, bogwood etc) and fish meshes to .X files (start with .X imports and we can move to .3DS, MD3 or a custom format later). The plants etc don't need to be rendered, we just need high quality pictures of them to use as billboards.
I just want to stress this for people that want to get involved but don't have coding experience - WE NEED HIGH QUALITY PHOTOS OF INDIVIDUAL OBJECTS. For example, let's say we wanted to use this plant and I've just taken a photo of it in my tank:
Ignore the quality issue for a moment, the actual photo itself is unnacceptable for a program like this. More than just the plant is visible - you can see substrate, another plant etc etc. This is actually what we need:
It's just a representation, but hopwefully it gets the point across. Every single plant and photo we use MUST BE CROPPED PRECISELY. The reason for this is so we can use it in the app - the plant must be on a uniform background so that it can be made transparent and placed in the virtual aquarium.
That is *a hell of a lot of work*. Be under no illusion. If I'm coding this and you send me an image that isn't good enough, I'll send it back to you because "nearly right" looks VERY WRONG on screen. If you don't think you can deal with this precision, then please don't volunteer.
Here's what we need:
- A modeller, who knows their stuff and can use 3ds/maya and export .X files
- Many photographers, who can produce high quality photos of plants, rocks, substrate, bogwood etc for use as billboards and textures
- Many art people, who will spend their time enhancing, cropping and editing the images for the coders to use.
- Data people like dirtydogg who will create XML data for the properties of each and every object; plants, fish, water parameters, wood, substrate etc
On the code side, it sounds like jflowers has the necessary knowledge so far. I propose the base system is based on a Visual C++ 6 workspace, as that can be read in VC6, VC.Net and VC2003. DirectX 9 will be used for the 3D view, and MFC will be used for the interface. It is coded to accept plugin api DLL's (through late binding or similar) so we can easily extend the functionality and code 2 things at the same time.
Up front, I'll say that this is going to be a LOT of work, no matter what anyone thinks. I've coded commercial DirectX apps and trust me - it's not easy. I'm happy to even do this project for free (ie: give it away to board members etc), but only as long as we have a bunch of people who are DEDICATED to getting the job done. I don't want to get halfway through this and have everyone bugger off because they got bored. That will be a waste of everyone's time, and I know from experience that managing virtual teams is not easy.
If enough people commit, we'll get the mods to create a new forum where we can post updates and keep in touch. Meanwhile, post to this thread if you'd like to work on it.
