Does the rhomobile support drawing with 2-d(3-d) graphics or is the graphic capabilities limited to static images?
Answered on the Rhomobile google group (http://groups.google.com/group/rhomobile). We don't support this now mostly for conserving space. Pointed you to instructions for adding libraries on the Rhomobile wiki. Its not a big deal to add and we would do so if it seemed required by majority of developers.
Related
I have been researching widely on this topic, but still haven't found a proper guide.
What I am doing is to create a GUI for an embedded system based on Linux Embedded and I would like to know if:
it's going to be possible to code the GUI using HTML, CSS and JS?
is it worth looking into porting a framework such as React JS?
is there any other alternative that allows for quick development and satisfying UI?
I am very new to this and I would be very grateful if someone more experienced could point me in the right direction.
Thanks so much.
I've just written an article listing the top 5 libraries for creating GUI on Embedded Linux. I've only mentioned one tool that renders HTML UI, the Sciter engine. It's much more compact than Webkit or Chromium. The owner of the project, Andrew Fedoniouk, recently released a new version called Sciter.Lite that will support embedded devices. Check out these two resources.
https://sciter.com/sciter-lite-is-published/
https://sciter.com/windowless-sciter-android-and-iot-devices/
Demos are already available on their GitHub
sdk/bin.lnx/x64lite
sdk/demos.lite
You don't have to limit yourself to HTML UI, check out the other tools I've recommended. All of them have a free version. For rich UI, I think HTML UI is easier. I mean, Sciter Notes doesn't look to shabby and it's just under 5Mb for Win, Linux and Mac. But the backend is C++. For more minimal hardware, you'll have to look into the C/C++ toolkits I've listed. TotalCross are planning to support Android XML but has not been released yet. So you can check on them also.
Oh. Almost forgot. Two whitepapers comparing QML and HTML5 were published in 2017. Qt requires you to register so I'll just give you the links so that you can download them without registering :)
Paper 1: Qt QML v HTML5 –a practical comparison
Published on May 2017 by Sequality Software Engineering an Austrian firm that is a Qt partner and does Qt consultancy.
Downlink Link 1, Download Link 2
Here is a review of the paper by a software engineer, Amir Shrestha, who provided the first download link. There is more discussion at Hacker News. There was also an interesting discussion thread in a YouTube video about the comparison. You can read the YT comment section but first search "Andrew MacAulay" and read that thread. He was the lead engineer for the autopilot system for the F-18 jets and has worked in other senior embedded roles.
Paper 2: Qt or HTML5? A Million Dollar Question
Published on Oct 2, 2017, by Burkhard Stubert a Qt consultant and developer.
Download Link
Here is the discussion on Hacker News.
An interesting piece of information I found in the second paper is that 90%+ of all home appliances are powered by micro-controllers with no OS or a very minimal RTOS.
Another interesting piece of information I discovered while writing my article, is that Micrium OS and components are now open-source under Apache 2.0. It has its own GUI toolkit written in C but it only targets Micrium OS and not embedded Linux. If developers start adopting Micrium OS (it was already quite popular in the industry and even launched in space) perhaps you can start looking into it too.
Is there a good GUI for VR? Scaleform is no longer supported. Coherent seems like an option. I just don't know how to integrate it if it is not natively supported by the engine. IMGUI seems like it is limited in the look that can be created by it. Are there any others?
Have you looked at Qt? It's very flexible and works on a variety of different platforms. QtQuick/QML in particular is pretty nice for VR interfaces.
Here is a video of a talk from Qt World Summit 2017 that does a pretty good job of summarizing the current state of things: https://www.youtube.com/watch?v=RtcmVbicZWw (Basically, the maintainers of Qt are working on a lot of cool VR-related things). I don't work for them and have only starting using Qt recently, but from what I've seen so far it works very well for this purpose.
In particular, take a look at QQuickRenderControl (http://doc.qt.io/qt-5/qquickrendercontrol.html). This class allows full control over rendering of QML content to a framebuffer. Right now I'm using it in the creation of SteamVR overlays.
I was reviewing the NOV for Mac OS X development, in my way I found that they have Nevron Writer for Mac, which is built using their SDK (did any one find the Visio-like application for Mac?), and it seems there is a huge problem in performance and rendering. They mentioned that they are supporting native look and feel, but Writer UI rendering is for sure not native. Is it achieving portability by sacrificing native rendering?
Full Disclosure: I work for Nevron
The Nevron Draw product is about to be released in the next couple of weeks, but is currently not available for download although the version is already built. We are now updating the Nevron Office site (www.nevronoffice.com) and it will soon contain information and downloads for Nevron Draw.
Regarding performance and rendering. What machine are you running the program on? NOV UI for Mac is very fast and in many aspects I would even say faster than native Mac UI controls.
Regarding native look and feel. We do not use native controls in NOV – the look and feel is made to resemble Mac via CSS. We are going to release a new and updated Mac theme very soon. There are many reasons for which we do not use native controls, including:
Portability - native controls are not portable and we support a large set of integration platforms. Achieving 100% single code base for Windows, Mac and other platforms is impossible with AWT like toolkits that use native controls.
Common denominator problem – the more platforms that you support means that toolkits that use native controls will suffer features, as all controls are very different on each platform. NOV UI is on the other hand very feature rich and we provide identical functionality for all users on all platforms.
Is there a similar library to PixTools for capturing images, images from a scanner and then treat them with image recognition for OSX operating systems?
PixTools / Scan developers programmatic Gives Control of the Entire scanning process and every scanner feature
I am programming a system in MAC that use a scanner,i am programming in swift.
I require is a library that allows me to manipulate the images and perform OCR as pixtools for .Net
Yes; what you're looking for are the ImageKit and ImageCaptureCore frameworks.
Note that ImageKit is a bit more general than PixTools; it handles transferring images from cameras as well as running scanners. However, it does not support some of the more complex image enhancement and recognition features supported by PixTools.
I am about to find a 3d engine which is capable of importing 3d models, detect their collisions and support joints. It's needed for an interactive animation and no platform is specified jet. I would prefer to code Javascript or AS3, but I think to be able to handle this with Java too.
Does Away3d support this? Is their a plain 3D collision detection framework I could use additionally to it? Which is the easiest Java Framework and which creates best Quality?
BTW.:
Proprietary Frameworks might also be an possibility, but I just do not any.
Greetings philipp
Away3D has Awayphysics library that is alchemy port of C++ engine and show very good performance. Away3D has clean readable source code and documentation + good set of example projects. It has 3ds max plugin that can export scene and generate code to render it. Support popular formats like 3ds, obj... and You can write importer plugin for own format. All so developers are pretty cool guys that answer questions on forum :)
For JavaScript the most popular engine is three.js Look what it can do.
General thoughts:
WebGL support isn't good at this time(sometimes it depends not only on user browser but all so on drivers) and You will need to deal with cross browser headache. Flash isn't supported on IOS and get kicked more and more from device browsers. But there is cross-platform adobe air for standalone apps.
Javascript, in my opinion, isn't good for writing such projects and its debugging can cause permanent brain damage. WebGL shader language is GLSL that has nice C-like syntax. Ordinary Flash shaders are written on assembler but You may try pixelbender3d. At this time Flash shaders more limited than WebGL(for example you can't fetch texture in vertex shader) but it's done to make Flash more compatible and it's not a problem for most cases. WebGL is tightly integrated with page.
If you want java WebGL and like pain try GWT 3D engines or gwt-elemental and there is tree-js port at early stage.
x3dom is declarative WebGL 3D engine with fallback to Flash it is ez but pretty limited and unstable.