Where to get OpenCL SDK? - sdk

I tried to download OpenCL SDK. But no way. I have an AMD GPU so I searched on Google for AMD SDK but all the links from google and some tutorials are broken and there is no possibility to find the sdk through AMD developpers site.
Well I tried then Intel OpenCL SDK ... but there is no direct link. I tried to register and apply for the sdk but I don't see any download links in my mail box :(
Is it that difficult to get hands on OpenCL SDKs? Something I am missing?
UPDATE:
Finally as for OpenGL SDK is only needed for headers and libs. And tools ...
Looks like the SDK is deprecated and we should just use OpenCL SDK Lite that only contains headers and library files:
https://github.com/GPUOpen-LibrariesAndSDKs/OCL-SDK/releases Pretty disappointing.
There are still tools related to ROCm for linux but not much for Windows. CodeXL should work on windows thow.
For Intel SDK I finally received a confirmation mail and could download it. It took a couple of days.

Full Installers:
32-bit: http://amd-dev.wpengine.netdna-cdn.com/app-sdk/installers/APPSDKInstaller/3.0.130.135-GA/full/AMD-APP-SDKInstaller-v3.0.130.135-GA-windows-F-x86.exe
64-bit: http://amd-dev.wpengine.netdna-cdn.com/app-sdk/installers/APPSDKInstaller/3.0.130.135-GA/full/AMD-APP-SDKInstaller-v3.0.130.135-GA-windows-F-x64.exe

On Win64, you can build the SDK yourself. Get the OpenCL headers from Khronos, and build the import library from the OpenCL.dll using the VC++ tools. See csidev.net.

Related

Does the Webpack Vivado 2020.2 include the SDK

I am confused. Does Vivado 2021.2 include the SDK using the Webpack license?
Digilent says yes: https://reference.digilentinc.com/programmable-logic/guides/vivado-create-hdl-wrapper
(the hdl wraper is part of the SDK, peta linux, etc)
But, my installation does not seem to include it.
Where did I go wrong?
Thanks
Err... Turns out that Vivado went from SDK to Vitis. So now they are using Vitis for their higher level languages.
https://forum.digilentinc.com/topic/21686-does-anybody-have-the-xilinx-vivado-20194-sdk-xilinx-doesnt-have-in-the-archive/?tab=comments#comment-62563
https://forums.xilinx.com/t5/Vitis-Acceleration-SDAccel-SDSoC/Vitis-Community-How-can-we-help/td-p/1117972
Here are some examples:
https://reference.digilentinc.com/programmable-logic/guides/getting-started-with-ipi
https://github.com/Xilinx/HLS-Tiny-Tutorials

Boost Graph Library, stable on iOS?

The Boost Graph Library is a header-only library and does not need to be built to be used.
it contains some pre-made functions that I want for my iOS application:
dijkstra_shortest_paths
astar_search
However, Boost is not officially supported on iOS (or Android) due to lack of regression testing, and I'm not experienced enough with either Boost or iOS to be comfortable testing it.
Is anyone using Boost.Graph on iOS, and if so is it stable?
Yes, most definitely it's stable.
Apple's LLVM compiler (which is what you should be using for iOS apps) is one of the most C++11 compliant compilers available today, and as such can compile and run boost with absolutely no issues.
Not to mention the fact that iOS is also fully POSIX compliant, which makes most frameworks out there viable for use on iOS. When in doubt, try it and then if it doesn't work, come back and ask a question.
Leverage what is already there, and use it well!
As such boost libraries are stable, but Boost Graph libraries are failing on all darwin flavours (darwin- 4.4_0x intel- darwin- 12.0 intel- darwin- 11.1 darwin- 4.4)
http://www.boost.org/development/tests/release/developer/summary.html
May be you can try other alternatives, like LEDA (http://www.algorithmic-solutions.com/news/archive.htm)
Thanks.

Blackberry AES encryption error

I have implemented AES encryption in BlackBerry using code from the Knowledge Base Article "How to - Use Advanced Encryption"
I keep getting error:
Module 'net_rim_crypto' not found.
In the code I have imported net.rim.device.api.crypto.*; package, but I can't find the package net.rim.crypto.
Do I need to add any external jar for that?
You should install JRE 4.6 and set the API level to 4.6 and OS of the target device should be higher than JRE version.
If you have problem with BlackBerry API or OS version, you may consider to use bouncy castle for j2me.

How to compile a DirectX 11 app in MinGW

I've looked, and I can't find any material relating to using Direct3d 10 or 11 with MinGW. What do I have to do to get things working? I'm getting errors in the header files supplied by the DX SDK.
And don't anyone even think of suggesting Visual Studio.
It seems that here it is explained how to get DX headers from SDK work with MinGW: http://d.hatena.ne.jp/tbk/20110115/1295080728
Use Google Translate to translate page to english.
Alternatively try using DirectX headers from mingw64 project from here: http://sourceforge.net/projects/mingw-w64/files/
You'll need "Download headers and CRT source v2.0.1 (r4623) (7.0 MB)". Or you can check out them directly from svn: https://mingw-w64.svn.sourceforge.net/svnroot/mingw-w64/trunk/mingw-w64-headers/
May be this is answer.
http://www.gamedev.net/topic/619239-directx-11-mingw-w64/page_view_findpost_p_4908066

iPhone and OpenCV

I know that OpenCV was ported to Mac OS X, however I did not find any info about a port to the iPhone.
I am not a Mac developer, so that I do not know whether a Mac OS X port is enough for the iPhone.
Does anyone know better than me?
OpenCV does indeed work on the iphone. Use the configure script here to compile the library: cross compiling for iphone
You just have to cross-compile just as you do your apps.
OpenCV now (since 2012) has an official port for the iPhone (iOS).
You can find all of OpenCV's releases here.
And find install instructions here:
Tutorials & introduction for the new version 3.x
The latest build script from Computer Vision Talks works great for Xcode 4 and iOS 4.3 . I have tried the script myself and it is just awesome!
Here is opencv2.0 on iPhone iphone opencv test
OpenCV is now available as a framework for iOS. Just drag and drop into your project. It supports video capture too. See the article and get the example project here: opencv framework for ios
For the sake of transparency, I wrote this article and it is hosted on my company's website.
I haven't tried using OpenCV specifically, but I do dev for the iPhone and can say that most libraries I've tried that work on OS X DO NOT work on the iPhone out of the box. Some of them just needed a little tweaking/configuration to be done and then it was fine on the iPhone, but the reality is that the phone is missing quite a few backend components that OS X supports. Most complex libraries (OpenCV sounds like one of them) aren't going to work without a major effort - particularly since OpenCV sounds like it depends on several other external libraries as well...so those would have to be ported too.
All you need is to generate XCode project for OpenCV project using cmake or cmake gui tool.
Remember to set option to generate XCode project instead of the default option to use CMakeFiles.
Open generated project, change the base SDK to iPhone SDK, and hit build!
Since OpenCV does not support iOS at now (but they has announced iPhone support in version 2.2), highgui library won't compile. So if you need camera access you have to write it yourself.
Anyway, other libraries should compile and work on the device. (Works for me).
iPhone do support OpenCV if you want to use it first go to the best OpenCV on iPhone Documentation on the web: Yoshimasa Niwa's
I used it and i already have an app on the AppStore that uses Face Detection and Image Processing: Flags&Faces if you have any doubts please contact me.
Note that OpenCV runs very fast on Intel chips but the iPhone is arm. Of course OpenCV is extremely useful but it won't be that fast. Also, there's no way to get a live video stream on the iPhone so all of the normal potential CV applications sort of lose their appeal, don't they?
You can also install OpenCV using a package manager like Cocoapods.
To quote the installation guide:
You want to add pod 'OpenCV', '~> 3.0' similar to the following to
your Podfile:
target 'MyApp' do
pod 'OpenCV', '~> 3.0'
end
Then run a pod install
inside your terminal, or from CocoaPods.app.
Here's modified script (based on the one from LambdaJive) that builds universal OpenCV framework for iPhone/iPhone Simulator - universal-i386arm opencv framework
The following post by Yoshimasa does indeed work with I OS and the IPhone 4 and is able to access both the front and back cameras.
The link to the project is using opencv on iphone en
and the sample code is at webgit and it is linked from this article. I really encourage to read the article before getting the source code.
A project utilizing opencv on the iPhone (Lucas-Kanade optical flow to be exact). Source code available and app is on the AppStore as well -

Resources