Saturday, April 12, 2008

Brahma - Calculations within the GPU

I have the test library setup now for Brahama, the GPU wrapper for Microsoft Accelerator.  It looks fast.  Really fast.  And the ability to mix GPU & CPU execution is a killer feature.  Match this up with an Nvidia Tesla and you have one kickass computer.

But does it play Flight Simulator?

Sorted 4096 Vector4s on the GPU in 0.1739 seconds, average time for each iteration was 0.0000
    Validating sort results: Error
    Array was set up in 0.0004 seconds
    Sorted 4096 Vector4s on the  CPU in 1.1536 seconds, average time for each iteration was 0.0003

I hope the focus stays on Brahma, the GPU-executing class wrapper.  Do I want to get rid of WPF and keep XAML?  No, I just want to see some more examples of Brahma with real-world Cuda-like samples.

How about some Silverlight integration too?

XAML without WPF = XOML

For a long time, I’ve been writing about how messed up Xaml is because of WPF-interdependency, and how Microsoft should have decoupled the Xaml bits from WPF, etc (Some of these rants were on my old blog). Well, I finally decided to do something about it, and I’m writing a XAML-like library that loads, and writes markup using its own XomlLoader and XomlWriter. At this point I’m calling the markup eXtensible Object Markup Language (I know that the Workflow Foundations’ designer files have that extension…) because that’s what its meant to be. It is also aimed to run on the .NET Framework, Mono and the .NET CF. :)

Using Xoml, you will be able to data-bind, forward reference, resource reference, write markup-extensions and more. I will be putting this project up for download shortly, look out for more news on it.

Ananth’s very own blog

No comments: