Bookmark and Share


Nvidia Corp. and Portland Group last week introduced a special compiler that can make software originally developed for Nvidia CUDA architecture to x86 and presumably vice versa. This allows software developers to ensure broad compatibility of their programs and offer certain advantages for highly-parallel GPU architecture. However, it will hardly make the life of software makers much easier, according to Alex Herrera, an analyst with Jon Peddie Research.

There are many reasons why different applications, including those in supercomputer space, are not remade for graphics processors, such as ATI Radeon or Nvidia GeForce, despite of bright prospects of higher performance. One of the main reasons is legacy code that continues to be used and that will hardly be dropped since it does already work. The compiler jointly developed by PGI and Nvidia will allow developers to test the CUDA-based software approach on x86 platforms and determine its reliability. Nonetheless, it may work stably enough, but performance of CUDA-based software on x86 will hardly be optimal, claims Mr. Herrera.

Just like GPU-based Nvidia's PhysX tools that rely on CUDA do not support SIMD extensions like SSE2, the new compiler may not support things like AVX found in AMD Bulldozer and Intel Sandy Bridge microprocessors. As a result, the application will not run with maximum possible performance on x86 platforms.

"CUDA on x86 is going to be slower than an application optimized to run on x86 without CUDA, probably a lot slower. So a developer running a CUDA application on x86 and then on Fermi is going to see a larger speed-up than he might otherwise have had had he first optimized on a conventional, non-CUDA x86 platform. Bigger speedup numbers serve Nvidia’s purposes of showcasing how much faster GPUs are than CPUs on many floating-point intensive applications," said the analyst.

In the end, designers of both special-purpose and commercial software for consumers will still have to implement different code-paths for different hardware, something that they already do.

Tags: Nvidia, x86, CUDA, Bulldozer, AVX, JPR, Geforce, Radeon, OpenCL


Comments currently: 2
Discussion started: 09/29/10 02:16:13 AM
Latest comment: 09/29/10 05:49:08 AM


Yes, in the last ten years, you could say that compilers were everything. The fact that Intel's compilers are the most popular and AMD has no compiler of its own, proved to be very detrimental for AMD's performance and AMD users in general as Intel has done quite few tricks just to make AMD performance worse in any application compiled with an Intel compiler.
0 0 [Posted by: East17  | Date: 09/29/10 02:16:13 AM]

So like, putting lipstick on the pig, except they cheated out on the lipstick.

Why am I not surprised? They hobble PhysX for the CPU, disable it when there is a non nVidia GPU in the system, and now they pull this kind of crap.

If I was a programmer/dev, I'd be pretty pissed off by that.
0 0 [Posted by: RtFusion  | Date: 09/29/10 05:49:08 AM]


Add your Comment

Related news

Latest News

Tuesday, July 22, 2014

10:40 pm | ARM Preps Second-Generation “Artemis” and “Maya” 64-Bit ARMv8-A Offerings. ARM Readies 64-Bit Cores for Non-Traditional Applications

7:38 pm | AMD Vows to Introduce 20nm Products Next Year. AMD’s 20nm APUs, GPUs and Embedded Chips to Arrive in 2015

4:08 am | Microsoft to Unify All Windows Operating Systems for Client PCs. One Windows OS will Power PCs, Tablets and Smartphones

Monday, July 21, 2014

10:32 pm | PQI Debuts Flash Drive with Lightning and USB Connectors. PQI Offers Easy Way to Boost iPhone or iPad Storage

10:08 pm | Japan Display Begins to Mass Produce IPS-NEO Displays. JDI Begins to Mass Produce Rival for AMOLED Panels

12:56 pm | Microsoft to Fire 18,000 Employees to Boost Efficiency. Microsoft to Perform Massive Job Cut Ever Following Acquisition of Nokia