FirePro W8100 Professional Graphics Solution Announced by AMD

AMD has just launched the AMD FirePro W8100 professional graphics card, bringing a new generation of AMD Graphics Core Next architecture as well as even grater workstation performance, which is dubbed to be 38x better than any other competing product on the market.

Featuring OpenCL, a best-in-class 8 GB of GDDR5 memory, 4.2 TFLOPS single-precision compute performance, 2.1 TFLOPS double-precision compute performance and of course OpenCL, the AMD FirePro W8100 is said to be designed for Media & Entertainment workflows, as well as engineering analysis and supercomputing applications.

In terms of productivity, the FirePro W8100 is said to be the ideal professional-grade solution for 4K Computer Aided Design (CAD) and other applications dedicated towards video editing, colour correction, compositing, design visualisation and GPU-accelerated compute tasks. Thanks to its super-fast memory and vast capacity, users are said to be able to load large datasets or handle ultra-HD video frames on the GPU’s internal memory for processing purposes, resulting in reduced lag and improved overall responsiveness.

Users are also said to have the option to combine up to four AMD FirePro W8100 cards in a single system in order to have over 16 TFLOPS compute performance, in addition to increased productivity with up to four 4K display support and certified application performance with differentiated graphics features.

The FirePro W8100 is said to deliver support for increasing real-time 4K productivity with single-GPU and multi-GPU configurations from various system integrators, including Armari, CARRI, Colfax, Exxact, Mouse Computers, PSSC Labs, Scan International, Tarox, Versatile Distribution Services, Workstation Specialists and Wortmann.

AMD has stated that it will start shipping in July from SAPPHIRE Technology, AMD FirePro Ultra Workstation providers as well as major workstation providers.

Thank you TechPowerUp for providing us with this information
Image courtesy of TechPowerUp

Leaked Benchmarks Show NVIDIA’s GTX 750 Ti Slower Than The GTX 660

After a few rumors about the Maxwell chipset and its release date in February, we now hear about statistics regarding the GeForce GTX 750 Ti suggesting that NVIDIA’s first Maxwell GPU is slower than the current GTX 660, according to WCCF‘s article. Take the news with a grain of salt though, nothing official has been released regarding the Maxwell GPU, nor its performance. However, nobody can confirm, nor deny its authenticity.

We see the GeForce GTX 750 Ti is around 10 to 15 percent slower than the GTX 660. A bit of a good news comes from WCCF, stating that the benchmarks were made using Single Precision format, while the new Maxwell GPU is known for its Double Precision GPU. Also, it makes no sense for NVIDIA to release a graphics card which is slower than its previous series.

Whether the benchmarks are valid or not, it remains to be seen. In the meantime, we are awaiting more official information about the Maxwell GPU and possibly some more benchmarks of the GTX 750 Ti after its release. Maybe we will even see the GTX 750 Ti perform in the new PCMark 8 that was released a few days ago.

Thank you WCCF for providing us with this information
Image courtesy of WCCF