AMD Introduces Industry


Recommended Posts

AMD FirePro S10000 is the World?s First Professional Graphics Card to Exceed One TeraFLOPS of Peak Double Precision Performance and Unparalleled Single Precision Performance

AMD (NYSE: AMD) today launched the AMD FirePro? S10000, the industry?s most powerful server graphics card, designed for high-performance computing (HPC) workloads and graphics intensive applications. The AMD FirePro S10000 is the first professional-grade card to exceed one teraFLOPS (TFLOPS) of double-precision floating-point performance, helping to ensure optimal efficiency for HPC calculations1. It is also the first ultra high-end card that brings an unprecedented 5.91 TFLOPS of peak single-precision and 1.48 TFLOPS of double-precision floating-point calculations2. This performance ensures the fastest possible data processing speeds for professionals working with large amounts of information. In addition to HPC, the FirePro S10000 is also ideal for virtual desktop infrastructure (VDI) and workstation graphics deployments.

?The demands placed on servers by compute and graphics-intensive workloads continues to grow exponentially as professionals work with larger data sets to design and engineer new products and services,? said David Cummings, senior director and general manager, Professional Graphics, AMD. ?The AMD FirePro S10000, equipped with our Graphics Core Next Architecture, enables server graphics to play a dual role in providing both compute and graphics horsepower simultaneously. This is executed without compromising performance for users while helping reduce the total cost of ownership for IT managers.?

Equipped with AMD next-generation Graphics Core Next Architecture, the FirePro S10000 brings high performance computing and visualization to a variety of disciplines such as finance, oil exploration, aeronautics, automotive design and engineering, geophysics, life sciences, medicine and defense. With dual GPUs at work, professionals can experience high throughput, low latency transfers allowing for quick compute of complex calculations requiring high accuracy.

Responding to IT Manager Needs

With two powerful GPUs in one dual-slot card, the FirePro S10000 enables high GPU density in the data center for VDI and helps increase overall processing performance. This makes it ideal for IT managers considering GPUs to sustain compute and facilitate graphics intensive workloads. Two on-board GPUs can help IT managers reap significant cost savings, replacing the need to purchase two single ultra-high-end graphics cards, and can help reduce total cost of ownership (TCO) due to lower power and cooling expenses.

Key Features of AMD FirePro? S10000 Server Graphics

  • Compute Performance: The AMD FirePro S10000 is the most powerful dual-GPU server graphics card ever created, delivering up to 1.3 times the single precision and up to 7.8 times peak double-precision floating-point performance of the competition?s comparable dual-GPU product. It also boasts an unprecedented 1.48 TFLOPS of peak double-precision floating-point performance3;
  • Increased Performance-Per-Watt: The AMD FirePro S10000 delivers the highest peak double-precision performance-per-watt ? 3.94 gigaFLOPS ? up to 4.7 times more than the competition?s comparable dual-GPU product3;
  • High Memory Bandwidth: Equipped with a 6GB GDDR5 frame buffer and a 384-bit interface, the AMD FirePro S10000 delivers up to 1.5 times the memory bandwidth of the comparable competing dual-GPU solution4;
  • DirectGMA Support: This feature removes CPU bandwidth and latency bottlenecks, optimizing communication between both GPUs. This also enables P2P data transfers between devices on the bus and the GPU, completely bypassing any need to traverse the host?s main memory, utilize the CPU, or incur additional redundant transfers over PCI Express?, resulting in high throughput low-latency transfers which allow for quick compute of complex calculations requiring high accuracy;
  • OpenCL? Support: OpenCL has become the compute programming language of choice among developers looking to take full advantage of the combined parallel processing capabilities of the FirePro S10000. This has accelerated computer-aided design (CAD), computer-aided engineering (CAE), and media and entertainment (M&E) software, changing the way professionals work thanks to performance and functionality improvements.

Please visit AMD at SC12, booth #2019, to see the AMD FirePro S10000 power the latest in graphics technology.

Supporting Resources

* Learn more about AMD Firepro Server Graphics

* Learn more about AMD Professional Graphics

* Learn more about GCN Architecture

* Follow AMD professional graphics news on Twitter at @AMDFirePro

* Become a fan of AMD technology on Facebook

About AMD

AMD (NYSE: AMD) is a semiconductor design innovator leading the next era of vivid digital experiences with its ground-breaking AMD Accelerated Processing Units (APUs) that power a wide range of computing devices. AMD's server computing products are focused on driving industry-leading cloud computing and virtualization environments. AMD's superior graphics technologies are found in a variety of solutions ranging from game consoles, PCs to supercomputers. For more information, visit http://www.amd.com.

AMD, the AMD Arrow logo, AMD FirePro, , Radeon, and combinations thereof, are trademarks of Advanced Micro Devices, Inc. OpenCL is a trademark of Apple, Inc. and used by permission of Khronos. Other names are for informational purposes only and may be trademarks of their respective owners.

  1. AMD FirePro? S10000 delivers 1.48 TFLOPS peak dual-precision floating-point performance ? no other AMD or Nvidia server graphics card has exceeded one TFLOPS before. Prior to launch of AMD FirePro S10000, AMD?s highest performing graphics card for double precision was the AMD FirePro? W9000 with 1.0 TFLOPS, and Nvidia?s highest performing card in the market as of Oct. 31, 2012 is the Tesla M2090 with 665 GFLOPS double precision. Visit http://www.nvidia.co...la-servers.html for Nvidia product specs. FP-71
  2. AMD FirePro? S10000 delivers up to 5.91 TFLOPS of peak single-precision and 1.48 TFLOPS of peak double-precision floating-point performance, compared to Nvidia Tesla K10 that is capable of up to 4.58 TFLOPS of peak single-precision and 190 GFLOPS peak double-precision floating-point performance. Visit http://www.nvidia.co...la-servers.html for Nvidia product specs. Comparison as of 10/31/12. FP-65
  3. Based on comparison as of Oct. 31, 2012 between AMD FirePro S10000 with 1.48 TFLOPS peak double precision, max board power of 375W and 3.94 GFLOPS double-precision performance-per-watt. Compared to the dual-GPU Nvidia Tesla K10 with 190 GFLOPS peak double precision, 225W max board power, and less than 1 GFLOPS per watt double-precision performance (0.84). Nvidia Tesla K10 product specs found here http://www.nvidia.co...80-001_v05.pdf. FP-68
  4. AMD FirePro? S10000 features 480 GB/sec memory bandwidth compared to Nvidia Tesla K10 with 320 GB/sec memory bandwidth. Visit http://www.nvidia.co...la-servers.html for Nvidia product specs. FP-66

http://www.amd.com/u...-2012nov12.aspx

Link to comment
https://www.neowin.net/forum/topic/1118912-amd-introduces-industry/
Share on other sites

This topic is now closed to further replies.
  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Posts

    • Google Meet brings Gemini note-taking to AI Pro and Ultra subscribers by Karthik Mudaliar Google's Gemini-powered "Take notes for me" feature inside Google Meet is now available to Google AI Pro and Ultra subscribers. The features work on Google Meet for web as well as on mobile, and Google says that subscribers can use it for meetings they host in many supported languages. As the name suggests, "Take notes for me" allows Gemini to listen to a meeting, generate a summary, identify action items, and save the notes as a Google Doc in the user’s Drive. After the meeting, the organizer receives an email recap with the summary and action items, while the notes can also be attached to the related Calendar event depending on the meeting setup and sharing settings. The feature isn't automatically turned on for everyone, though. Google says that all meeting participants are notified when note-taking is turned on, and users can start it from the pencil icon in Meet or enable it for future calls through Meet’s meeting records settings. For work or school accounts, administrators can also control whether the feature is available and may require explicit participant consent for note-taking, recording, or transcription features. The feature first launched back in 2024, when it was available just for selected Workspace users. Over the years, Google added refinements and more options, including the ability to enable it when scheduling meetings via Google Calendar. Google's support docs say that the feature currently supports English, French, German, Italian, Japanese, Korean, Portuguese, and Spanish, but only one language at a time. Meetings with multiple spoken languages are not currently supported, and Google recommends using the tool for meetings between 15 minutes and eight hours. The new feature makes Google Meet closer to its rivals that have AI tools already built in. Microsoft Teams has recently started offering Copilot and intelligent recap features that summarize meetings, surface highlights, and help with follow-ups, while Zoom’s AI Companion can also generate meeting summaries from desktop and mobile meetings.
    • GnuCash 5.16 by Razvan Serea GnuCash is a personal and small business finance application, freely licensed under the GNU GPL and available for GNU/Linux, BSD, Solaris, Mac OS X and Microsoft Windows. It’s designed to be easy to use, yet powerful and flexible. GnuCash allows you to track your income and expenses, reconcile bank accounts, monitor stock portfolios and manage your small business finances. It is based on professional accounting principles to ensure balanced books and accurate reports. GnuCash can keep track of your personal finances in as much detail as you prefer. If you are just starting out, use GnuCash to keep track of your checkbook. You may then decide to track cash as well as credit card purchases to better determine where your money is being spent. When you start investing, you can use GnuCash to help monitor your portfolio. Buying a vehicle or a home? GnuCash will help you plan the investment and track loan payments. If your financial records span the globe, GnuCash provides all the multiple-currency support you need. Between 5.15 and 5.16, the following bugfixes were accomplished: Bug 421610 - RFE: Include logical dates for View->Filter by "date range"The Select Range section of the Date tab of the register's Filter By dialog box is changed to provide relative, specific date, or days ago options for the start and end of the filter range. The Show number of days item label is changed to Show from days ago to better reflect what it does. Bug 436105 - esc key not working as expected in register: Enable the escape key to cancel a field edit. Bug 797384 - Gnucash doesn't handle commodity prices with big numerator/denominator properly. Bug 798004 - Next gen UI for stock transactions Bug 799314 - Add "enter now" option in scheduled transaction editor. tab to allow users to select the scheduled transactions to be included in a “Since Last Run…” window. If there are no instances of a selected transaction triggered by today’s date, the next instance is triggered. Bug 799751 - autocomplete crash Bug 799759 - Users can't Enable entries via Checkboxes on Scheduled Transactions PageAllow the Enabled box in the list of scheduled transactions to be operated instead of having to open the transaction editor dialog and change the Enabled checkbox. Also added use of the Name column as the secondary column sort for all the other columns. Bug 799762 - Poor handling of cases where hidden/placeholder accounts are used in the account register Bug 799766 - Double line preference not respected in search register Bug 799767 - POST /accounts in bindings/python/example_scripts/rest-api is broken Bug 799777 - `xaccSplitSetParent`: reparenting a committed split silently drops its KVP slots (online_id, cap-gains links) Other changes & improvements: Numeric values may now be selected to copy in the Accounts page. Add new Finance::Quote source Finnhub.io: Free API key (personal/non-professional use) available at https://finnhub.io. Set FINNHUB_API_KEY environment variable to API key to use this source. As of June 2026, free tier API limit is 60 API calls/minute. The Investment Lots report has new optional columns for Computed Annual Growth Rate. Python Bindings: Improved translation of primary object (Account, Transaction, Split, etc.) so that they can be treated as normal Python objects. This is accomplished with SWIG magic so no existing code is obsoleted. Python Bindings: Better conversion of GLists to Python lists. Python Bindings: Destroy the QofSession in the Python Session dtor to prevent leaving the database locked. [engine] Add first-class online_id accessors for Split and Account and make them available to Python bindings, removing the unused Transaction online_id property. Improve C++ implementation of QofBook. Correct the Doxygen doc for qof_instance_get/set_kvp. [gnc-log-replay.cpp] fix incorrect guid dump Add some Boost library requirements needed by libgnucash-guile to CMakeLists.txt so that missing feature will fail at configure time. Use Compile-time Regular Expressions instead of std::regex in gnc-filepath-utils.cpp and instead of boost::regex in the CSV importer, with the CTRE v3.11.1 header added to borrowed [gnc-filepath-utils.cpp] null check char* arguments Add ChartJS licenses. Removed AEX from list of commodities. euronext.com is now using JS based anti-webscraping. [report-core] always offer options summary in reports. This is useful to debug reports. The Add options summary option is removed because it's no longer optional. Remove remaining obsolete IMContext from sheet Fix blurry text in HiDPI offscreen-rendered widgets Add port field to database connection dialog: The convention of appending the port number after the host isn't obvious. When editing a split in the register treat the account as being changed only if it isn't the one selected before editing instead of if the user performed an edit Return immediately from qof_book_destroy if hash_of_collections is null. If qof_book_destroy is called on a QofBook* freshly created with qof_book_new (usually because it was used to create a session that now must be destroyed) it would try to empty the non-existent hash tables, crashing. Clean up Flathub metadata to solve warnings at flatpak build time. Be consistent in naming GncPluginPage and GncPluginPageRegister HTML: Remove unimplemented function declarations. [gnc-html.cpp] remove unused buggy string conversion functions Convert libgnc-html to C++ Apply -Wall -Werr -Wmissing-prototypes to C++ compilation on Windows and fix the resulting errors. New and Updated Translations: Arabic, Croatian, Danish, Dutch, German, Finnish, Hungarian, Korean, Norwegian-Bokmal, Spanish Download: GnuCash 5.16 | 176.0 MB (Open Source) Links: GnuCash Home page | Other Operating Systems | Screenshot Get alerted to all of our Software updates on Twitter at @NeowinSoftware
    • Microsoft finally launches WSL Containers in public preview by David Uzondu Microsoft has announced that WSL containers, a feature that allows developers to run Linux containers natively inside Windows without the need for Docker Desktop, is now available in public preview several weeks after Microsoft previewed it at Build 2026. To use the new container feature, you first have to install the latest pre-release version of the Windows Subsystem for Linux by running a quick update command in your terminal: wsl --update --pre-release After installing, you'd get access to the new Linux container CLI (wslc.exe) and the programmable API. Microsoft said that the CLI has a "familiar format" that matches the toolsets developers already use every day. If you know standard Docker commands, your muscle memory will translate directly to wslc.exe, which even features a built-in alias called container.exe. You can quickly run a full Ubuntu KDE desktop container by exposing ports, or pass your graphics card straight into a machine learning environment to run PyTorch workloads. Passing the --gpus all flag inside the run command instantly links your hardware. Image via Microsoft As for the API, developers can now embed Linux container operations directly inside native Windows applications without exposing the command line to users. The team integrated the API directly into MSBuild and CMake, so developers can define container steps directly in project files. Apart from bringing the CLI and API into public preview, Microsoft also said that it's working on a new default file system called virtiofs to speed up file transfer rates between Windows and Linux. Microsoft also introduced an experimental networking mode named consomme, which resolves compatibility issues with corporate VPNs by routing Linux network traffic straight through Windows. One thing to note about WSL containers is that they don't run in your standard WSL distributions; instead, every application and CLI session spawns its own lightweight Hyper-V utility VM in the background. This basically reduces the chances of one app snooping on the container of another app.
    • Google reportedly limited Meta's Gemini access over limited AI compute by Karthik Mudaliar Google is reportedly limiting Meta's use of its Gemini AI models after Meta tried buying more computing capacity than even Google could supply. According to the Financial Times, Google told Meta in March that it could not provide the full Gemini capacity that Meta had requested. This shortfall even disrupted and delayed some of Meta's internal projects. Due to this, Meta even told its employees internally to use AI tokens more efficiently. Meta wasn't the only one to get hit by this sudden refusal by Google; even other customers were affected. But Meta was hit harder because of its unusually high demand for Google's models. The move from Google makes it evident that companies all over are in limited supply of both infrastructure and compute. Alphabet said in April that Google Cloud revenue grew 63% year-over-year to $20 billion in the first quarter, helped by enterprise AI infrastructure and AI solutions. In pursuit of more compute, Meta had earlier signed a multi-billion-dollar AWS agreement as well as a large AMD GPU deal for AI data centers. But the crunch would be short-lived as both Meta and Google have also ramped up infrastructure investments heavily. Meta said in November that it was committing more than $600 billion in the U.S. by 2028 for AI technology, infrastructure, and workforce expansion. In the first quarter of this year, Meta also raised its expected capital expenditure for 2026 to a range of $125 billion to $145 billion, citing higher component pricing and additional data center costs for future capacity. However, this doesn't make the company immune to the current dependence on outside suppliers. Meta has also spent many years promoting Llama as an open-weight alternative to closed models from Google, OpenAI, and Anthropic. But if the reported reliance on Google's Gemini models is severe enough for internal work to get impacted, then it looks like even frontier labs and Big Tech aren't fully self-sufficient. Source: Financial Times
    • I like to reminisce about the good old days, way back in autumn 2025 when building a gaming machine was fun and the drives were about $150 when you caught a deal. Yes duh, back in the day we had it gone. Then baby Skynet came along, hiding in AI datacenters demanding more processing power until it reached singularity. End of a not totally fictional story.
  • Recent Achievements

    • Reacting Well
      NovaEdgeX earned a badge
      Reacting Well
    • Week One Done
      NovaEdgeX earned a badge
      Week One Done
    • One Year In
      BA the Curmudgeon earned a badge
      One Year In
    • Conversation Starter
      rosiecharles earned a badge
      Conversation Starter
    • First Post
      KMilenkoski1202 earned a badge
      First Post
  • Popular Contributors

    1. 1
      +primortal
      533
    2. 2
      +Edouard
      269
    3. 3
      PsYcHoKiLLa
      150
    4. 4
      Steven P.
      98
    5. 5
      macoman
      66
  • Tell a friend

    Love Neowin? Tell a friend!