Language Selection

English French German Italian Portuguese Spanish

NVIDIA launches GeForce 7800 GTX - reviews are rolling in

Filed under
Hardware

Like we already reported weeks ago NVIDIA today launched the GeForce 7800 GTX (G70). The GeForce 7800 GTX is made on a 110nm process and uses 303 million transistors. The graphics core runs at 430MHz and has 24 pixel pipelines and 8 vertex shaders.

The card's architecture is based on the previous generations but does feature some new things such as an improved shader engine which NVIDIA called CineFX 4.0. Transparant Dynamic Ultra Sampling (TSAA) is another new feature, to improve AA quality.

After taking a look at some of the reviews I noticed that the GeForce 7800 GTX perform better than a GeForce 6800GT SLI system in most benchmarks. Mostly the difference accounts for about 10 percent but in UT 2004 Icetomb (1600x1200: 0x AA/0x AF) the ATI X850XT scored 42 percent less than the new 7800GTX!

However, it should also be noticed that the GeForce 6800GT SLI and ATI X850XT seem to be a bit faster in some benchmarks.

Fortunately the new GeForce 7800 GTX uses single slot cooling and uses 'only' 100W. This is pretty impressive as a GeForce 6800 Ultra SLI system uses about 220W and takes four slots.

TBREAK says the 7800 GTX consumes less power than the GeForce 6800. The cooling solution of the 7800 GTX reference card produces only 24dB and keeps the card at 47 degrees idle and 55 degrees on load.

Here are the first reviews so you can take a look at the specifications and performance of NVIDIA's new breed: (more reviews will be added over the day)

  • TBREAK
  • Tom's Hardware Guide
  • HTPCNews
  • NV News
  • NV News (some SLI results)
  • Xtremesystems (Overclocked SLI tests -15990 3DMark05)
  • Bjorn3d
  • Digit Life
  • Hardware Secrets
  • TrustedReviews
  • AnandTech
  • Neoseeker

    NVIDIA claims limited quantities of the GeForce 7800 GTX will be immediately available in stores for a cost of about $599.

    dvhardware.net

  • More in Tux Machines

    Linux and Graphics: AMD, Linux 4.14 LTS, Etnaviv Gallium3D

    • Linux 4.14 Ensures The "Core Performance Boost" Bit Gets Set For AMD Ryzen CPUs
      Recently making waves in our forums was talk of a kernel patch to address a case where the AMD CPB (Core Performance Boost) isn't being exposed by Ryzen processors. Here's more details on that and some benchmarks. Being talked about recently is f7f3dc0: "CPUID Fn8000_0007_EDX[CPB] is wrongly 0 on models up to B1. But they do support CPB (AMD's Core Performance Boosting cpufreq CPU feature), so fix that."
    • Linus Torvalds Is Confident That Linux Kernel 4.14 LTS Will Arrive on November 5
      Development of Linux 4.14, the next LTS (Long Term Support) kernel series, continues with the fifth RC (Release Candidate) milestone, which was announced by Linus Torvalds himself this past weekend. According to Linus Torvalds, things have finally starting to calm down for the development of the Linux 4.14 LTS kernel, and it looks like the RC5 snapshot is smaller than he would have expected, at least smaller than last week's RC4, which is a good thing, meaning that there won't be need for eight RCs during this cycle.
    • Etnaviv Gallium3D Is Almost To OpenGL 2.0 Compliance
      The Etnaviv Gallium3D driver that provides reverse-engineered, open-source graphics support for Vivante graphics hardware is almost to exposing OpenGL 2.0. Etnaviv contributor Christian Gmeiner today posted a set of patches for adding occlusion queries support to the driver. The code at just over one thousand lines of code is the last major feature needed for exposing desktop OpenGL 2.0 capabilities with this community-driven driver.
    • AMD Developers Begin Making Open-Source FreeSync/AdaptiveSync Plans
      While the AMDGPU DC code is expected to land for Linux 4.15 with goodies like Vega display support, HDMI/DP audio, and atomic mode-setting, one of the sought after display features won't be initially supported: FreeSync or the VESA-backed AdaptiveSync. As we've known for a while, while AMDGPU DC fills out the requirements for being able to support FreeSync, the last bits of the implementation are not present as the interfaces are basically yet to be decided among the open-source driver developers. While AMD can post their existing FreeSync code as found in AMDGPU-PRO hybrid driver, they are trying to come up with a more standardized interface that will satisfy the other upstream Linux driver developers too that might want to support AdaptiveSync.

    Servers and Red Hat: Cloud Foundry, Docker, CRI-O 1.0, Alibaba and Elasticsearch

    • How to deploy multi-cloud serverless and Cloud Foundry APIs at scale
      Ken Parmelee, who leads the API gateway for IBM and Big Blue’s open source projects, has a few ideas about open-source methods for “attacking” the API and how to create micro-services and make them scale. “Micro-services and APIs are products and we need to be thinking about them that way,” Parmelee says. “As you start to put them up people rely on them as part of their business. That’s a key aspect of what you’re doing in this space.”
    • Docker Opens Up to Support Kubernetes Container Orchestration
      There's been a lot of adoption of Kubernetes in the last few years, and as of Oct. 17 the open-source container orchestration technology has one more supporter. Docker Inc. announced at its DockerCon EU conference here that it is expanding its Docker platform to support Kubernetes. Docker had been directly competing against Kubernetes with its Swarm container orchestration system since 2015. The plan now is to provide a seamless platform that supports a heterogenous deployment that can include both Swarm and Kubernetes clusters. "Docker adapts to you because it's open," Docker founder Solomon Hykes said during his keynote address at DockerCon.
    • Introducing CRI-O 1.0
      Last year, the Kubernetes project introduced its Container Runtime Interface (CRI) -- a plugin interface that gives kubelet (a cluster node agent used to create pods and start containers) the ability to use different OCI-compliant container runtimes, without needing to recompile Kubernetes. Building on that work, the CRI-O project (originally known as OCID) is ready to provide a lightweight runtime for Kubernetes.
    • Red Hat brings its open source solutions to Alibaba Cloud
      Alibaba Cloud has joined the Red Hat Certified Cloud and Service Provider program, with Red Hat solutions to become directly available to Alibaba Cloud customers in the coming months.
    • Elasticsearch now on Alibaba Cloud, eyes China market
      The Amsterdam-based company behind Elasticsearch and Elastic Stack said the new offering would be available to Alibaba Cloud customers as an add-on, giving them access to real-time search, logging, and data analytics capabilities.

    Software: VirtualBox 5.1.30, Cockpit 153, GNOME Mutter 3.27.1, KDE Neon

    • Oracle Releases VirtualBox 5.1.30 to Patch Glibc 2.26 Compile Bug on Linux Hosts
      Oracle released VirtualBox 5.1.30, a minor maintenance update to the open-source and cross-platform virtualization software that addresses a few important issues reported by users from previous versions. Coming one month after the VirtualBox 5.1.28 release, which probably most of you out there use right now on your personal computers, VirtualBox 5.1.30 contains a fix for a Glibc 2.26 compilation bug for Linux hosts and a 3D-related crash for Windows guest that use the Windows Additions package.
    • Cockpit 153
      Cockpit is the modern Linux admin interface. We release regularly. Here are the release notes from version 153.
    • GNOME Mutter 3.27.1 Brings Hybrid GPU Support
      Mutter 3.27.1 has just been released as the first development release for the GNOME 3.28 cycle of this compositor / window manager. The change most interesting to us about Mutter 3.27.1 is support for hybrid GPU systems. The context for the hybrid GPU system support is explained via this bug report, "supporting systems with multiple GPUs connected to their own connectors. A common configuration is laptops with an integrated Intel GPU connected to the panel, and a dedicated Nvidia/AMD GPU connected to the HDMI ports."
    • #KDE #KDENEON Release bonanaza! Frameworks, Plasma, KmyMoney and Digikam

    Intel Ads as 'Articles'