Language Selection

English French German Italian Portuguese Spanish

Graphic cards to get water cooling

Filed under
Gaming

There are very few components which have seen similar steep price increases as the graphic card space. Admitted, performance of high-end chips has increased dramatically over the past years, but so have price points: Just three years ago, flagship models were typically priced in the $400 range. With gamers demanding more performance and willing to shell out more cash for more pixel power, the list price of top end cards climbed to $500 and $600 and most recently even to $1000 for a 512 MByte model of Nvidia's 6800 Ultra.

One of the companies profiting from this trend is Lake Forest, Ill., based BFG, which is one of the few very visible American graphics cards manufacturers. While the 3-year old company employs a full-time staff of just 16 people, it is the leading supplier of Nvidia cards in US retail - most recently accounting for a market share of about 25 percent. Despite BFG recently launched entry-level products under the "3DFuzion" brand, the company caters mainly to gamers and carries most Nvidia chips in its product line. It was one of the very few suppliers that was able to ship GeForce 6 products into retail stores such as BestBuy or CompUSA before fall of last year.

Given the firm's success and its deep relationship with Nvidia, we were sure we are at the right address to inquire about upcoming trends in this industry. Co-founder and executive vice president of marketing Scott Herkelman and Ric Lewis, executive vice president of operations, heat dissipation is not just a problem of microprocessors. Also graphics card will be affected soon and will have to be equipped with more efficient cooling technologies.

"With form factors of PCs getting smaller we have to come up with new cooling designs to get to the performance level the chip is able to deliver. This is especially the case when people are overclocking their systems," Herkelman said. While he does not believe that mainstream cards will require a cooling method other than a fan, he said that high- end cards may go a different way: "The most promising approach to cool a graphics card is water cooling. There are other forms of cooling concepts we cannot talk about yet, but all of them are part of a non-air cooling approach."

Current grapohic chip generations will continue to work well with traditional fans, Herkelman said: "We are not yet at a point where consumers have to be concerned about buying water cooling systems. But in about two or five years, consumers and even OEMs will have to think about how to cool a whole box to ensure stability and performance."

Just one sign for this trend may be some more or less prototype-cards that were shown at this year's Computex tradeshow. Several Taiwanese card manufacturers showed dual-chip graphic cards with up to 1 GByte of memory and a power consumption of 230 watts when idle. Another indication is Nvidia's SLI technology: Ric Lewis said that many BFG customers are ordering not one, but two SLI-capable graphic cards right away. "We were definitely surprised how many SLI units we can sell overall," indicating that dual graphic cards are quickly going mainstream in the higher-end gaming segment. But two graphic cards of course consume more power and create much more heat than just one device: "It is interesting that a lot of people, who order SLI graphics cards, only have 300 or 250 watt power supplies. But if you run such a system you want to have at least a 500 watt power supply," he said.

Gamers also spend much more money on their graphics system than ever before. Cards that listed for about $600 last year, fetched almost $1000 on eBay due to their short supply. This year, the first appeared that listed for a stunning $1000. Nvidia and ATI already mentioned that $1000 may not be the final word for future cards - and BFG agrees. "We proved that we can sell a $1000 graphics card. We sold out of every one of our $999 512 MByte cards very quickly," Herkelman said. For the next generation, he expects a range of $1000 to $1200 for high-end versions - which will sell "as long as you can show that the hardware is good for something." He declined to comment on where the pricing limit for such cards may be.

Herkelman justified the price for high-end cards and said that such products are not just about bragging rights. "Golf is a great analogy. There are people who buy the latest driver every year to get an extra one or two yards or more consistency. We refer in the same way to this industry: We spend a lot of money on computers ourselves, since we want the extra frame rate or visual clarity in a game to see things a little earlier and better. These cards really bring an advantage, especially in multiplayer gaming."

Source.

More in Tux Machines

Trisquel 9.0 Development Plans and Trisquel 8.0 Release

  • Trisquel 9.0 development plans
    Just as we release Trisquel 8.0, the development of the next version begins! Following the naming suggestions thread I've picked Etiona, which sounds good and has the fewest search results. We currently do our development in a rented dedicated server in France, and although it is functional it has many performance and setup issues. It has 32 gigs of RAM, which may sound like plenty but stays below the sweet spot where you can create big enough ramdisks to compile large packages without having to ever write to disk during the build process, greatly improving performance. It also has only 8 cores and rather slow disks. The good news is that the FSF has generously decided to host a much larger dedicated build server for us, which will allow us to scale up operations. The new machine will have fast replicated disks, lots of RAM and two 12 core CPUs. Along with renewing the hardware, we need to revamp the software build infrastructure. Currently the development server runs a GitLab instance, Jenkins and pbuilder-based build jails. This combination was a big improvement from the custom made scripts of early releases, but it has some downsides that have been removed by sbuild. Sbuild is lighter and faster and has better crash recovery and reporting.
  • Trisquel 8.0 LTS Flidas
    Trisquel 8.0, codename "Flidas" is finally here! This release will be supported with security updates until April 2021. The first thing to acknowledge is that this arrival has been severely delayed, to the point where the next upstream release (Ubuntu 18.04 LTS) will soon be published. The good news is that the development of Trisquel 9.0 will start right away, and it should come out closer to the usual release schedule of "6 months after upstream release". But this is not to say that we shouldn't be excited about Trisquel 8.0, quite the contrary! It comes with many improvements over Trisquel 7.0, and its core components (kernel, graphics drivers, web browser and e-mail client) are fully up to date and will receive continuous upgrades during Flidas' lifetime. Trisquel 8.0 has benefited from extensive testing, as many people have been using the development versions as their main operating system for some time. On top of that, the Free Software Foundation has been using it to run the Libreplanet conference since last year, and it has been powering all of its new server infrastructure as well!

today's howtos

FOSS Events in Europe: Rust, foss-north, KubeCon + CloudnativeCon Europe 2018

  • Rust loves GNOME Hackfest: Day 1
    This is a report of the first day of the Rust loves GNOME Hackfest that we are having in Madrid at the moment. During the first day we had a round of introductions and starting outlining the state of the art.
  • Madrid GNOME+Rust Hackfest, part 1
    I'm in Madrid since Monday, at the third GNOME+Rust hackfest! The OpenShine folks are kindly letting us use their offices, on the seventh floor of a building by the Cuatro Caminos roundabout. I am very, very thankful that this time everyone seems to be working on developing gnome-class. It's a difficult project for me, and more brainpower is definitely welcome — all the indirection, type conversion, GObject obscurity, and procedural macro shenanigans definitely take a toll on oneself.
  • Five days left
    I use to joke that the last week before foss-north is the worst – everything is done, all that is left is the stress.
  • KubeCon + CloudnativeCon Europe 2018
    The Cloud Native Computing Foundation’s flagship conference will be taking place in Copenhagen from May 2-4. It will cover Kubernetes, Prometheus OpenTracing, Fluentd, Linkerd, gRPC, CoreDNS, and other key technologies in cloud native computing.

Programming: Taxonomy of Tech Debt, Python and More

  • A Taxonomy of Tech Debt
    Hi there. I’m Bill “LtRandolph” Clark, and I’m the engineering manager for the Champions team on LoL. I’ve worked on several different teams on League over the past years, but one focus has been consistent: I’m obsessed with tech debt. I want to find it, I want to understand it, and where possible, I want to fix it. When engineers talk about any existing piece of technology - for example League of Legends patch 8.4 - we often talk about tech debt. I define tech debt as code or data that future developers will pay a cost for. Countless blog posts, articles, and definitions have been written about this scourge of software development. This post will focus on types of tech debt I’ve seen during my time working at Riot, and a model for discussing it that we’re starting to use internally. If you only take away one lesson from this article, I hope you remember the “contagion” metric discussed below.
  • 6 Python datetime libraries
    Once upon a time, one of us (Lacey) had spent more than an hour staring at the table in the Python docs that describes date and time formatting strings. I was having a hard time understanding one specific piece of the puzzle as I was trying to write the code to translate a datetime string from an API into a Python datetime object, so I asked for help.
  • Getting started with Anaconda Python for data science
  • How to install the Moodle learning management system
  • Anatomy of a JavaScript Error
  • Is DevOps compatible with part-time community teams?