Nuclear Monster

  • Dave James reviewing the new $999 graphics card from Nvidia:

    Inside that chunky chip are 24 streaming microprocessors (SMM) in six graphics processing clusters (GPC). With the Maxwell design running to 128 CUDA cores in each SMM that makes for a grand total of 3,072 cores in the GTX Titan X. Completing the core configuration are 192 texture units and 96 ROPS.

    That’s a whole heap of graphics processing power right there.

    Nvidia have also doubled the size of the frame buffer compared to the previous Titan cards, maxing out at 12GB GDDR5 memory, running across six 64-bit memory buses to deliver an aggregated 384-bit total memory bus.

    That memory capacity might well look a little bit like big numbers for the sake of it, but we thought it would be a long time before the original Titan’s 6GB frame buffer was anywhere near fully utilised. Yet right now Shadow of Mordor is filling up around 5.7GB with the HD texture pack at Ultra 4K settings; we may only be a couple of years away from 12GB actually getting used. Right now, 12GB is more future-proofing than anything else.

    I’ll take ten.

  • Mike Fahey:

    This morning Nintendo announced a partnership with Japanese company DeNA to produce games using Nintendo properties on mobile platforms. As the owner of mobile platform Mobage, DeNA is one of the most powerful names in mobile gaming, but the sort of games it’s known for won’t make traditional console gamers happy.

  • Chris DiBona announces the shutdown of Google Code:

    When we started the Google Code project hosting service in 2006, the world of project hosting was limited. We were worried about reliability and stagnation, so we took action by giving the open source community another option to choose from. Since then, we’ve seen a wide variety of better project hosting services such as GitHub and Bitbucket bloom. Many projects moved away from Google Code to those other systems. To meet developers where they are, we ourselves migrated nearly a thousand of our own open source projects from Google Code to GitHub.

    As developers migrated away from Google Code, a growing share of the remaining projects were spam or abuse. Lately, the administrative load has consisted almost exclusively of abuse management. After profiling non-abusive activity on Google Code, it has become clear to us that the service simply isn’t needed anymore.

    Beginning today, we have disabled new project creation on Google Code. We will be shutting down the service about 10 months from now on January 25th, 2016.

    There are a ton of abandoned but still useful projects on Google Code, most of which will be lost after 2016 if nobody clones them and puts them online somewhere else. At least an Export to GitHub button on every Google Code site now.

    This is your continued reminder that Google, and start-ups funded by VC money, are not a safe place to store your work. Own your shit before GitHub starts inserting malware into downloads or sells out in some original and disruptive way. Get a domain, some shared hosting, maybe a Linux or BSD VPS if you’re rich. With git it is easy enough to move a project if you have cloned the project locally and have established a web presence that people can check for updates. At the very least, don’t make the GitHub page the public-facing home for your project.

    Even Google isn’t stupid enough to put their most important projects on another company’s servers:

    Google will continue to provide Git and Gerrit hosting for certain projects like Android and Chrome. We will also continue maintaining our mirrors of projects like Eclipse, kernel.org and others.

    You can be sure their internal code for things like search aren’t hosted on GitHub, either.

    Donate to the Internet Archive.

  • https://www.youtube.com/watch?v=-v2mvO7Yq48

    How did Dinosaurs go extinct when they could rap like this?

  • Valve recently put up an early list of Steam Machines coming this November from various gaming computer makers. Paul Tassi of Forbes is here predicting doom and gloom in an article comparing the chances for Valve’s success against traditional gaming consoles:

    So if Steam Machines aren’t for console players, will existing PC gamers bite? I really don’t see a reason for them to do so. PC gamers like their mouse and keyboard, their ability to sit close to their monitor, and the probably multi-thousand-dollar rig they already have. The avid PC gamers I’ve seen look at Steam Machines as “good for people who want to check out PC gaming,” but almost none of them seem to be considering it for themselves. And why would they? It’s just a pre-built gaming rig that hooks up to their TV and runs Steam Big Picture, something all the do-it-yourself-ers out there could have made themselves for years now if they really wanted to. With the release of the new store, many veterans of the scene are looking over some of the machines and laughing about the price, knowing they could get the exact same level of performance for much, much cheaper if they did it on their own. Other than the ability to express their endless love and devotion for Valve’s Gabe Newell, patron saint of PC gaming, I don’t know what PC gamers get out of Steam Machines.

    This is but one of the many arguments Paul Tassi has that they won’t find their market.

    It’s a bit early to predict the death of Steam Machines, eight months before they ship. Among Paul Tassi’s other arguments is this one:

    Why? Before I flood your screen with a deluge of reasons, first and foremost what the new store page shows is a huge range of prices, ranging from slightly above asking price for the PS4 and One ($460 for iBuyPower’s box) to typically absurd gaming PC levels (as high as $5,000 for top-of-the-line units from Falcon Northwest and Origin). Across the fifteen(!) different machines on the page, only two are around the $500 mark, while the rest probably average between $700-$900, if I’m being generous.

    There is definitely a large selection and I can see how it might steer some people away and back to their consoles. It makes sense to have a small selection of products that people can choose from. A good machine, a better machine, a best machine.

    That is the one valid criticism in Paul Tassi’s article. The others are ridiculous. PC gamers aren’t going to buy these? If some weren’t already buying pre-built gaming machines these companies wouldn’t be in business to now sell Steam Machines.

    There are other, more valid, criticisms that Paul Tassi misses.

    These machines run Steam OS, and although you could replace that and install Windows at your own expense there will be a more limited selection of games available for the default Linux-based operating system.

    Though he does point out that they’re more expensive than consoles, the reason why is extremely important. Consoles are subsidized and initially sold at a loss to be later offset by software sales with a cut going back to the manufacturer. Valve isn’t selling these directly and can’t do that. Subsidies are the one way that Valve could really improve the initial turnout on Steam Machines but then they would have to produce their own and be competing with these third parties.

    The advantage to Valve in the current Steam Machine situation is that they lose almost nothing if Steam Machines fail to find users.

    The one thing Valve loses if Steam Machines fail is their edge against Microsoft. Steam Machines (and SteamOS) exist as a hedge against the possibility of a strong Windows App Store in future versions of Windows where customers won’t or can’t choose Valve’s Steam marketplace.

    The computer manufactures have something to lose too. With Steam Machines they don’t have to deal with Microsoft and pay for copies of Windows. As far as we know, Valve charges them nothing to include SteamOS.

    This is the reason why Steam Machines aren’t really here to challenge consoles. They’re here to take on Windows.