XNA: The Power and the Potential

Following recent comments from coding shop Binary Tweed, the Xbox Live Community Games project appears to be in dire peril due to a combination of factors. The Clover developer says that the size of the XBLCG market can't sustain the required development costs, while gamers might argue that there's little point visiting the portal at all when the quality of content is, by and large, seriously sub-par and in many cases, somewhat over-priced. A 400 point virtual fireplace any one?

Put very simply, the tools give hobbyist programmers and pro-developers alike a rich resource of coding potential. XNA effectively allows for full access to the Xenos GPU along with four of the six processing threads on the main CPU. While it's always difficult to draw analogies in processing and GPU power, development sources speculate that XNA performance on the console is roughly equivalent to the Intel Atom CPU commonly found in netbooks, but with the added support of a NVIDIA 7800/8800 class GPU. Not a bad combination, and a whole lot of horsepower for a USD 99 buy-in.

In terms of the XNA engine itself, one Xbox 360 and PC developer described it to me as being "well thought through, showing experience and pedigree. Most engines I've seen at fellow PC studios are much worse in terms of architecture on the elegance/sloppiness scale."

Of course there are drawbacks. Available memory is an ever-present issue, especially acute for developers used to working with PC. The C# programming language used compounds this problem by serving as a "nanny OS" that takes care of memory management itself. Coders call this the "Garbage Collector", the in-built controller that reclaims memory used by data that is no longer being used by the code.

Despite these limitations, let's attempt to give some context here. XNA should easily be able to cope with a massive range of not-so-old PC titles and there is no reason why any PSP, Wii, iPhone or Flash game should not translate beautifully into a high-def XNA game. And with that in mind we approach the crux of the Community Games problem from the gamer's perspective. With a few exceptions, the content is, by and large, a waste of time and nowhere near close to realising the potential of the platform.

So, bearing in mind the richness of potential in the tools, what can be done? Perhaps Microsoft should consider opening up the community games portal to match the length and breadth of scope inherent in the iPhone App Store. Why not allow proper internet access? Why not get Google Earth running on Xbox 360? Build a range of must-have free content, not necessarily exclusively games-based, and then you'll enough of an audience so that development of more impressive software could perhaps be financially viable.

As it is, the portal is in desperate need of finding its own distinct identity, simply because the alternatives – full retail games and the Live Arcade – offer so much more value and enjoyment. Without a unique selling point, better marketing and irresistable content, XBLCG is threatened with extinction, a massive shame bearing in mind the power and the potential of the tools on offer.

Comments (10) Latest comment 3 years ago

Comments threads automatically close after 30 days, but please feel free to continue chatting on the forum!

  • Calgon #1 3 years ago

    Unlucky being hit with the double RROD Richard, by "fixed" I take it you mean "refurbished" then?
  • Crazyreyn #2 3 years ago

    I agree - XNA could become a console App Store. After all, you can only get a game rejected if it's technically unplayable or has offensive content. Kodu is a great start in bringing attention to the service, but MS could do more in highlighting what exactly this can do for gamers.
  • MeBrains #3 3 years ago

    that really is a shame.

    every since the days of the C64, subsequently the Amiga and ST, I have thought that programming is beyond reach of too many a child, while I started when I was 12 (speaking dutch and needing to learn english). Only with easy-to-access platforms is there a slight chance that kids pick it up early.

    then again: I learned it because of necessity. There were no easy to get to games. There was not the overdose kids have now...Why invest time when you can download them in an instant. I learned from copying code from magazine to machine.
    Edited by 1 at 08/07/09 @ 21:00
  • seasidebaz #4 3 years ago

    Sorry, this is the worst article I've ever read.

    XNA is NOT an engine. It's a framework. There's a big difference: you need to build an engine on TOP of XNA to see any real performance.
    Shawn Hargreaves, one of the people who made XNA, has said previously that you can get equivalent performance out of something built with XNA as you can from the proper XDK. And if you design and code properly, the garbage collector isn't an issue. The documentation even tells you when the garbage collector kicks in.
    As for memory, you have 512Mb total to play with. The filesize limit on a community game is 150Mb. Unless you assign masses of unnecessary data, you should never run out.
    And finally, it's very silly to compare a quad-core PowerPC chip (which is what you have acces to) with an Atom processor.
  • Jabjabs #5 3 years ago

    Just a quick correction to post #4, 360 runs on a tri core processor.
  • seasidebaz #6 3 years ago

    @Jabjabs:

    3 cores with 2 hardware threads each, so in essence you have 6 hardware threads. I meant to say it's LIKE having a quad-core, as you can run 4 threads in parallel on XNA. Thanks for pointing out my mistake though. It's still far more power than an Atom processor, which doesn't have any hardware multithreading support.
  • Alkeno #7 3 years ago

    I believe the biggest problem XNA has to face is competition from retail games and mostly, Live Arcade games. Many small developers (or lonely guys) that try to develop something tend to look up to other major developments on the 360 and moan I can never get anything near that type or performance/art/polish. It can be discouraging.

    On the iPhone, on the other hand, most games are tetris clones and simple applications (goes with the nature of the system, it's a phone, I believe high-end hardcore titles belong to PSP). It's easy to feel motivation to build an iPhone game, because you know that if you put in enough weeks you can come up with something that doesn't scream amateur turf.

    At least it happened to me, I downloaded XNA a year ago and began toying around, but quit, not much motivation. Now I'm starting to develop for iPhone and feeling compeled to at least trying to bring a 0,99 game into the market to call my own...
  • zooms #8 3 years ago

    I hope we will see one break through title which will raise the profile of the Indie ( Community ) games channel. A few decent games will make all the current dross seem totally unimportant.

    I would love to see professional developers somehow start to allow game mods via XNA - it would have all the benefits that developers currently feel they get from a moding community but within the safe confines of XNA.
  • dr.apocalipsis #9 3 years ago

    (...)but with the added support of a NVIDIA xxxx/8800 class GPU. (...)

    Just not true.
  • NeoTechni #10 3 years ago

    "While it's always difficult to draw analogies in processing and GPU power, development sources speculate that XNA performance on the console is roughly equivalent to the Intel Atom CPU commonly found in netbooks, but with the added support of a NVIDIA 7800/8800 class GPU."

    Which is pretty much the opposite of PS3's Linux, where people PS3's as a supercomputer, with out a GPU. Though people have SDL running on the SPUs to make up for it. The article mentions XNA on 360 has no internet access. Linux on PS3 has access to everything except the GPU.