| Machiavellian said: Having everything compressed keeps it effecient. No wasted time encoding and decoding thus improving performace. As everything has a cost, depending on how many jobs are being processed and sent over the wire, the nano seconds in encoding decoding can mean a lot if you have 100s of jobs being processed. I do not see how memory will be an issue with 8GB. Even if the OS takes 2 to 3 still gives a lot of memory for use. If the data is compressed it also helps to keep memory available. Non immediate stuff can be temporarly stored on the HDD. When I write web applications, I am always prefetching information from databases and caching them before users actually select and object which will be used. SImple things like Google Autocomplete is one such thing that I use when creating dynamic Eforms for line of business scenerios. For some queries hitting databases, it could take some time to filter through information so its always best to work with a local cache to improve performance. As to your point about having an instance copy ready, you should take a look at MS Orleans Platform. Here is a link from MS Research. http://research.microsoft.com/en-us/projects/orleans/ MS stated that for each X1 sold, developers will have the capability to leaverage 3 X1 compute power in the cloud. Basically this means that Developers will have a persistance instance of their code available to leaverage for cloud compute. The Orleans is one project that is created to provide that type of distributed, persistance resource which was in development since 2008. Read the part from the Programming model and the grains concept to see how MS has built a platform to solve that solution. Current the Halo team is using the platform so we might actually see real world product using the cloud compute from them. You are right that this type of setup is extreamly complex and not something that was thought up in a few months because of Sony reveal. On a geek from I am very interested to see how this technology will play out and if it actually will be the next coming as MS is prone to hype. |
I wasn't arguing the usefullness of working with compressed data, but the particular use of standardized compression tools as proposed in the move engines. It makes sense for web applications, http requests and responses, form data, etc are terribly wasteful in memory size. It is a lot less useful for binary game data though, especially with properly optimized (or already compressed) data structures.
It's the same vibe I get from the New Orleans platform. An abstract layer build for easily scaleable business web applications. Applying it to games, especially single player games seems a bit of an extra money grab. As if someone came up with an idea to make more money of the Azure platform by offering it to enhance games. Charge developers to use the web hosting, charge players the live gold fee for enhanced single player.
From a game development standpoint it's great to have to have access to an already established widely distributed server network. But the grains themselves still need to be build from the ground up. Developing a fully fledged server side version of the game (which probably can't rely on specific hardware and needs to be ported to the available apis) is a huge undertaking for a couple of enhancements.
8gb ram seems like a lot now (or 5 or 7 whatever is left after the OS) but so was 512mb in 2004 after the ps2 with only 32+4mb available. Yet it didn't take long before developers ran out of memory again. (Remember Bill Gates' famous quote in '81 640kb should keep everyone happy for the next 10 years...)
In my experience ram is always the bottleneck, together with read/write speed. Working directly with bitstream compressed optimized data and letting the processor do some extra work is preferable to sacrificing memory and bandwidth to pre storing too many things. I don't see it being very practical for preparing light maps or complex animation sequences. More useful for mmo type settings where you offload memory and caching by letting the server keep track of the world and send you the relevant cell data when needed. Yet that's not applicable to single player games that need to keep working when the connection drops.
The first games won't have any memory problems ofcourse, they won't max out the 8 processor cores either. So I wonder what (if anything) they'll show at E3 for cloud computing enhancements in single player games, and whether that couldn't be done locally with a bit of optimizing.
