[GDC 2008] We were invited to attend an interesting discussion held by the PC Gaming Alliance, a group of companies that includes Acer/Gateway, Activision, AMD, Dell, Electronic Arts, Epic, Intel, Microsoft and NVIDIA. The Alliance has been created to solve one problem: Make PC Gaming easier to “consume” and extend the reach of PC Gaming beyond veteran/hardcore gamers.
What is the problem?
Unlike game consoles, the PC is an open platform with a multitude of hardware and software choices. This is a strength when it comes to high-end gaming systems because you always get the latest technology, but it is also a weakness because it confuses consumers who buy multi-purpose computers. There are examples of 2000 dollars PCs that are not capable of decent gaming and from a consumer’s perspective, it’s ridiculous – most would expect a $2000 PC to be somewhat gaming-capable. On the software side, games all have different minimum requirement specs (and sometimes completely bogus ones).
Having worked in the PC graphics industry, it is not hard to see where a big chunk of the problem is: integrated graphics. What is it? Integrated graphics is the graphics processor that comes attached to your motherboard. It is very low-cost, and has mediocre gaming performance. More than half of the computers sold use integrated graphics – including expensive ones. Ironically, it costs only about $12 to $20 to go from poor gaming (integrated graphics) to entry-level gaming (low-end GPU). On the CPU (processor) front, mid-range computers ship with a decent processor, so I don’t even consider this to be a real show stopper right now.
Why is it that way?
The current situation is the legacy of decades of selling computers to IT departments (and tech savvy customers), based on features: CPU speed, Memory capacity, Hard disk capacity and so on… That worked back then, but that’s not how many consumers shop: you have to sell the functionality (“what it does”) and not the feature set (“what it is”).
Many PC buyers don’t even know that they should upgrade their graphics when they buy a relatively expensive computer. After all, the integrated graphics is often Intel’s “Extreme Graphics”. People think: “wow, it’s extreme, so it must be good, right?” Wrong! It’s extremely… poor for gaming – simply because it was not designed for that purpose. The line that I hear is: “anyone who cares about gaming knows that integrated graphics is lame”. Wrong! Again, many buyers do not even know what integrated graphics is.
How to fix it?
The group has yet to define its policy, but here’s our take on it: it is not as hard as it seems: right now, “bad gaming experience” mainly comes down to a lack of graphics horsepower. In my opinion, the low hanging fruit is a scoring system of some sort, based on current games performance. PCs with crappy graphics should be labeled “gaming level 0”. Let’s say that, today, the highest level of gaming would be Level 5. Future hardware would be 6 and beyond. Accordingly, games should be labeled with the appropriate “gaming level”. From there, the “gaming level” could be communicated like any other feature: RAM, Mhz, GB. It’s not perfect, but it would be a far cry from what we have today: walk into a computer store and most of the time, the graphics capabilities is not even mentioned on the specifications sheet.
How to communicate it?
Someone in the attendance proposed to have a logo. The idea was somewhat discarded by the group, because it was deemed “too expensive”. That’s a typical answer from engineers, and there lies a small part of the problem: this is all driven by geeks.
First of all, it is also hard to believe that Acer/Gateway, Activision, AMD, Dell, Electronic Arts, Epic, Intel, Microsoft and NVIDIA combined can’t afford a logo. Secondly, if a logo is easier for consumers, go for it! At least, stay open minded.
Sounds too easy?
The idea of having a clear segmentation in a (PC) world where everyone struggles to differentiate is not going to be easy. Look at how many CPU there are, or at how many different graphics cards and motherboard there are. There will be a lot of resistance from board manufacturers that will be afraid to be “commoditized” by a standard.
Conclusion
In the short term, “bad PC gaming” mostly equates to “poor graphics performance”. As a consumer who is interested by gaming, try to be sensitive to this and avoid integrated graphics like the plague. The difference between “awful” and “good” gaming performance is often less than $100 away, and great graphics cost only $200.
The good news is that integrated graphic will get a lot better, a lot sooner than you think (we have already published about an integrated graphics processor capable of running Crysis). Better integrated graphics alone will solve a lot of problems.
How do you perceive the problem, and how would you solve it?
Disclaimer: I do not own shares of any of the companies mentioned in this article.