Tom’s Hardware ran a great article today comparing AMD HD3D and Nvidia’s GeForce 3D Vision software solutions. The intent of the piece was to determine which of the software/hardware mixes offered the best bang for the buck in 3D video game compatibility and software support. However, what caught our eye were their remarks about MTBS’ 3D Game Analyzer:
We originally considered reporting the MTBS3D certification ratings for this article. MTBS3D.com is an independent 3D certification and advocacy group, which provides an online tool that asks users questions about their 3D game experience. We ended up deciding against this because the rating assigns a significant percentage of its final score to the presence of 3D convergence control. Consequently, a game with flawless 3D reproduction but no convergence control can have a lower score than a game with significant visual anomalies. We don’t think this scoring equation yields ideal results, but we do give MTBS3D props for at least trying to address the state of 3D.
We are very flattered that Tom’s Hardware (TH) took notice of M3GA, and we think it’s important to clarify why the analyzer works the way it does, and what we have in mind for the (near) future.
Tom’s Hardware is correct that our rating system places great importance on convergence controls. Convergence is the setting in stereoscopic 3D drivers and some native S-3D games that determines how much of the experience is inside and outside the screen. When used properly, it marks the difference between a memorable 3D experience, and a tedious “been there – done that” video game.
According to U-Decide 2011, over 76% of experienced stereoscopic 3D gamers want a combination of depth and out of screen visuals in their FPS 3D games at varying degrees. Unfortunately, it’s not easy for all games to accomplish this because convergence controls open the door to camera angle problems and visual anomalies which take time to root out.
To rectify this, the driver developer and/or the game developer will occasionally lock out the convergence controls to save on production time. When 2D+Depth (e.g. Trioviz, Crysis 2) technologies are used, effective convergence is nearly impossible because flaws would be too revealing and harm the experience. More often than not, the inability to have convergence controls is less an artistic choice, and more a limitation of technology or production time.
In the case of M3GA, we have not one, but two methods of rating games. The first is a direct relationship between visual flexibility, required eye candy reductions, and remaining anomalies. So a Platinum game has full visual flexibility, full eye candy capability, and no anomalies. The bottom of the four stage scale is Bronze, and these games have no convergence control, are depth-only, and may or may not have visual anomalies.
The second scale is numeric and works completely differently. Games start with a score of 100, and lose points for visual anomalies and eye candy reductions. There is a deduction for not having convergence flexibility, but it is far more forgiving. This is why the numeric score can seem higher than the award rating. It’s not an error – they are just completely different methodologies.
We think it’s important to maintain both scales because the first gives a snap-shot description of the game’s quality and visual flexibility, and the second gives an alternative method of rating the game for those preferring a detailed QA format, and aren’t as picky about convergence flexibility.
Now here’s the kicker! The current implementation of MTBS’ 3D Game Analyzer is out of date. While the existing database will be converted and maintained, the new service due out soon will be far more user friendly and fully featured. We have been giving private demonstrations to industry and gamers alike, and the response has been very positive. We’re even having the service renamed with its own URL!
While we encourage visitors to check the current analyzer out and submit games where possible, this is eons behind what is currently being developed. In addition to being much more user friendly, it is going to have most if not all the features that members have been begging for. We hope you are as excited as we are.