I'd wait until they update the app to run on the new iPad so that we can get more accurate results but this doesn't seem out of the ordinary to me. The CPU is the same, as is the memory speed, and the GPU is not tested - hence the similar score
Since Geekbench only tests CPU performance on calculations, the score will be similar to the iPad2. Unfortunately, no matter what kind of benchmark system you use, you will get different results that may not mean anything. You have to take a look at real-world speeds in order to see a difference. In the case of the 2012 iPad, the real wonder of it is the rendering speed. Rendering a webpage for example will be faster than the iPad2 just due to the extra graphical power. It's all about real-world benchmarks such as, if you browsed a website to gather information for a paper you are writing, which would get it done faster? A. iPad B. Netbook C. Mac D. Windows PC. It will also be a certain amount of trade-off performance that you will be looking at.
I've seen GPU benchmarks too (ones that looked legit at least). 50%+ faster than the ipad2 in real world tests, which likely means 2x faster or more in terms of pure GPU power. 2-3x faster than tegra 3 running the GLBenchmark off-screen tests, which are about the closes we can get to direct comparisons
I'm expecting very little improvement in game graphics as far as shaders/lighting/poly count/etc., with all the upgrades going straight into running things at retina resolution and the higher res textures to go with that.
Given that double the resolution is a big deal, that would be fine.. and actually good news for iPad 1/2 owners since it'd be unlikely for any apps to be iPad 3 exclusive for power reasons. And have fewer games not compatible with the iPad 1 then perhaps there otherwise would have been.
Of course thats just relatively uninformed speculation, who knows.