Because I can’t get it to properly work. This computer has 4G RAM and dual 2.73GHz processors. It’s more than fast enough to handle just about anything Google Earth can throw at it. But when I try to load Ancient Rome, which really should just be another city, it slows down to a crawl. If I click on one of the yellow building icons I don’t see the building in ancient time, but a popup photograph, admittedly of high quality, but now one that lets me “fly” anywhere. Instead there are the same links to load ‘terrain’, basic set of ‘250 monuments’, and ‘5000 buildings’. But nothing seems to really work. If I try to zoom in on an area of the city, I get Street View camera icons that lead to contemporary urban scenes.
Can I get this to work, or do I need more money, or more processing or memory capacity? So far I"ve done better with the phototours of some of the contemporary sites, some of which provide extroardinar levels of detail.
Well, I’ve got a 2.6 GHz dual core, 2 GB RAM and a broadband connection and it loads slow as hell for me. I don’t just mean slow, it damn near crippled my computer. I have a crappy video card, however, so that might have something to do with it.
I took a screenshot of what it looks like after 10 mintues of loading:
Just WAGging my ass off here, but normal Google Earth images are just that - images. Easy to load.
Based on the pic that gladtobeblazed posted, every single building in there has to be rendered by the computer. I don’t know if the 3D information sent through the 'net is bigger than an image file, and perhaps this is where the bottleneck occurs? Or, as gladtobeblazed stated about having the crappy video card, maybe that’s what’s causing the slowdown…
Still, in most American and Kiwi cities, plus some in Australia, and a few in Europe, you can zoom down to Street View and that takes hardly any time at all. Other major world cities usually provide numerous 3-D buildings and oblique aerial views.
Similarly, oblique aerial views of New York, Los Angeles, or San Francisco, with most famous buildings clearly identifiable and realistically rendered, take less than a minute to load. Why does Ancient Rome take 10 minutes? You’d think they could just take some images of that, and then process it any other city.
Heck, even non-famous buildings get quickly rendered!
Isn’t that obvious? Isn’t it obvious that a 3D representation of an object carries vastly more information than just a photo of it?
I remember my computer slowing down considerably in the center of Washington DC with the White House and other monuments and buildings being in 3D. I just turned that feature off.
Not necessarily. It depends on the complexity of both. A high-res photo could conceivably be larger in size than a simple, untextured (or simply textured) 3D model, which is what Google’s Ancient Rome looks to be. Compare this 3D model of the White House (1 MB) to this high-res picture of it (~3 MB), for example. Of course, if it’s a 3D model with equivalent texture resolution of the photograph, it’ll be much larger.
A simple 3D model is often just a collection of points, polygons, and curves – it doesn’t take much space to store the data; it’s the rendering process that’s usually slow.
Yup, that’s what I should have said. Rendering 3D takes a lot more processing power. And every time you move the observer’s point of view you need to recalculate the entire thing which is why games require so much video processing power.
Once you get it loaded, does it perform well? Is there any way to save newly acquired data from the Ancient Rome app so it isn’t so slow the next time?
