As for the environment (including the nearby existing buildings and so on) of one designed building in an architecture animation, when is it necessary to be modeled, when not?
When modeling the real environment, how to realize it? By modeling software like Maya, 3Ds Max, etc.? Or using an image-based modeling software like Canoma, Photomodeler? Or any other methods?
And should an architecture animation has a high resolution? How high at least then (pixel)? Does anyone have any good example of architecture animation including the real environment?
in my experience, if you are after out-and-out realism, you will spend vast amounts of time getting it right. it's a law of diminishing returns thing. unless you have a really good reason (instructions from a client and a signed cheque), you're going to want to reach some kind of compromise.
the reason i say this is that this decision effects how you model the site. it depends on what kind of stylisation is acceptable.
the cleanest way to do it would probably be to model the site in whatever modelling program you are familiar with. start by blocking in the surroundings, and slowly build up detail.
image-based modelling is going to give unpredictable, but possibly cool results.
resolution: it depends on what your animation is for. if you want to put it on the web, you need it small (320x240 or less), if it's going to be played on a fast machine in studio, you might want it a lot bigger (up to 1024x768). high-def video is about twice that again. the amount of data being pumped per second (and hence the amount of worrk the computer has to do) goes up exponentially with larger frame sizes. you can compensate by lowering the frame-rate, but anything below 20fps is going to be noticeably jerky.
high resolution = exponentially higher rendering times (think days, not hours). also, think about how you will compress your rendered animation. some codecs will shred your beautiful rendering.
the best quality would be something like 1024x768, uncompressed, 30fps. good luck trying to get that to play!
... these are only random notes. you're basically asking for rendering 101.
Wow, agfa8x, thank you so much! You've answered my questions so quickly and in depth.
One more question here, and I hope you don't mind :) ... If (only an assumption) one could quickly model the envionment (maybe this should be called reconstruction) with high resolution (like 1024x768), what kind of fields can this be applied, besides the architecture animation?
And what about if it has only resolution 320x240 or less? (I think you may say that the resolution doesn't matter, as it will depend on the clients' demand, right?)
you can make a digital model once, and render it as many times as you want, at high or low resolution.
the resolution is not set by the model, but is chosen when you render.
the resolution for your animation will be limited by the power of the machine you will be showing the animation on. the more powerful the machine, the higher resolution the animation can be.
there isn't a rule (not even a rule of thumb) for how big to render. it all depends on what it is you're trying to show.
There's another option...
You can make a bunch of high-quality still-renderings and turn them into a movie in after-effects/premiere. This allows you to make a 1024x768 30fps animation without going completely insame...
A really quick way to cheat with detailing out the surrounding site would be to make boxes to match the approximate heights/ widths of buildings and then map an image of the elevation on the surface. Ive done this in 3dmax with very good results...this is especially convinient for an animation as it shows enough details (depending on quality and accuracy of mapped image) without adding too much on rendering time.
Mar 31, 05 12:01 pm ·
·
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.
Any experience in real scene modeling for an architecture animation?
As for the environment (including the nearby existing buildings and so on) of one designed building in an architecture animation, when is it necessary to be modeled, when not?
When modeling the real environment, how to realize it? By modeling software like Maya, 3Ds Max, etc.? Or using an image-based modeling software like Canoma, Photomodeler? Or any other methods?
And should an architecture animation has a high resolution? How high at least then (pixel)? Does anyone have any good example of architecture animation including the real environment?
in my experience, if you are after out-and-out realism, you will spend vast amounts of time getting it right. it's a law of diminishing returns thing. unless you have a really good reason (instructions from a client and a signed cheque), you're going to want to reach some kind of compromise.
the reason i say this is that this decision effects how you model the site. it depends on what kind of stylisation is acceptable.
the cleanest way to do it would probably be to model the site in whatever modelling program you are familiar with. start by blocking in the surroundings, and slowly build up detail.
image-based modelling is going to give unpredictable, but possibly cool results.
resolution: it depends on what your animation is for. if you want to put it on the web, you need it small (320x240 or less), if it's going to be played on a fast machine in studio, you might want it a lot bigger (up to 1024x768). high-def video is about twice that again. the amount of data being pumped per second (and hence the amount of worrk the computer has to do) goes up exponentially with larger frame sizes. you can compensate by lowering the frame-rate, but anything below 20fps is going to be noticeably jerky.
high resolution = exponentially higher rendering times (think days, not hours). also, think about how you will compress your rendered animation. some codecs will shred your beautiful rendering.
the best quality would be something like 1024x768, uncompressed, 30fps. good luck trying to get that to play!
... these are only random notes. you're basically asking for rendering 101.
Wow, agfa8x, thank you so much! You've answered my questions so quickly and in depth.
One more question here, and I hope you don't mind :) ... If (only an assumption) one could quickly model the envionment (maybe this should be called reconstruction) with high resolution (like 1024x768), what kind of fields can this be applied, besides the architecture animation?
And what about if it has only resolution 320x240 or less? (I think you may say that the resolution doesn't matter, as it will depend on the clients' demand, right?)
ok, from the top...
you can make a digital model once, and render it as many times as you want, at high or low resolution.
the resolution is not set by the model, but is chosen when you render.
the resolution for your animation will be limited by the power of the machine you will be showing the animation on. the more powerful the machine, the higher resolution the animation can be.
there isn't a rule (not even a rule of thumb) for how big to render. it all depends on what it is you're trying to show.
There's another option...
You can make a bunch of high-quality still-renderings and turn them into a movie in after-effects/premiere. This allows you to make a 1024x768 30fps animation without going completely insame...
A really quick way to cheat with detailing out the surrounding site would be to make boxes to match the approximate heights/ widths of buildings and then map an image of the elevation on the surface. Ive done this in 3dmax with very good results...this is especially convinient for an animation as it shows enough details (depending on quality and accuracy of mapped image) without adding too much on rendering time.
Block this user
Are you sure you want to block this user and hide all related comments throughout the site?
Archinect
This is your first comment on Archinect. Your comment will be visible once approved.