I've recently met Vilmos Zsigmond ( http://www.imdb.com/name/nm0005936/ ) recently when he came to town for a screening of Close Encounters and did a lighting demo with Q&A. He went on at length about film vs digital. He essentially said with digital his job really don't change. It's as simple as working with another film stock. However he does prefer film because he felt digital was actually TOO sharp.
There is no film stock that lets you shoot both 24fps and then shoot thousands of FPS under the same lighting conditions. I suspect there aren't any film cameras that do 1000+ fps AND 24 with the same motor. You have to have the right tool for the Job. So you switch to a Phantom Flex type camera for your real slowmo stuff.
I've been told the current method for archival is actually just that. They use 3 B&W stocks to separate the color because B&W is far more stable and last longer under ideal conditions.
The only reason film is still used in Hollywood productions is because it's currently cheaper to do so. Their entire workflow from shooting to archival to restoration is based on using film in key spots. The studios are in it to make money. When an all digital workflow becomes cheaper than their current model then you bet your ass they will switch over.
I'm a student film maker in an experimental (focused more on art house than typical Hollywood) film program. There are tons of students here madly in love with film for various reasons. Most of them are purely sentimental and no doubt unable to tell the difference if you put them side by side. I'm not one of them. I prefer the advantages of digital.
However, there are some legitimate reasons to prefer film. Digital workflow has it's downsides. It can be more complicated to get started editing with. Film you can just cut and tape together. You can hold the medium in your hand and see how it all works together. Some people prefer something they can physically touch. It can be a more enjoyable process for them to work with and problem solve with.
Digital can require significantly more complicated just to get your footage to play nice with your NLE software. Also, being able to see your image instantly can give students the impression they can cut corners in planning stages. When you can't see your final image until days/weeks later after processing it really forces you to make sure you plan everything out in more detail. You just have understand everything going on to avoid mistakes that will cost you both in time and money.
You don't need xyz information. The objects data is in a tree structure. The root has some XYZ position and everything else is relative to it. It's organized in a fashion where each level of the tree contains more fine detail. So you can calculate when the detail is smaller than a pixel and stop rendering. Chance are you never need to store the entire object in memory unless you have a very high resolution display.
This engine supposedly doesn't use voxels but I suspect it's something very similar. Check out these videos for a better explanation
It would be a huge time saver. Pulling focus gets complicated quickly. Not having to worry about it is one less thing to have go wrong. You can do less prep work, less takes, and require less staff. Save $$$ and now complicated things become more approachable for a lower budget film.
Also, having that control after the fact in post production would be great. Got a poor performance from an extra in the background? Well, now there's a very simple way to make it less noticeable. A shot is too confusing and the audience gets confused? Well, now you can guide them where to look.
I guess I was trying to say you don't necessarily need grammar for a language used by computers. Grammar for them is just a hack, or addon to allow a language to communicate more than the originally was intended.
Humans don't want to reinvent the wheel every time we need to expand our language and thus grammar works well for this. Computers don't have that issue and so grammar (at least as we know it) isn't important.
I was just trying to point out that having a grammar isn't required for a language.
Well, depending on the number of communications they need to make to each other it's very possible 4 level words could map out every possible communication they could have with it each other.
Think of it like Chinese symbols.. They aren't just one word but complex ideas.
Grammar exists because we are unable to store such large amounts of data. We can't have 1 word/symbol map to a unique complex concept. A computer might not have such limitations. Especially if their entire universe of ideas/concepts can be enumerated easily.
We submitted our (The Amateur Monster Movie http://www.imdb.com/title/tt1717690/ ) zero budget film to be added IMDB a few times in the past while still doing principal photography. They denied it until we released our trailer and started getting more press coverage. We don't have any distribution deal (yet) or submitted it to any festivals.
I think if they just released a trailer and got more press (which Slashdot should fix) they will be added. I think it's a matter of them simply trying to avoid adding films that most likely will never be seen by anybody but the people involved in creating them.
Here is a link to our trailer in case your curious.
My understanding is that the sharpest part of the lens is in the center. So the crop factor actually reduces the vignetting effect of a poor lens.So a crappy lens might become usable when combined with a smaller sensor.. It's all a trade off anyway. Smaller sensor = lower cost to manufacture but lens can potentially be more expensive. You just need to find the sweet spot.
Right now full frame seems like the way to go because it's easy to find cheap but good new and used lens. However, not too many cameras (in video) offer such a solution.
I've been watching the Apertus/Elphel project for some time now. I too wanted a larger sensor to allow a nice shallow DOF. However, the more research I do the more I realize it can still be achieved. The digital cameras that were used to shoot the Star Wars prequels were 2/3" and they achieved a very cinematic look. The C-mount means you can get lens that are capable of doing this at a fraction of the price.
Really, the only true advantage of a larger sensor is having potentially larger pixels. This allows each photosite to capture more light and thus work better in lower light conditions. Think of a bucket capturing water. The larger the bucket the more water you can get.
Now, right now nobody seems to be manufacturing a full frame sensor that does what you need and is cheap enough. Once they exist I'm sure the Elphel/Apertus team will start using them.
Here is another interesting camera project that seems to debunk many of the perceived size sensor limitations. It' called "Drama" and is designed to be a camera capable of uncompressed video. http://dvxuser.com/V6/showthread.php?t=219424
Finally!!! A news article I can slip in a promo for the zero budget, comedy/horror film I've been working on over the last year. It is called "The Amateur Monster Movie" and was shot in and around Milwaukee Wisconsin. You can view our trailer on Youtube via the link below. Let me know what you guys think.
If Pi is infinite and has no discernible patterns then wouldn't that basically fall into the infinite monkeys and typewriter category? If you look long enough you will find any encoded message you want because it contained every possible coded message?
In a brain like system we provide the stimuli which effectively act as seeds, but depending on other, non-deterministic events we can't guarantee that it will be the same.
I guess I don't understand how the non-deterministic nature really has any affect on it's equivalency to a computer. Like how a DFA and NFA are shown to be equivalent.
I've just recently taken my first AI course but unfortunately it just turned out to be a very boring algorithms course. It wasn't nearly as comprehensive as the required (for CS degree) algorithms course either. I am very interested in the subject and know just the very basics of neural networks but do stumble on a few articles about them from time to time
Can you recommend any books on both neural networks and the brain itself?
I agree it's pattern matching but I also consider that to be a mathematical process.
A machine/computer can only be as accurate as it is because we isolate it. We limit the number of things that can go wrong. A brain doesn't have that luxury. Practice seems to be learning to isolate a task from unknown and unrelated external forces/input.
I see no reason why a brain can't be trained to perform the exact same task as a computer/machine being as accurate and consistent. It's just we don't do it. We find short cuts. We make tools. If we had to do it that way our brains would eventually adapt to do just that.
I think there is numerical analysis involved it's just we don't directly have access to it. Our conscience mind doesn't have root access and thus is limited to some sort of API and there is private data structures I simply can not access via my conscience mind.
True, in the case of the basketball we only limited feedback from our throw function. It returns make/miss overthrow/underthrow type results. However, even with our initial throw we have some basis for how far we can throw something. We might not think of it directly in terms of feet but there is some built in relation to some unit of measurement in our heads. This basic unit is what guides all internal calculations/conversions.
This is how when somebody says toss the ball twenty feet I can get it close to twenty feet. I have some idea of what twenty feet is as a unit of measurement. I have some idea of how far I can throw is ball. I have some idea of how much force I need to exert to throw it twenty feet. Sure it's not going to be dead on accurate/consistant without extensive practice.
You might say that you could build a robot to throw the ball exactly twenty feet every time. Sure, you can factor in things like wind, elevation, etc etc. However there is always going to be something else that could cause your robot to fail. It could just fail mechanically over time. I could unplug it or just push it over. Or some other influence simply not accounted for.
The reason our brains can't do these things accurately is because we don't need them to. The time and effort to train yourself to do such a thing is insanely large compared to building (or in modern times) buying a tool to achieve the desired results. If a tool didn't exist then our brains would develop the ability to do just that over time. It might not happen in a single generation but it would happen.
it's quite easy in computers, because we have explicit memory addresses
Yes, it's true you can find something quickly if you have the memory address. However, you still need to know the address of what you want. A dictionary is a great way to verify the spelling of a word but you still need to have a good idea of how the word is spelled to find it.
A computer could simply check every address via brute force to find what it needs but that isn't generally what happens. I assume our brain works somewhat similar to a database. You make a query and get results. Those results can be used to make another query and so on and so forth until you may eventually find your desired result.
Where are my keys? I dunno... Is it where i normally put them? No. Let's retrace my steps.. Found them. I don't have a specific "address" to find my keys but I can usually locate them via a similar method. However, if I always put my keys in the same place then I have the location already. Like a computer I have the place to find them and it's only a one step process. However, as things get more complicated you simply can't juggle that many direct addresses to information and nor does a computer.
True, we don't say oh I left my keys at 0xffa12b012 but that is because our conscience mind doesn't have root access. It has to access everything some sorta limited access API. I suspect we will continue to hack our brains until we gain such access.
Being able to throw a ball right some, or even most of the time may be fine for a game of basketball, but is this ability good enough to calculate the values for some complex and precise engineering application? Absolutely not.
I would argue that are brains can and will. It's just a matter of time before they naturally evolve to do such things. I believe autism/savant types are simply evolution trying to give us those abilities. However, I wouldn't downplay the precise engineering that goes into things like sports or arts. It's just our brains were able to attempt to these easier as they relate much closer to things we already can do well.
It will be.. We just haven't gotten there yet. Our brains will evolve to do such tasks it's just we haven't needed them to do it.
Again, absolutely I agree, but the issue is consistently, you may be able to train it but when it's so vulnerable to minor changes in the way it works, can it do it consistently?
Well, computers seem consistent but they do fail. It could be a software bug, it would be a hardware bug. Given enough time bits will randomly flip and may cause inconsistency in calculation. A computer flips a sign bit and suddenly 1 * 9 = -9. Sure some (or even many) of these can be avoided by adding error detection/correction but it will never catch them all. Perhaps a human brain is doing significantly more calculations than modern computers. I'm guessing the probability of random bit error in the brain/computer are close and it's simply a matter of the brain doing significantly more calculations per second that it happens frequently.
I think I've decided to no longer consider their brains severely damaged. Their brains seem to be wired in a manor that puts communication (and other things) at a lower priority. The link below is a video demonstrating how a person originally classified as severly mentally disabled was simply unable to communicate in the manor we are accustom to.
http://www.wimp.com/autisticgirl/
I think it's simply evolution. While these people who never last in the hunter/gather world in modern times we care for them and in some cases they are able to lead "normal" lives and reproduce. These traits get past on and eventually perfected to the point where it isn't such a resource drain on their brain. Now, they can pick up those other skills like being able to communicate and function "normally" but with a new specialized part of the brain.
but hopeless for calculating with a reasonable degree of accuracy the actual distance to that object- the margin of error for most people is on average going to be quite large.
I disagree. How can we learn to throw a basketball into a tiny hoop from far away without having very accurate estimates? Think of any sport and just how many good estimates are done VERY quickly and pretty damn accurately. How can a painter look at any scene recreate (to scale) what they see on canvas? I'd say are brains are pretty damn good at calculating with very high accuracy.
Just because I can't say the hoop is exactly 32.74578453 feet from me doesn't mean I don't know how far it is away. If I can throw the ball into the hoop then I have accurately calculated/predicted the distance.
Look at how sometimes people are mid-conversation, talking about something they know in depth and suddenly they forget what they were going to say- this is because processing in the brain has gone completely off track.
I'm having a hard time coming up with a good analogy but I suspect these situations are similar to interrupts in computers. Something more important requires the brains resources at that time. It's not like the information is forgotten it's simply not accessible at that movement in time. The information is never "lost" it's just unavailable for a time. If it was lost you wouldn't have the "oh yeah" moments when you remember it or look it up again. You recognize it because you already knew it.
While I agree the brain isn't as effective at large scale number crunching I do believe it's something the brain can be trained to do. There are plenty of people out there who can do insanely complex arithmetic in their heads. I suspect the reason we all don't have such skills is because we don't need them.
but there's a lot that current computers can do that the brain can't- serious large scale number crunching for example
There is no real reason in the survival of the fittest terms for us to be able to accomplish such tasks. So those resources in the brain were put to use on other tasks like accurately processing visual and audio data. I can hear or spot a predator very quickly and accurately in all types of environments and lighting conditions. If we use a computer to perform these tasks we realize just how much computation is required. There is no reason these resources couldn't be allocated to general number crunching. It's just evolution says they are better used for other tasks.
Slashdot Mirror
I've recently met Vilmos Zsigmond ( http://www.imdb.com/name/nm0005936/ ) recently when he came to town for a screening of Close Encounters and did a lighting demo with Q&A. He went on at length about film vs digital. He essentially said with digital his job really don't change. It's as simple as working with another film stock. However he does prefer film because he felt digital was actually TOO sharp.
There is no film stock that lets you shoot both 24fps and then shoot thousands of FPS under the same lighting conditions. I suspect there aren't any film cameras that do 1000+ fps AND 24 with the same motor. You have to have the right tool for the Job. So you switch to a Phantom Flex type camera for your real slowmo stuff.
I've been told the current method for archival is actually just that. They use 3 B&W stocks to separate the color because B&W is far more stable and last longer under ideal conditions.
I'm a student film maker in an experimental (focused more on art house than typical Hollywood) film program. There are tons of students here madly in love with film for various reasons. Most of them are purely sentimental and no doubt unable to tell the difference if you put them side by side. I'm not one of them. I prefer the advantages of digital.
However, there are some legitimate reasons to prefer film. Digital workflow has it's downsides. It can be more complicated to get started editing with. Film you can just cut and tape together. You can hold the medium in your hand and see how it all works together. Some people prefer something they can physically touch. It can be a more enjoyable process for them to work with and problem solve with.
Digital can require significantly more complicated just to get your footage to play nice with your NLE software. Also, being able to see your image instantly can give students the impression they can cut corners in planning stages. When you can't see your final image until days/weeks later after processing it really forces you to make sure you plan everything out in more detail. You just have understand everything going on to avoid mistakes that will cost you both in time and money.
You don't need xyz information. The objects data is in a tree structure. The root has some XYZ position and everything else is relative to it. It's organized in a fashion where each level of the tree contains more fine detail. So you can calculate when the detail is smaller than a pixel and stop rendering. Chance are you never need to store the entire object in memory unless you have a very high resolution display.
This engine supposedly doesn't use voxels but I suspect it's something very similar. Check out these videos for a better explanation
http://www.youtube.com/watch?v=gNZtx3ijjpo
It would be a huge time saver. Pulling focus gets complicated quickly. Not having to worry about it is one less thing to have go wrong. You can do less prep work, less takes, and require less staff. Save $$$ and now complicated things become more approachable for a lower budget film.
Also, having that control after the fact in post production would be great. Got a poor performance from an extra in the background? Well, now there's a very simple way to make it less noticeable. A shot is too confusing and the audience gets confused? Well, now you can guide them where to look.
I guess I was trying to say you don't necessarily need grammar for a language used by computers. Grammar for them is just a hack, or addon to allow a language to communicate more than the originally was intended.
Humans don't want to reinvent the wheel every time we need to expand our language and thus grammar works well for this. Computers don't have that issue and so grammar (at least as we know it) isn't important.
I was just trying to point out that having a grammar isn't required for a language.
Well, depending on the number of communications they need to make to each other it's very possible 4 level words could map out every possible communication they could have with it each other. Think of it like Chinese symbols.. They aren't just one word but complex ideas. Grammar exists because we are unable to store such large amounts of data. We can't have 1 word/symbol map to a unique complex concept. A computer might not have such limitations. Especially if their entire universe of ideas/concepts can be enumerated easily.
Same with Aliens from outer space/other dimensions!
We submitted our (The Amateur Monster Movie http://www.imdb.com/title/tt1717690/ ) zero budget film to be added IMDB a few times in the past while still doing principal photography. They denied it until we released our trailer and started getting more press coverage. We don't have any distribution deal (yet) or submitted it to any festivals.
I think if they just released a trailer and got more press (which Slashdot should fix) they will be added. I think it's a matter of them simply trying to avoid adding films that most likely will never be seen by anybody but the people involved in creating them.
Here is a link to our trailer in case your curious.
http://www.youtube.com/watch?v=3aZquMQhAmo
My understanding is that the sharpest part of the lens is in the center. So the crop factor actually reduces the vignetting effect of a poor lens.So a crappy lens might become usable when combined with a smaller sensor.. It's all a trade off anyway. Smaller sensor = lower cost to manufacture but lens can potentially be more expensive. You just need to find the sweet spot.
Right now full frame seems like the way to go because it's easy to find cheap but good new and used lens. However, not too many cameras (in video) offer such a solution.
Adapters from C/CS to any other mount are cheap. Going the reverse is expensive.
I've been watching the Apertus/Elphel project for some time now. I too wanted a larger sensor to allow a nice shallow DOF. However, the more research I do the more I realize it can still be achieved. The digital cameras that were used to shoot the Star Wars prequels were 2/3" and they achieved a very cinematic look. The C-mount means you can get lens that are capable of doing this at a fraction of the price.
Really, the only true advantage of a larger sensor is having potentially larger pixels. This allows each photosite to capture more light and thus work better in lower light conditions. Think of a bucket capturing water. The larger the bucket the more water you can get.
Now, right now nobody seems to be manufacturing a full frame sensor that does what you need and is cheap enough. Once they exist I'm sure the Elphel/Apertus team will start using them.
Here is another interesting camera project that seems to debunk many of the perceived size sensor limitations. It' called "Drama" and is designed to be a camera capable of uncompressed video. http://dvxuser.com/V6/showthread.php?t=219424
Finally!!! A news article I can slip in a promo for the zero budget, comedy/horror film I've been working on over the last year. It is called "The Amateur Monster Movie" and was shot in and around Milwaukee Wisconsin. You can view our trailer on Youtube via the link below. Let me know what you guys think.
http://www.youtube.com/watch?v=3aZquMQhAmo
If Pi is infinite and has no discernible patterns then wouldn't that basically fall into the infinite monkeys and typewriter category? If you look long enough you will find any encoded message you want because it contained every possible coded message?
http://www.lumiera.org/
Does linux support HFS+ w/ journaling now?
In a brain like system we provide the stimuli which effectively act as seeds, but depending on other, non-deterministic events we can't guarantee that it will be the same.
I guess I don't understand how the non-deterministic nature really has any affect on it's equivalency to a computer. Like how a DFA and NFA are shown to be equivalent.
I've just recently taken my first AI course but unfortunately it just turned out to be a very boring algorithms course. It wasn't nearly as comprehensive as the required (for CS degree) algorithms course either. I am very interested in the subject and know just the very basics of neural networks but do stumble on a few articles about them from time to time
Can you recommend any books on both neural networks and the brain itself?
I agree it's pattern matching but I also consider that to be a mathematical process.
A machine/computer can only be as accurate as it is because we isolate it. We limit the number of things that can go wrong. A brain doesn't have that luxury. Practice seems to be learning to isolate a task from unknown and unrelated external forces/input.
I see no reason why a brain can't be trained to perform the exact same task as a computer/machine being as accurate and consistent. It's just we don't do it. We find short cuts. We make tools. If we had to do it that way our brains would eventually adapt to do just that.
I think there is numerical analysis involved it's just we don't directly have access to it. Our conscience mind doesn't have root access and thus is limited to some sort of API and there is private data structures I simply can not access via my conscience mind.
True, in the case of the basketball we only limited feedback from our throw function. It returns make/miss overthrow/underthrow type results. However, even with our initial throw we have some basis for how far we can throw something. We might not think of it directly in terms of feet but there is some built in relation to some unit of measurement in our heads. This basic unit is what guides all internal calculations/conversions.
This is how when somebody says toss the ball twenty feet I can get it close to twenty feet. I have some idea of what twenty feet is as a unit of measurement. I have some idea of how far I can throw is ball. I have some idea of how much force I need to exert to throw it twenty feet. Sure it's not going to be dead on accurate/consistant without extensive practice.
You might say that you could build a robot to throw the ball exactly twenty feet every time. Sure, you can factor in things like wind, elevation, etc etc. However there is always going to be something else that could cause your robot to fail. It could just fail mechanically over time. I could unplug it or just push it over. Or some other influence simply not accounted for.
The reason our brains can't do these things accurately is because we don't need them to. The time and effort to train yourself to do such a thing is insanely large compared to building (or in modern times) buying a tool to achieve the desired results. If a tool didn't exist then our brains would develop the ability to do just that over time. It might not happen in a single generation but it would happen.
it's quite easy in computers, because we have explicit memory addresses
Yes, it's true you can find something quickly if you have the memory address. However, you still need to know the address of what you want. A dictionary is a great way to verify the spelling of a word but you still need to have a good idea of how the word is spelled to find it.
A computer could simply check every address via brute force to find what it needs but that isn't generally what happens. I assume our brain works somewhat similar to a database. You make a query and get results. Those results can be used to make another query and so on and so forth until you may eventually find your desired result.
Where are my keys? I dunno... Is it where i normally put them? No. Let's retrace my steps.. Found them. I don't have a specific "address" to find my keys but I can usually locate them via a similar method. However, if I always put my keys in the same place then I have the location already. Like a computer I have the place to find them and it's only a one step process. However, as things get more complicated you simply can't juggle that many direct addresses to information and nor does a computer.
True, we don't say oh I left my keys at 0xffa12b012 but that is because our conscience mind doesn't have root access. It has to access everything some sorta limited access API. I suspect we will continue to hack our brains until we gain such access.
Being able to throw a ball right some, or even most of the time may be fine for a game of basketball, but is this ability good enough to calculate the values for some complex and precise engineering application? Absolutely not.
I would argue that are brains can and will. It's just a matter of time before they naturally evolve to do such things. I believe autism/savant types are simply evolution trying to give us those abilities. However, I wouldn't downplay the precise engineering that goes into things like sports or arts. It's just our brains were able to attempt to these easier as they relate much closer to things we already can do well. It will be.. We just haven't gotten there yet. Our brains will evolve to do such tasks it's just we haven't needed them to do it.
Again, absolutely I agree, but the issue is consistently, you may be able to train it but when it's so vulnerable to minor changes in the way it works, can it do it consistently?
Well, computers seem consistent but they do fail. It could be a software bug, it would be a hardware bug. Given enough time bits will randomly flip and may cause inconsistency in calculation. A computer flips a sign bit and suddenly 1 * 9 = -9. Sure some (or even many) of these can be avoided by adding error detection/correction but it will never catch them all. Perhaps a human brain is doing significantly more calculations than modern computers. I'm guessing the probability of random bit error in the brain/computer are close and it's simply a matter of the brain doing significantly more calculations per second that it happens frequently.
What's really going to blow your mind is that analog doesn't really exist ;)
I think I've decided to no longer consider their brains severely damaged. Their brains seem to be wired in a manor that puts communication (and other things) at a lower priority. The link below is a video demonstrating how a person originally classified as severly mentally disabled was simply unable to communicate in the manor we are accustom to.
http://www.wimp.com/autisticgirl/
I think it's simply evolution. While these people who never last in the hunter/gather world in modern times we care for them and in some cases they are able to lead "normal" lives and reproduce. These traits get past on and eventually perfected to the point where it isn't such a resource drain on their brain. Now, they can pick up those other skills like being able to communicate and function "normally" but with a new specialized part of the brain.
but hopeless for calculating with a reasonable degree of accuracy the actual distance to that object- the margin of error for most people is on average going to be quite large.
I disagree. How can we learn to throw a basketball into a tiny hoop from far away without having very accurate estimates? Think of any sport and just how many good estimates are done VERY quickly and pretty damn accurately. How can a painter look at any scene recreate (to scale) what they see on canvas? I'd say are brains are pretty damn good at calculating with very high accuracy.
Just because I can't say the hoop is exactly 32.74578453 feet from me doesn't mean I don't know how far it is away. If I can throw the ball into the hoop then I have accurately calculated/predicted the distance.
Look at how sometimes people are mid-conversation, talking about something they know in depth and suddenly they forget what they were going to say- this is because processing in the brain has gone completely off track.
I'm having a hard time coming up with a good analogy but I suspect these situations are similar to interrupts in computers. Something more important requires the brains resources at that time. It's not like the information is forgotten it's simply not accessible at that movement in time. The information is never "lost" it's just unavailable for a time. If it was lost you wouldn't have the "oh yeah" moments when you remember it or look it up again. You recognize it because you already knew it.
While I agree the brain isn't as effective at large scale number crunching I do believe it's something the brain can be trained to do. There are plenty of people out there who can do insanely complex arithmetic in their heads. I suspect the reason we all don't have such skills is because we don't need them.
but there's a lot that current computers can do that the brain can't- serious large scale number crunching for example
There is no real reason in the survival of the fittest terms for us to be able to accomplish such tasks. So those resources in the brain were put to use on other tasks like accurately processing visual and audio data. I can hear or spot a predator very quickly and accurately in all types of environments and lighting conditions. If we use a computer to perform these tasks we realize just how much computation is required. There is no reason these resources couldn't be allocated to general number crunching. It's just evolution says they are better used for other tasks.
I stumbled upon this one recently. http://img.oddstuffmagazine.com/2010/05/scientific-tattoo-ideas-03-580x488.jpg