Jump to content
  • Advertisement
Sign in to follow this  
Ashkan

Depth of Field

This topic is 3809 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Considering the current state of technology, I've been always wondering how depth of field could be of any meaningful use in games. It sure results in impressive screen shots, but the problem is that depth of field is trying to manipulate view parameters in a direction that players have no control over, so developers usually end up with fixed focal points, beyond which everything is blurred. Our eyes can dynamically adjust the focal point so the resulting effect is more dramatic. That's really the kind of effect that I expect when I hear the term "Depth of Field", but with such limited use as I've seen in games (I haven't actually played any of the newer games due to limited system resources - I'm judging based on screen shots here so I might be missing something here) I can't help myself but to wonder... what's the fuss? Do we have algorithms / input devices that allow such dynamic manipulation of the focal point? Any insights are welcome.

Share this post


Link to post
Share on other sites
Advertisement
I don't really see the big deal either, but like many things in graphics, physical accuracy isn't as important as the illusion of accuracy. The brain sees something that slightly resembles depth-of-field and it's fooled into thinking it's there in true form. The same thing happens all the time in various algorithms for lighting and shadowing, where you only need to "nudge" the brain in the right direction and it fills in the rest. SSAO is a excellent, fairly recent example of a pseudo-shadowing algorithm that aims to fool the brain into perceiving a properly shadowed scene.

Share this post


Link to post
Share on other sites
Quote:
Original post by Ashkan
Considering the current state of technology, I've been always wondering how depth of field could be of any meaningful use in games. It sure results in impressive screen shots, but the problem is that depth of field is trying to manipulate view parameters in a direction that players have no control over, so developers usually end up with fixed focal points, beyond which everything is blurred. Our eyes can dynamically adjust the focal point so the resulting effect is more dramatic. That's really the kind of effect that I expect when I hear the term "Depth of Field", but with such limited use as I've seen in games (I haven't actually played any of the newer games due to limited system resources - I'm judging based on screen shots here so I might be missing something here) I can't help myself but to wonder... what's the fuss? Do we have algorithms / input devices that allow such dynamic manipulation of the focal point?

Any insights are welcome.

Personally, I agree. Why "simulate" DOF in-game when the human eye automatically focuses on certain objects? I hate it when a game uses DOF and "forces" me to look at, say, my weapon. What, I can't look at the few meters ahead of me while recharging my gun, or something [grin]?

I think it's the same issue as (over-done) HDR was(is). I'd rather not look at a completely white, glowing wall just because there's some sunlight hitting it. Nor do I want Alex's face to glow in the darkest of places...

@ Zipster: It's not so much the illusion anymore, it's like ... overdoing everything [wink].

Share this post


Link to post
Share on other sites
I agree that DOF is mostly nonsense in games nowadays..It's best use is for cutscenes that emulate camera work.

I think using it to blur far away stuff as Gothic 3 is sort of a hack.

Share this post


Link to post
Share on other sites
You can dynamically adjust the focal point of DOF based on what the Camera is looking at...Crysis does this.

Share this post


Link to post
Share on other sites
DOF is there to hide the low-detail level geometry in the distance. Nothing else :-) ... so it is a technique that helps to render large-scale worlds with a very aggressive LOD system.
The only way to make it work in a senseful way other than this is using eye / head tracking software ... but then you will have problems with the low-lod geometry :-)

Share this post


Link to post
Share on other sites
Quote:
Original post by agi_shi
Quote:
Original post by Ashkan
Considering the current state of technology, I've been always wondering how depth of field could be of any meaningful use in games. It sure results in impressive screen shots, but the problem is that depth of field is trying to manipulate view parameters in a direction that players have no control over, so developers usually end up with fixed focal points, beyond which everything is blurred. Our eyes can dynamically adjust the focal point so the resulting effect is more dramatic. That's really the kind of effect that I expect when I hear the term "Depth of Field", but with such limited use as I've seen in games (I haven't actually played any of the newer games due to limited system resources - I'm judging based on screen shots here so I might be missing something here) I can't help myself but to wonder... what's the fuss? Do we have algorithms / input devices that allow such dynamic manipulation of the focal point?

Any insights are welcome.

Personally, I agree. Why "simulate" DOF in-game when the human eye automatically focuses on certain objects? I hate it when a game uses DOF and "forces" me to look at, say, my weapon. What, I can't look at the few meters ahead of me while recharging my gun, or something [grin]?

I think it's the same issue as (over-done) HDR was(is). I'd rather not look at a completely white, glowing wall just because there's some sunlight hitting it. Nor do I want Alex's face to glow in the darkest of places...

@ Zipster: It's not so much the illusion anymore, it's like ... overdoing everything [wink].
Maybe I don't understand, but the computer screen is a flat surface, so the human eye looking at a rendered image can only focus at one depth - the distance to the screen. If the game didn't provide any focal length variation (ala Crysis) then there wouldn't be any blurring at all...

Share this post


Link to post
Share on other sites
Quote:
Original post by Jason Z
Quote:
Original post by agi_shi
Quote:
Original post by Ashkan
Considering the current state of technology, I've been always wondering how depth of field could be of any meaningful use in games. It sure results in impressive screen shots, but the problem is that depth of field is trying to manipulate view parameters in a direction that players have no control over, so developers usually end up with fixed focal points, beyond which everything is blurred. Our eyes can dynamically adjust the focal point so the resulting effect is more dramatic. That's really the kind of effect that I expect when I hear the term "Depth of Field", but with such limited use as I've seen in games (I haven't actually played any of the newer games due to limited system resources - I'm judging based on screen shots here so I might be missing something here) I can't help myself but to wonder... what's the fuss? Do we have algorithms / input devices that allow such dynamic manipulation of the focal point?

Any insights are welcome.

Personally, I agree. Why "simulate" DOF in-game when the human eye automatically focuses on certain objects? I hate it when a game uses DOF and "forces" me to look at, say, my weapon. What, I can't look at the few meters ahead of me while recharging my gun, or something [grin]?

I think it's the same issue as (over-done) HDR was(is). I'd rather not look at a completely white, glowing wall just because there's some sunlight hitting it. Nor do I want Alex's face to glow in the darkest of places...

@ Zipster: It's not so much the illusion anymore, it's like ... overdoing everything [wink].
Maybe I don't understand, but the computer screen is a flat surface, so the human eye looking at a rendered image can only focus at one depth - the distance to the screen. If the game didn't provide any focal length variation (ala Crysis) then there wouldn't be any blurring at all...


That's true. But the game doesn't know what I'm trying to look at! Consider one of the earlier games that really demonstrated this problem: Zelda Wind Waker. I'm not even in 1st person perspective, I'm just a disembodied camera following Link around the world. I notice something that looks interesting off in the distance. I am absolutely powerless to look at it, because the focal length of the camera is hard-coded to some arbitrary (short) distance. If the game could read my mind, then I might not care so much. But until that point, I don't really want it deciding where my eyes are allowed to focus.

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!