Oh, and I just thought of another way the animation algorithm could get smarter: recognizing cyclic changes. In a lot of old games, there were cyclic changes in sprites. Every 1/2 to 1 second, the sprite cycles between one of two frames to represent shimmering water, people walking, fire flickering, etc. That could be easily detected and be used to set the animation speed.
Well, if you really want to get technical, there are several other things you need to factor again, both of which work against your ability to resolve:
Things aren't either "resolved" or "unresolved"; there's a loss of contrast until the brain can no longer fiddle out what the data is supposed to be. This is modeled by a modulation transfer function. The limit for a person to process between two stark black and white lines is when it gets down to the ~10% of original (varies significantly from person to person). The full human response is presented by the csf equation:
CSF( f ) = 2.6 * (0.0192 + 0.114 * f ) * exp(-0.114 * f )^1.1
f is the frequency in cycles per degree (line pairs), and CSF is a 0-1 contrast measurement. For real-world images which don't have such stark contrast lines, the limiting contrast is notably higher, and thus, the practical resolving resolution is higher. Also, the lower the mtf contrast, the longer it takes the brain to figure out what it's seeing; hence, motion dramatically increases the practical resolving resolution. On top of that, human mtf response yields lower contrast in dimmer light conditions; given two alternating colors of brightness A and A*100, and B and B*100, where B is 10x A, you may get double the apparent contrast with the second pair.
You also have to factor in depth-of-focus limits. When you're talking about having to bring screens inches from your eyes for the pixel density to be relevant, you're actually making the problem worse because the closer an object gets to the eye, the more trouble the eye has focusing on it. The resolution numbers above are only approached as an object's distance increases toward infinity (although are closely approximated at reasonable TV viewing distances).
All of this means that while yes, you're correct concerning the pixels / line pairs mixup, which means a higher resolution is practical, there's also a lot of factors making a *lower* resolution the practical limit.
1080 is just fine vertical resolution unless you're wanting your device to extend well into the peripheral.
The animation would start instantly; it would only take the, say, 1/4 second (which is not very long at all -- about as fast your average person can sweep their fingers up and down) to complete. And if that time is too long for you, well, that's what the word "customizeable" was there for -- aka, the user can change the default. Lastly, if the "smart" algorithm is implemented, this problem, if it ever existed, would generally go away. Creature goes into its "I'm about to attack you" mode, then quickly goes into "attacking you" mode. The algorithm would learn that "I'm about to attack you" needs a fast animation.
Said dead pixel lowers the average lightness of all pixels around it. And anyway, it's not *that* much more than what the average person can perceive. 1080 is actually a very good standard for practical usage.
Looking at the images, I don't think it'd be that hard to detect the level of antialiasing in the image. In fact, they're already kind of doing it when they decide whether to blend areas or to leave them distinct. To me, that leaves the obvious solution: rather than a binary between "sharp" and "blurred" being determined by the level of color difference, you use the level of color difference (with a particular focus on matching hue) to decide the maximum transition distance. So, for example, a stark color change will still essentially be a clean line, and a gradual color change will still be smooth, but a moderate color change will have a small blending distance.
The maximum theoretical human eye resolution (limited by aperture and focal length) is about 0.35mm per line pair at 1m. For most people with "normal" eyesight, the limit is closer to 0.7 mm at 1m. So for a TV that's 6 feet away and 2 feet tall, there's no point to a vertical resolution higher than ~1800 pixels, and 1080 is more than what most people can perceive. 4320 pixels at 6 feet makes no sense unless your TV is a minimum of 5 feet tall, and for most people, 15 feet;) So yes, it really does take wall-sized TVs for that sort of resolution to even be useful. A 7" panel (~4" tall) with 1600 pixels vertical resolution? That's 0.06mm per pixel. The maximum theoretical distance in which that could be useful is under 7 inches, and more practically, closer to 3 inches away from the viewer. If you want that kind of resolution, you might as well make it into a planetarium;)
I can easily picture taking this one step further. With a couple additional steps, your sprite drawer should be able to recognize matching parts in other frames of the same sprite. You could then allow a customizeable time period -- say, 1/4 to 1/2 second -- where there's a linear interpolation between the two SVG graphics. Aka, when Mario jumps, his legs don't just instantly appear apart; they move there. A smart drawing algorithm could even intelligently determine the appropriate time period to have for the animation. Aka, if it oft tries to switch frames again before the animation is completed, the animation for the transition between those particular frames gets sped up.
Rental prices vary hugely. We got our car reserved from CheapJeep for 9,000 ISK/day (under $50/d) midseason -- by Icelandic standards, that's a steal. I don't know about accomodation fees, as we're only camping and couchsurfing. I could quote you prices for remote trail cabins, though, lol;)
Well, either way, I'm sure you'll have a great time:)
Lol, I have to get used to the concept of staying in a single location on a trip;) When we travel, we always move every night. At least on this trip we'll have a car to act as a mobile home base. What a luxury, not having to have all of our gear on our backs at once!;) Esp. food.
Come back in August? So are you going before then? Because we'll be there in mid-late July.:) If that's the case, lol, we could meet up for lunch or something.
* The pictures I've seen of the beaches near Vík are stunning.:)
* If you're going to rent a 4x4, you can drive through the Landmannalaugar area.
* Skógafoss is amazing, and you can camp right by it. Of course, you said you're already going up the Skóga....
* Svartifoss is a bit more out of the way and smaller, but beautifully framed by columnar basalt cliffs.
* I've heard the Morsá valley recommended, and it definitely looks like an easy hike, although the pictures I've seen don't impress me as much as elsewhere.
* Jökulsárlón is an obvious candidate, and kind of hard to avoid anyway;)
* I assume you're not planning to get on top of Vatnajökull;)
But you probably already know that stuff;) Ooh, actually, here's something you may or may not know: in August, thishappens at Jökulsárlón.:) Wish I could be there for it, but I had to choose between that and Bræðslan.
Takk!:) Hehe, ég get ruglað Íslensku mjög vel stundum;) Ég verð að aefa mig. Fyrsti ferð mín til Íslands er í Júlí. Ég ekki að allir tala ensku, en mér finnst gaman að læra.:)
Old Norse (or more accurately, Proto-Norse) is the origin point of the North Germanic languages. Hence, it branched off from the West Germanic languages when each went their separate ways (aka, were no longer mutually intelligible)
English has a number of Norse-origin loanwords (several hundred), but has more in common through the original Germanic connection.
I don't see how Iceland has anything to do with your involvement in the Cold War. Or was your goal to change from a discussion of a dispute between Iceland and the UK into a dispute between the UK and the US in order to avoid discussing your long history of (over) exploiting the resources of a poorer country which couldn't defend itself?
Oooh, someone who knows their Icelandic geography and actually types their accents, umlauts, and eths;) I'm actually hoping to hike Fimmvörðuháls in July. It should be pretty neat, seeing the lava that móði and magni laid down fresh over the trail.
Random side note: her name is Jóhanna Sigurðardóttir. Do you know the proper way to address her?
"Jóhanna".
You similarly address your teachers, elders, etc by first name. Even phone books in Iceland are sorted by first name. The "last name" is a patronymic -- "Sigurðardóttir" literally means that her father's name was Sigurðar.
Slashdot Mirror
Oh, and I just thought of another way the animation algorithm could get smarter: recognizing cyclic changes. In a lot of old games, there were cyclic changes in sprites. Every 1/2 to 1 second, the sprite cycles between one of two frames to represent shimmering water, people walking, fire flickering, etc. That could be easily detected and be used to set the animation speed.
Well, if you really want to get technical, there are several other things you need to factor again, both of which work against your ability to resolve:
Things aren't either "resolved" or "unresolved"; there's a loss of contrast until the brain can no longer fiddle out what the data is supposed to be. This is modeled by a modulation transfer function. The limit for a person to process between two stark black and white lines is when it gets down to the ~10% of original (varies significantly from person to person). The full human response is presented by the csf equation:
CSF( f ) = 2.6 * (0.0192 + 0.114 * f ) * exp(-0.114 * f )^1.1
f is the frequency in cycles per degree (line pairs), and CSF is a 0-1 contrast measurement. For real-world images which don't have such stark contrast lines, the limiting contrast is notably higher, and thus, the practical resolving resolution is higher. Also, the lower the mtf contrast, the longer it takes the brain to figure out what it's seeing; hence, motion dramatically increases the practical resolving resolution. On top of that, human mtf response yields lower contrast in dimmer light conditions; given two alternating colors of brightness A and A*100, and B and B*100, where B is 10x A, you may get double the apparent contrast with the second pair.
You also have to factor in depth-of-focus limits. When you're talking about having to bring screens inches from your eyes for the pixel density to be relevant, you're actually making the problem worse because the closer an object gets to the eye, the more trouble the eye has focusing on it. The resolution numbers above are only approached as an object's distance increases toward infinity (although are closely approximated at reasonable TV viewing distances).
All of this means that while yes, you're correct concerning the pixels / line pairs mixup, which means a higher resolution is practical, there's also a lot of factors making a *lower* resolution the practical limit.
1080 is just fine vertical resolution unless you're wanting your device to extend well into the peripheral.
Humans don't stay perfectly still in static tests of the ability to resolve lines, so that's clearly already factored in.
The animation would start instantly; it would only take the, say, 1/4 second (which is not very long at all -- about as fast your average person can sweep their fingers up and down) to complete. And if that time is too long for you, well, that's what the word "customizeable" was there for -- aka, the user can change the default. Lastly, if the "smart" algorithm is implemented, this problem, if it ever existed, would generally go away. Creature goes into its "I'm about to attack you" mode, then quickly goes into "attacking you" mode. The algorithm would learn that "I'm about to attack you" needs a fast animation.
Said dead pixel lowers the average lightness of all pixels around it. And anyway, it's not *that* much more than what the average person can perceive. 1080 is actually a very good standard for practical usage.
Looking at the images, I don't think it'd be that hard to detect the level of antialiasing in the image. In fact, they're already kind of doing it when they decide whether to blend areas or to leave them distinct. To me, that leaves the obvious solution: rather than a binary between "sharp" and "blurred" being determined by the level of color difference, you use the level of color difference (with a particular focus on matching hue) to decide the maximum transition distance. So, for example, a stark color change will still essentially be a clean line, and a gradual color change will still be smooth, but a moderate color change will have a small blending distance.
The maximum theoretical human eye resolution (limited by aperture and focal length) is about 0.35mm per line pair at 1m. For most people with "normal" eyesight, the limit is closer to 0.7 mm at 1m. So for a TV that's 6 feet away and 2 feet tall, there's no point to a vertical resolution higher than ~1800 pixels, and 1080 is more than what most people can perceive. 4320 pixels at 6 feet makes no sense unless your TV is a minimum of 5 feet tall, and for most people, 15 feet ;) So yes, it really does take wall-sized TVs for that sort of resolution to even be useful. A 7" panel (~4" tall) with 1600 pixels vertical resolution? That's 0.06mm per pixel. The maximum theoretical distance in which that could be useful is under 7 inches, and more practically, closer to 3 inches away from the viewer. If you want that kind of resolution, you might as well make it into a planetarium ;)
I can easily picture taking this one step further. With a couple additional steps, your sprite drawer should be able to recognize matching parts in other frames of the same sprite. You could then allow a customizeable time period -- say, 1/4 to 1/2 second -- where there's a linear interpolation between the two SVG graphics. Aka, when Mario jumps, his legs don't just instantly appear apart; they move there. A smart drawing algorithm could even intelligently determine the appropriate time period to have for the animation. Aka, if it oft tries to switch frames again before the animation is completed, the animation for the transition between those particular frames gets sped up.
Rental prices vary hugely. We got our car reserved from CheapJeep for 9,000 ISK/day (under $50/d) midseason -- by Icelandic standards, that's a steal. I don't know about accomodation fees, as we're only camping and couchsurfing. I could quote you prices for remote trail cabins, though, lol ;)
Well, either way, I'm sure you'll have a great time :)
Lol, I have to get used to the concept of staying in a single location on a trip ;) When we travel, we always move every night. At least on this trip we'll have a car to act as a mobile home base. What a luxury, not having to have all of our gear on our backs at once! ;) Esp. food.
Come back in August? So are you going before then? Because we'll be there in mid-late July. :) If that's the case, lol, we could meet up for lunch or something.
South and east, eh? And minimal hiking?
* The pictures I've seen of the beaches near Vík are stunning. :) ;) ;)
* If you're going to rent a 4x4, you can drive through the Landmannalaugar area.
* Skógafoss is amazing, and you can camp right by it. Of course, you said you're already going up the Skóga....
* Svartifoss is a bit more out of the way and smaller, but beautifully framed by columnar basalt cliffs.
* I've heard the Morsá valley recommended, and it definitely looks like an easy hike, although the pictures I've seen don't impress me as much as elsewhere.
* Jökulsárlón is an obvious candidate, and kind of hard to avoid anyway
* I assume you're not planning to get on top of Vatnajökull
But you probably already know that stuff ;) Ooh, actually, here's something you may or may not know: in August, this happens at Jökulsárlón. :) Wish I could be there for it, but I had to choose between that and Bræðslan.
*Campers* are *juicy*.
No heavy industry or smelters? Their main export is aluminum. What do you think the Kárahnjukar hydro plant was built for?
Nah... Björk is part húldufólk. ;)
Takk! :) Hehe, ég get ruglað Íslensku mjög vel stundum ;) Ég verð að aefa mig. Fyrsti ferð mín til Íslands er í Júlí. Ég ekki að allir tala ensku, en mér finnst gaman að læra. :)
(hmmm... Slashdot borðar "thorns". "Ég 'th'ekki...", etc)
Most people in Iceland find it disgusting, and it's not eaten often. Pizza, on the other hand...
Old Norse (or more accurately, Proto-Norse) is the origin point of the North Germanic languages. Hence, it branched off from the West Germanic languages when each went their separate ways (aka, were no longer mutually intelligible)
English has a number of Norse-origin loanwords (several hundred), but has more in common through the original Germanic connection.
I assume you mean tallest, not longest ;) And is Vinnufossen a waterfall or did someone leave a tap dripping? ;)
In terms of sheer power, Iceland's easily got you beat. You have no Dettifoss, for example.
Unless you're one of the Sámi people, you Norwegians are just a bunch of Germans who moved north while the Pharaohs were ruling Egypt.
See how this silly game goes?
I don't see how Iceland has anything to do with your involvement in the Cold War. Or was your goal to change from a discussion of a dispute between Iceland and the UK into a dispute between the UK and the US in order to avoid discussing your long history of (over) exploiting the resources of a poorer country which couldn't defend itself?
Actually, I probably should get myself out of the habit of translating buinn & similar as "finished" and simply use the past tense....
Should a straight person go by "breederminister"? I mean, it's such a stupid joke.
Oooh, someone who knows their Icelandic geography and actually types their accents, umlauts, and eths ;) I'm actually hoping to hike Fimmvörðuháls in July. It should be pretty neat, seeing the lava that móði and magni laid down fresh over the trail.
Random side note: her name is Jóhanna Sigurðardóttir. Do you know the proper way to address her?
"Jóhanna".
You similarly address your teachers, elders, etc by first name. Even phone books in Iceland are sorted by first name. The "last name" is a patronymic -- "Sigurðardóttir" literally means that her father's name was Sigurðar.
I'm sure you're mispronouncing it. It's basically "GREEMS-vih-dn", with the ö pronounced as the eux in the French "deux" and the 'r' rolled.
And anyway, Grímsvötn actually refers to the subglacial lakes in the area (vötn = lakes, plural of vatn)