I hope that is sarcasm there because if you think that is a computer anyone would do anything on especially minecraft you are more than just delusional. He wasn't even doing anything remotely evil unless you consider honest criticism some form of satanic power. Now if he would have said your screenshot sucks because you hate cute puppies then it's just mindless hate but what you did right now what completely unnecessary. I suggest you step away from your computer and wait four or five years to grow up then come back here and make a proper response but hey lets all be butthurt cause you know why not!
It is sarcasm, but yeah, I gone a bit too far, I tend to do that.
But tell me honestly, was it a valid critique? How else would you showcase illumination if not in a dark place? Besides, you can clearly see everything here, maybe that's my fault that I tuned my monitor...
My monitor is five or six years old, it's not even a full hd as you can guess. But you know why I play on moody? Because I can see everything perfectly well!
If you are able to see on moody, then I would assert that your monitor is not properly calibrated.
How else would you showcase illumination if not in a dark place? Besides, you can clearly see everything here, maybe that's my fault that I tuned my monitor...
What do you mean, you tuned your monitor? What sort of calibration standard are you using? Did you actually tag your picture with a ICC profile?
Color calibration is NOT a trivial thing. If you just moved a slider to adjust "These bars look like this grey scale", then be aware that those tests can be seriously in error. If the bars are vertical -- which is true on the NVidia calibration tests I've seen -- then the test is a complete failure, and that's just the first potential problem. And if it was a dotted pattern to a solid, throw it out as worthless.
1. Did you test the black point? Is your black point calibrated to any sort of standard? Black point does not mean "is (0,0,0) black". It means "At what point can I visually distinguish black from non-black dark grey". By the nature of the test, it will vary based on whether you are in a bright room, or a dark room -- and this is part of what is specified in the ICC profiles that tag pictures.
2. Did you test the white point? Many computers have a very blue "white".
3. Did you actually test that the range is smooth? Sure, a new monitor might -- MIGHT have a smooth response from bright to dark, it isn't always going to be that way.
4. Are you CRT, LCD, LED, etc? Different monitor technologies have different base responses to voltage.
5. Does your system have anything to alter that base response? Even if your display hardware has a linear native response, you can still have an OS level, or driver level, or even hardware level lookup table to alter the voltage to look like something else.
The actually resultant behavior of the monitor is something else that goes into the ICC profiles.
6. Even if all of that is done, and everything is fine, I have two questions for you:
#1: If you have alternating horizontal bars of (dark point, white point), what is the value of white that you expect to match? Example: If you have bars of (0,0,0) alternating with bars of (255,255,255), what value do you consider to be the matching solid grey? And why would you even think of using (0,0,0) if the "minimum distinguishable from grey" is (15,15,15)?
#2: If you then take that middle value, and compare it to white and dark, what values do you get on the first edge?
If you even think that there is only one possible answer to those two questions, you don't even deserve a seat at the table to discuss this. And if you do understand that there is no single answer (hint: dot matrix printers are not the same as inkjets are not the same as lasers are not the same as magazines etc; that doesn't even take into account the many different monitor standards, etc.)
Do not assume "sRGB is the answer, everyone should use it". sRGB has horrible Gamut. "Cyan" does not exist in sRGB. "sRGB" was nothing more than an attempt to determine how typical consumer-grade CRT's worked, and the idea of "Lets all mimic a bad technology" is a disaster.
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?
First of all I'm sorry, I'm being insufferable sometimes.
I don't know names of the tests, I just ran through calibrating program that goes with monitor a few times until I got the best result, I don't know the science behind that, but judging by your description it contained most of them. And I can see on moody in pitch dark (hardly but still), so I'm happy with my settings, they provide a wider visible range, hard to complain really.
The first test on that system has bars of different colors, at different steps. 32 steps in all. Supposedly, you should see all of the bars, evenly spaced visually.
I took out digital color meter, and read the RGB values. They are roughly linear. So at every step, the RGB numbers sent to the display increase linearly.
The only way that this will give you a linear visual response, with all bars being visible, is if you have a gamma of 1.0 and a black point basically at/below (2,2,2).
(Now, in fairness: This is on a mac. I have the display set in display preferences to LCD Display, which should be the system unaltered setting. I have confirmed that if I change the value there, I get different numbers on the meter. Whether I'm looking in Firefox, or in Preview, things change -- if I set the monitor to 1.8, for example, the numbers go down. So there is some internal OS level color sync going on that cannot be disabled. But even still, it seems that this site is flawed.)
That's not any reasonable standard. So, no, I don't think that site is valid for testing, at least, not all of it.
... Yea, looking over the rest: a (1,1,1) image is different from a (2,2,2) and different from (0,0,0) (that's out of 256) -- not accurate.
...And I can see on moody in pitch dark (hardly but still), so I'm happy with my settings, they provide a wider visible range, hard to complain really.
And why would there ever be an option that can not be used? I'm afraid it's the other way around.
Why would there be an option that cannot be used? Seriously?
How about: "It looks good on my system, so that's all that matters". Much later: "Gee, my monitor is a native gamma 1.2 LCD, and my OS does no color calibration or correction, so it looks dark to other people. Well, lets add in a "brightness" slider, that should help".
Doesn't help that the game slider marked "brightness" actually is "gamma". It affects midtones more than darks.
And, I have actually checked the game's color numbers on a windows system that had a 16 bit display (sigh). Sure enough, it wants to use (0,0,0) and (8,8,8) to draw smooth stone (on moody) -- and expects you to be able to tell them apart.
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?
I don't know if you were playing at all, but there are several blocks in vanilla Minecraft that don't take colored light to well. Some just look weird, coloring moving pistons will actually crash the game.
As for the other bazillion mods you also installed, I have no idea. Please be aware... you are trying to run a BETA in full release with other mods... I can't guarantee that will work, especially if there are already bugs in the interactions with vanilla. I take the WIP tag seriously, and a BETA means that it's not fully released yet =)
Try removing a couple of mods you think might be the problem, try generating a superflat world, mess around, see what replicates the error, and what doesn't.
One of the many issues:
I made a mistake involving the call to "getMixedLight" It returns the skylight value + blocklight value + the color of the light... This is breaking a lot of things, and I'm going to remove the color in the next version. That will stop a lot of the terrible crashes. I'll make another helper method that actually gives me the data when I need it, but for now, you'll have to be content with the beta status I'm able to put out... It's a beta... nothing more.
I would like to make a suggestion of adding a config option or options to allow people to individually customize their lighting experience, in terms of how your lighting is calculated.
For example, I don't really like simply tinting a block with a particular color as in, with a red light, the blocks next to the red light are almost entirely pure red in tint (or maybe not almost, but completely).
I'd like to propose an alternative formula for calculating colors. Let's say you have a red light. RGB is 15,0,0 and L is 15.
As the light propagates, I would like to suggest that the initial color next to the light starts out bright white with a hint of red, and as you go further from the light, the light gets darker, but the color becomes more apparent.
L(x) = 65520 + i - ((~RGB << 4) / (1 + (r - x)), where x is distance in steps from light source, r is the range of the light, RGB is color, and i is light intensity.
Not sure how that formula would look exactly, but it kinda represents what I am describing.
But in other words, what I am trying to describe is that lights should look more like lights and less like color tinters. Then again, maybe this could simply be a matter of using the right RGB values in the lights. Like a red light shouldn't be 15,0,0 but maybe 10,0,0.
I would like to make a suggestion of adding a config option or options to allow people to individually customize their lighting experience, in terms of how your lighting is calculated.
For example, I don't really like simply tinting a block with a particular color as in, with a red light, the blocks next to the red light are almost entirely pure red in tint (or maybe not almost, but completely).
I'd like to propose an alternative formula for calculating colors. Let's say you have a red light. RGB is 15,0,0 and L is 15.
As the light propagates, I would like to suggest that the initial color next to the light starts out bright white with a hint of red, and as you go further from the light, the light gets darker, but the color becomes more apparent.
L(x) = 65520 + i - ((~RGB << 4) / (1 + (r - x)), where x is distance in steps from light source, r is the range of the light, RGB is color, and i is light intensity.
Not sure how that formula would look exactly, but it kinda represents what I am describing.
But in other words, what I am trying to describe is that lights should look more like lights and less like color tinters. Then again, maybe this could simply be a matter of using the right RGB values in the lights. Like a red light shouldn't be 15,0,0 but maybe 10,0,0.
Config - Possible
Config that alters lighting behavior - Possible
Lighting behavior that looks a little more washed out? - Possible
What you described - Not quite possible. Although I'm spending a lot of time with another individual overhauling the smooth lighting renderer. What you are describing sounds something akin to just flattening out the color curve... which will kinda work.
Sadley, the issue is that r,g,b and l, are all separate channels of light. There wasn't a really easy way for me to get anything to behave quite right, so I had to make this compromise... and force colors to behave and degrade from 0 to 15, resulting in the odd color saturation.
There are a couple of things in the pipeline that will kinda address this issue.
One: colored sunlight. At the moment, these lights look terrible in the sunlight, because they are able to override white light, and they make things look darker and more saturated.
If I make it so sunlight has a white color (as opposed to NO color) then the colored lights won't be able to propagate through and make daylight areas look as terrible...
I also would make sunlight on a per-dimension basis, so the end could be blue, the nether purple, and the over-world white...
Two: If I don't like where the block light ended up, I can always make all colors across the board appear less saturated.... leaving everything else in place. That's going to be for later though.
Yes they behave like "tinters" at the moment, which works in all cases, except sunlight... which I agree, needs work.
I also would make sunlight on a per-dimension basis, so the end could be blue, the nether purple, and the over-world white...
Why that exact colors if not a secret? Purple nether and blue end make a little sense to me, especially interacting with main blocks in that dimensions it would create some weird visuals. Surely nether is hell, so why would not it be red-ish?
Also, can you make skylight color dependent of the time of day? Blue at night, white/bright yellow at day and red at dawn/sunset?
I also would make sunlight on a per-dimension basis, so the end could be blue, the nether purple, and the over-world white...
Would you please talk to Xcw about Mystcraft support?
Short summary: Mystcraft lets you create dimensions with multiple suns; each sun can have a different color. Right now, only the "sunset" color has any effect, but if this would allow for actual different colored light from each sun ...
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?
It is sarcasm, but yeah, I gone a bit too far, I tend to do that.
But tell me honestly, was it a valid critique? How else would you showcase illumination if not in a dark place? Besides, you can clearly see everything here, maybe that's my fault that I tuned my monitor...
If you are able to see on moody, then I would assert that your monitor is not properly calibrated.
What do you mean, you tuned your monitor? What sort of calibration standard are you using? Did you actually tag your picture with a ICC profile?
Color calibration is NOT a trivial thing. If you just moved a slider to adjust "These bars look like this grey scale", then be aware that those tests can be seriously in error. If the bars are vertical -- which is true on the NVidia calibration tests I've seen -- then the test is a complete failure, and that's just the first potential problem. And if it was a dotted pattern to a solid, throw it out as worthless.
1. Did you test the black point? Is your black point calibrated to any sort of standard? Black point does not mean "is (0,0,0) black". It means "At what point can I visually distinguish black from non-black dark grey". By the nature of the test, it will vary based on whether you are in a bright room, or a dark room -- and this is part of what is specified in the ICC profiles that tag pictures.
2. Did you test the white point? Many computers have a very blue "white".
3. Did you actually test that the range is smooth? Sure, a new monitor might -- MIGHT have a smooth response from bright to dark, it isn't always going to be that way.
4. Are you CRT, LCD, LED, etc? Different monitor technologies have different base responses to voltage.
5. Does your system have anything to alter that base response? Even if your display hardware has a linear native response, you can still have an OS level, or driver level, or even hardware level lookup table to alter the voltage to look like something else.
The actually resultant behavior of the monitor is something else that goes into the ICC profiles.
6. Even if all of that is done, and everything is fine, I have two questions for you:
#1: If you have alternating horizontal bars of (dark point, white point), what is the value of white that you expect to match? Example: If you have bars of (0,0,0) alternating with bars of (255,255,255), what value do you consider to be the matching solid grey? And why would you even think of using (0,0,0) if the "minimum distinguishable from grey" is (15,15,15)?
#2: If you then take that middle value, and compare it to white and dark, what values do you get on the first edge?
If you even think that there is only one possible answer to those two questions, you don't even deserve a seat at the table to discuss this. And if you do understand that there is no single answer (hint: dot matrix printers are not the same as inkjets are not the same as lasers are not the same as magazines etc; that doesn't even take into account the many different monitor standards, etc.)
Do not assume "sRGB is the answer, everyone should use it". sRGB has horrible Gamut. "Cyan" does not exist in sRGB. "sRGB" was nothing more than an attempt to determine how typical consumer-grade CRT's worked, and the idea of "Lets all mimic a bad technology" is a disaster.
Sorry. I may have gotten carried away here.
* Promoting this week: Captive Minecraft 4, Winter Realm. Aka: Vertical Vanilla Viewing. Clicky!
* My channel with Mystcraft, and general Minecraft Let's Plays: http://www.youtube.com/user/Keybounce.
* See all my video series: http://www.minecraftforum.net/forums/minecraft-editions/minecraft-editions-show-your/2865421-keybounces-list-of-creation-threads
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?
*looks at thread title*
........
*looks at post*
........
*shrugs*
.........
http://www.lagom.nl/lcd-test/
Game designer, forum stalker.
First of all I'm sorry, I'm being insufferable sometimes.
I don't know names of the tests, I just ran through calibrating program that goes with monitor a few times until I got the best result, I don't know the science behind that, but judging by your description it contained most of them. And I can see on moody in pitch dark (hardly but still), so I'm happy with my settings, they provide a wider visible range, hard to complain really.
And why would there ever be an option that can not be used? I'm afraid it's the other way around.
Did you ever make it into a game?
Sorry, no.
The first test on that system has bars of different colors, at different steps. 32 steps in all. Supposedly, you should see all of the bars, evenly spaced visually.
I took out digital color meter, and read the RGB values. They are roughly linear. So at every step, the RGB numbers sent to the display increase linearly.
The only way that this will give you a linear visual response, with all bars being visible, is if you have a gamma of 1.0 and a black point basically at/below (2,2,2).
(Now, in fairness: This is on a mac. I have the display set in display preferences to LCD Display, which should be the system unaltered setting. I have confirmed that if I change the value there, I get different numbers on the meter. Whether I'm looking in Firefox, or in Preview, things change -- if I set the monitor to 1.8, for example, the numbers go down. So there is some internal OS level color sync going on that cannot be disabled. But even still, it seems that this site is flawed.)
That's not any reasonable standard. So, no, I don't think that site is valid for testing, at least, not all of it.
... Yea, looking over the rest: a (1,1,1) image is different from a (2,2,2) and different from (0,0,0) (that's out of 256) -- not accurate.
Why would there be an option that cannot be used? Seriously?
How about: "It looks good on my system, so that's all that matters". Much later: "Gee, my monitor is a native gamma 1.2 LCD, and my OS does no color calibration or correction, so it looks dark to other people. Well, lets add in a "brightness" slider, that should help".
Doesn't help that the game slider marked "brightness" actually is "gamma". It affects midtones more than darks.
And, I have actually checked the game's color numbers on a windows system that had a 16 bit display (sigh). Sure enough, it wants to use (0,0,0) and (8,8,8) to draw smooth stone (on moody) -- and expects you to be able to tell them apart.
* Promoting this week: Captive Minecraft 4, Winter Realm. Aka: Vertical Vanilla Viewing. Clicky!
* My channel with Mystcraft, and general Minecraft Let's Plays: http://www.youtube.com/user/Keybounce.
* See all my video series: http://www.minecraftforum.net/forums/minecraft-editions/minecraft-editions-show-your/2865421-keybounces-list-of-creation-threads
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?
What's up?
True story... if it DOES work with Optifine, you'll have to report the bug immediatly :3
I don't know if you were playing at all, but there are several blocks in vanilla Minecraft that don't take colored light to well. Some just look weird, coloring moving pistons will actually crash the game.
As for the other bazillion mods you also installed, I have no idea. Please be aware... you are trying to run a BETA in full release with other mods... I can't guarantee that will work, especially if there are already bugs in the interactions with vanilla. I take the WIP tag seriously, and a BETA means that it's not fully released yet =)
Try removing a couple of mods you think might be the problem, try generating a superflat world, mess around, see what replicates the error, and what doesn't.
One of the many issues:
I made a mistake involving the call to "getMixedLight" It returns the skylight value + blocklight value + the color of the light... This is breaking a lot of things, and I'm going to remove the color in the next version. That will stop a lot of the terrible crashes. I'll make another helper method that actually gives me the data when I need it, but for now, you'll have to be content with the beta status I'm able to put out... It's a beta... nothing more.
Although, fancy screenshots are appreciated =D
I would like to make a suggestion of adding a config option or options to allow people to individually customize their lighting experience, in terms of how your lighting is calculated.
For example, I don't really like simply tinting a block with a particular color as in, with a red light, the blocks next to the red light are almost entirely pure red in tint (or maybe not almost, but completely).
I'd like to propose an alternative formula for calculating colors. Let's say you have a red light. RGB is 15,0,0 and L is 15.
As the light propagates, I would like to suggest that the initial color next to the light starts out bright white with a hint of red, and as you go further from the light, the light gets darker, but the color becomes more apparent.
L(x) = 65520 + i - ((~RGB << 4) / (1 + (r - x)), where x is distance in steps from light source, r is the range of the light, RGB is color, and i is light intensity.
Not sure how that formula would look exactly, but it kinda represents what I am describing.
But in other words, what I am trying to describe is that lights should look more like lights and less like color tinters. Then again, maybe this could simply be a matter of using the right RGB values in the lights. Like a red light shouldn't be 15,0,0 but maybe 10,0,0.
Config - Possible
Config that alters lighting behavior - Possible
Lighting behavior that looks a little more washed out? - Possible
What you described - Not quite possible. Although I'm spending a lot of time with another individual overhauling the smooth lighting renderer. What you are describing sounds something akin to just flattening out the color curve... which will kinda work.
Sadley, the issue is that r,g,b and l, are all separate channels of light. There wasn't a really easy way for me to get anything to behave quite right, so I had to make this compromise... and force colors to behave and degrade from 0 to 15, resulting in the odd color saturation.
There are a couple of things in the pipeline that will kinda address this issue.
One: colored sunlight. At the moment, these lights look terrible in the sunlight, because they are able to override white light, and they make things look darker and more saturated.
If I make it so sunlight has a white color (as opposed to NO color) then the colored lights won't be able to propagate through and make daylight areas look as terrible...
I also would make sunlight on a per-dimension basis, so the end could be blue, the nether purple, and the over-world white...
Two: If I don't like where the block light ended up, I can always make all colors across the board appear less saturated.... leaving everything else in place. That's going to be for later though.
Yes they behave like "tinters" at the moment, which works in all cases, except sunlight... which I agree, needs work.
Hopefully that addresses some of your concerns.
Why that exact colors if not a secret? Purple nether and blue end make a little sense to me, especially interacting with main blocks in that dimensions it would create some weird visuals. Surely nether is hell, so why would not it be red-ish?
Also, can you make skylight color dependent of the time of day? Blue at night, white/bright yellow at day and red at dawn/sunset?
Would you please talk to Xcw about Mystcraft support?
Short summary: Mystcraft lets you create dimensions with multiple suns; each sun can have a different color. Right now, only the "sunset" color has any effect, but if this would allow for actual different colored light from each sun ...
* Promoting this week: Captive Minecraft 4, Winter Realm. Aka: Vertical Vanilla Viewing. Clicky!
* My channel with Mystcraft, and general Minecraft Let's Plays: http://www.youtube.com/user/Keybounce.
* See all my video series: http://www.minecraftforum.net/forums/minecraft-editions/minecraft-editions-show-your/2865421-keybounces-list-of-creation-threads
(In regard to a mod that gives realistic animal genetics):
Would you really rather have bees that make diamonds and oil with magical genetic blocks?
... did I really ask that?