What Is Bit Depth and Why Should I Care?

Hi, I'm Ben, today we're talking about bit

depth

. Bit

depth

is a setting that is attached to every image you've opened in Photoshop and has a big impact on the quality of your image, so let's take a look at

what

it is. It is the smallest piece of computer memory that we can change and bit depth means how many small bits of memory we are going to use to describe each pixel that makes up our image. Well, that will determine how much we can change a pixel and let's get started. with the lowest we could go, let's say we only used one bit of memory to describe each pixel, well a bit can only be on or off, it can actually be a zero or a one, it can't be anything else and if those Those are all the options you have.

So this video wouldn't look like this because this isn't even a bit, it's a bit, it can only have pure black and pure white, so this is

what

it looks like, it's pretty low tech, but if you think about when we got it for the first time. As for graphics, when I started using computers we could only use text and when graphics came out memory wasn't cheap, even to create this video here it would be too many pixels, we couldn't afford a computer with that much memory. so let's reduce the number of pixels that make up this video so it looks more like back in 1972, when I played my first video game, which was pong and pong, I used solid black and solid white because that's all we had before.

It had text and pong it was kind of fun to play when you were used to checkers and chess, but then we progressed and in 1984 the Mac came out, that's when I bought my first Mac and we still only had solid black and white, but we had more memory. and therefore we could use more pixels and therefore these pixels would become smaller this way and back then we had a program called mac paint and mac paint was the precursor to photoshop but all the graphics we made were black and white solid because we were so limited in the amount of memory we could use, well, let's get out of this original looking video.

Look at what would be a modern use for 1-bit graphics. It looks like it belongs in the 1970s. Well, it largely does, but there are still some modern uses. Let's say you have a label printer on your desk. I have one that uses this little cartridge and this cartridge only has white tape and if you heat it it turns black, it can't produce anything but white or black. If I want to print my logo or graphic on this I need a one bit image equal to your restaurant receipt printer and then another example would be if you're driving down the road and you see a construction sign and all it is is a grid . of light bulbs and you have to tell it which one

should

be on and which one

should

be off.

Well, you need a one-bit file to do that, but let's see how we can do it today in Photoshop. Well, here's a photo I took in Lisbon and I think this would look pretty good in solid black and solid white. I think I could use it at the top of a restaurant receipt. If that's the case, let's remove all shades of gray so you have a couple of options. The first is that I can choose image settings and there is an option called threshold and the threshold will make the image from solid black to solid white and then I can move this slider to determine if we get more white or go the other way if we get more. black and somewhere in between and I can move it around until I think the image looks better and then click OK.

Just because this image now only contains solid black and white doesn't mean we're just going to use up a little bit of memory. to store it, in fact, if you look here at the file name and look to the right, it says rgb 16. This is a 16 bit image, we need to do something else if we really want to save some space when we store this image, so let's do it , let's go to the image menu, we'll choose the mode and here it says 16 bits per pixel, that's a lot, so let's go down to eight bits first and then go back there and see what we want. we end up in this thing called bitmap, which means just a grid of bits, individual bits, which can only be solid black and solid white, how do we get there?

Because it's dimmed well if you have a color image, which means it's in a mode. with that color you need to go to grayscale, it will ask you to remove your colors and then you can go into the bitmap here and when you do this it appears, it asks you for resolution because it assumes that you are going to send this. I go to a restaurant receipt printer or something similar and want to know how small dots that printer can make. My printer on my desk can do 360 dots per inch, so that's what I type and then down here you're supposed to have normal grayscale. photography and want to know what to do with all the shades of gray in your image, so if you click here you have a few options.

Now we have applied the threshold and it would be like this. The choice here. Threshold 50 means you just leave the slider. in the middle, well, if you have an image that looks more like a photograph than this one, that looks more like a graphic, then you'll want to use one of these options to try to simulate shades of gray, even if you only end up with Solid Black and White solid Let's try it on a different image. This one won't look good with solid black and solid white. You can find out, just come to the threshold and take a look.

It doesn't matter where you move this. I'm not necessarily going to be happy with this, so let's first convert the image to grayscale. I guess I can choose not to show again and then let's go to the bitmap, but this time let's use diffusion blur. What that will do is try to simulate. shades of gray even though we're just using solid black and solid white, let's see how that looks, that looks terrible, well, it only looks terrible because we're zooming out on the image that you want to zoom in at least one hundred percent. that's the only time you can see all the pixels that make up your image and now I can see that the image is made of solid black and white pixels, but it did a good job of trying to make it look like I still have something that looks like tones of gray, so that's what I would like to use if I wanted to put my face on the top of a restaurant receipt because it can only print solid black on solid white.

Now I realize that some of it is not very exciting. unless you're in the restaurant receipt printing business, so let's make a progression, let's start using more bits per pixel, we can go from one bit at the most basic level to 32 bits at the ridiculous end, so let's move slowly . and see when we want to use it and what it will do with our image, so every time you add another bit of memory to describe each pixel that makes up your image, it will double the number of options for what that pixel can look like instead of having just the possibility of solid black and solid white, suddenly you have four shades of gray and it looks like this and so with four shades of gray that could be used when memory started to get a little less cheap. than when we used solid black and solid white and that's when we could have four shades of gray.

Back then I played a game called sprint or sprit 2 because it had two wheels but it let you drive around this little track all that was on the screen was four shades of gray but damn that was way better than when we just had solid black and solid white, then someone figured out how inside the computer, why not define instead of having four shades of gray to find some colors? to use and then we started to have more fun arcade games like Space Invaders and there instead of having shades of grey, we had colors and that's because of something called a color table, but I'll describe it at a time when color appeared, suddenly we had more interesting games and they started spending more money on memory and suddenly we could have 16 colors and with 16 colors that means we were using four bits of color and that's when I had games like Donkey Kong Joust Defender Frogger uh Miss Pac- Man or regular Pac-Man and you had the arcade full of these games, well we could only use 16 colors because memory was still pretty expensive, although we were finally able to get memory cheap enough where we could use eight bits and with eight bits It had 256 different colors that we could use, that's also what we had when the original Macintosh 2 came out.

The first type of color Mac featured 8-bit graphics, but those 8 bits were 8 indexed color bits. Let's find out what the hell that means. All that means is that every possibility with those little bits that could be 0 or 1 was assigned a color and it was just put in a table and stored somewhere whenever it found all the zeros it turned black, once it turned white and some of the different combinations would create other colors based on the color chart, so let's get rid of these weird graphics, what are we using this for today? Because I don't want to think about going back to childhood.

Well, we still use these concepts in Photoshop today. time, so let's take a look at how we can go from one bit to eight bits of what is known as indexed color. indexed means there is a color table somewhere. Well, here goes, let's say I would like to save this image and display it. On the Internet you would have a couple of different options, let's explore the one that has to do with bit depth. To do this, I'll go to the File menu, choose export, and choose save. for web in save for web there are two file formats that allow us to choose how many bits we are going to use to store the image in those two file formats they are gif and the second one is ping eight, the eight represents eight bits, so we can choose any of which will generally offer the same options, so now we are set to two colors, which means a total of one bit of memory.

Any time you only have one bit of memory, you only have two options, but in this case, instead of having black and white, we have two colors, because look here is a color chart and that just means that you looked at the image to determine what colors he thought would be best and assigned them here that we could really influence. that when playing with this menu ends up determining how those colors are chosen, but when you only work with two colors I doubt you will see much difference between them, but then you notice how it simulates shades of gray, remember that choice.

This is called dithering because there is an option here called dithering and we can control how much there is by reducing the dithering to zero. You would see that it's turned off, but you might want to add a little bit because not everything looks like a photo, you might just have a logo that has some shading and a little bit of dairy might look better than a lot, or you can come here and turn it off for complete. I'm going to turn mine off completely so we can do It's pretty obvious that we're only at two colors, so let's go to the color menu and click on it.

Let's look at all the options in this menu, each one as you go down the menu doubles the number of colors, remember that each one. Every time you add another bit of memory, you double the number of options you have, so this goes from one bit, to two, to three, to four, to five, to six, to seven, to eight, and if we choose this , we will use eight. times more bits of memory than if we used this, so as we get up to here, the image will start to look better as we get more and more tones, but we will also increase the amount of memory and the amount of hard drive space . that this image takes up and I saved it using each of these settings and here are the different file sizes I ended up with and it's not going to perfectly equate to how many bits it's using because there's other information stored in a file like the date it was saved. created, what file format is that kind of thing, but in general you'll see that every time I go to more bits, it takes up more space on my hard drive.

Now down here is our color chart, this shows us all the colors. that are being used and this is 256 colors, if I go to 128 you will see that this is only half full and if I go to 64 it will be a quarter full and so on, so it just shows you how many colors you are using knowing that the smaller whatever number you choose the smaller your file size is going to be and the higher the number the bigger and it's a compromise and most of the time you come here and if it's a photograph I would set this to broadcast and I usually have it set at 100 and then I come in here and say, okay, how much can I get away with?

So let's save this image with 16 colors, there are 16, I'll hit the save button. and then I'm going to reopen this and speed it up so you don't have to look here is the image open in Photoshop and let's take a look, first take a look in the layers panel, look at everything. The icons at the bottom are grayed out because you can't use layers when you have this type of file. If I come here and try to apply a filter, none of them are available if I try to adjust the image many times. Some of the settings I'm used to using are not available and the ones that are will act a little strange.

If you want all of those things to work, then you have to go to the image menu, choose mode and notice that we are in a strange mode, whenever you have an eight-bit indexed color, you will bein index color mode and there will always be a color table involved. Remember that when we were in Save for web it showed you the colors that we can see. them again if we choose this there they are, we can even change them if you were to come in here and click on one. You could change this to blue and that sign suddenly turns blue because you just replaced it.

With whatever is here, I'm going to click cancel because I wanted the sign to be the original color. If I want the layers to be filters available and my settings to work normally, then I need to choose the rgb image mode rgb is what normal images are. and now all my normal settings are available, I can use layers and if I try to apply a filter, they're all here, so if I need it to just be 16 colors when I'm done using those functions, I just save it again. again as a png file or a gif file with 16 colors just like I originally did.

Now I'm not suggesting you use the gif or ping file format for photos most of the time, jpeg would look much better when you're saving something for the internet. but if you have a graphic like a logo and that logo has like a three-dimensional ball inside it, the rest is all text, well, having a little bit of dithering on that ball and reducing the number of colors would make it dramatically smaller and it would look a lot better than jpeg jpeg doesn't look good with graphics, it's good with photographs, but so far we've used the non-exciting bit depth part next time I'll show you how to go from eight bit indexed. color where that color table is involved, then something known as eight bits per channel, which means eight bits of red, eight bits of green, and eight bits of blue, then we can continue with 16 and 32 bits, but I want to make sure that it has a general idea. because a bit is just remember that a bit is the smallest part of memory that we can use, it is something that can be a zero or a one, if we only use one of them, we can only have solid black and solid white if we get two from them. we double our options to four, like four shades of gray, then if we add another bit, we double it again to eight and we can continue up to eight bits if we're going to use a color table, but once you've gotten past a color table, if you want more colors, suddenly you want to get into 8 bits per channel, where you don't have a color table, that's what we'll look at next week.