My good friend Carl F. Corneil has made a fantastic After Effects script that allows you to do lip sync in After Effects in a second! Usually, you'll have to key every mouths opacity manually every time you want the mouth shape to change. Luckily, not anymore! This scripts allows you to slide a controller in the Y-axis to automaticly change to your desired mouth.
This is a quick and easy tutorial on how to use the script:
- First, download the script at the bottom of this post. Unzip the .jsx-file, and add it to the scripts-folder in your After Effects folder.
- Open After Effects (should even work on old versions).
- Make a new composition and import your mouth layers. Position them as you like.
- Then, go to Layer->New->Solid (you can also use a null object). Make the solid a small size like 100x100 or something. Name the layer "controller".
- Select the "controller" layer first, and then select all the other mouth layers. Go to File->Scripts->CarlsOpacityYController
Voila! The opacity on your mouths are now magic! Try taking the "controller" and move it in the y-axis (up/down)! As you can see, the mouths change according to where the controller is positioned in the y-axis.
In the "effects"-panel, there is a slider control. Adjusting this value, changes the amount you'll have to move the controller for the mouths to change. A really handy feature if you use very many/few mouths!
Thank you so much to Carl F. Corneil for making our lives easier!
Download CarlsOpacityYcontroller_v.1.5
Download example project
Viser innlegg med etiketten tutorial. Vis alle innlegg
Viser innlegg med etiketten tutorial. Vis alle innlegg
24 august 2011
31 oktober 2010
Basic color correction in Photoshop CS4 and CS5
In this tutorial, I’ll explain how to do basic, but proffesional looking color correction in Photoshop CS4 and CS5. Photoshop isn’t that hard if you get the first step inside. And that is what I’ll try to help you with.
There is mainly two ways to work with a photo (either in photoshop or elswhere); destructive and non-destructive. What is destructive and non-destructive, you ask? Well, the easiest way to understand this, is to think a real life situation. You got a picture in front of you and you want to paint a tree on it. The destructive way to make that tree happen, is to pain right on the photo. The non-destructive way is to lay a transparent paper on top of the photo and paint on that one, so if you make a mistake, the original picture is not destroyed. This is really how we want to work in photoshop too. We’ll apply «filters» on top of the photo, instead of just working directly with the original photo. Adobe have made this quite easy with their newest versions of Photoshop.
I’ll use a .jpg in this tutorial, as an example, but I recommend you use a raw image (16-bits) if you can. Read one of my previous guide why.
I’ll start off using a picture of my friend Kine. The original photo is not really interesting, but with a few easy touches, this image will be a totally new experience. This is what we'll try to achieve:
What I’ll focus on in this tutorial, is adding adjustment layers (filters) to the image. I will not use any tools, retouching or other things for now. That’ll be another tutorial. So ok, lets open the picture and get started!
First, download the image of my friend Kine (click HERE). Open photoshop, and choose File > Open. Choose «Kine_Original» Now, in the layers window there is a button for adding different adjustment layers. Press this button, and a box will appear. These are the different adjustment layers we can use. I’ll cover the most used ones.
Brightness/Contrast
The brightness and contrast is quite self explainable. You adjust the brightness and contrast in the image. Usually, I don’t use this feature, because I think «Levels» and «Curves» do a better job, but sometimes the Brightness/Contrast is the best way to achieve the desired result.
Levels
The levels adjustment is a great way to work with the contrast and/or seperate colors in the image. You can work either with the blacks, the middle tone or the whites. If you take one slider and drag it, you see the contrast changes, and your chosen parameter starts to increase. This is a perfect way to for example remove grey areas.
At the bottom in the levels box, you’ll se the output colors. These are the main colors as a starting point. If you adjust these, the black will become lighter or the whites will become darker.
On top of the levels-box you can see a drop down bos that says «RGB». If this is correct, you’re working with all colors at the same time. In other words, you’re working with the contrasts. I you want to, you can choose either red, green or blue in stead, and now you can adjust only one color separately. This gives you enormous power, and is a feature we’ll use a lot in other adjustments layers too.
Curves
This is maybe my favourite adjustment layer. The possibilities is almost endless if you know how to handle it. The curves works a bit like the levels adjustment layer, but instead of just a line, you got both x and y to work with. To make this easy; down to the left is where the shadow (blacks) information lies, and on the top right the hilight (whites) lies. So if you make a point on the line by clicking on it, you may drag the point up or down to make the selected color range either darker or brighter. In example; if we make two points on the line, one a bit above the middle, and one under it. We drag the above upwards a bit, and the one below a bit down, the picture suddenly got a lot more contrast. The coolest thing about curves, is the ability to choose the different colors (Red, Green, Blue) and manipulate those separately. Try open the drop down menu that says «RGB» and choose either red, green or blue. If you now make a point and drag it up or down, you can see that the whole image is changing its colors. This is a great way to adjust your image. The Curves adjustment layer doesn’t add any color to the image, it simply adjusts the ones that’s already there.
Exposure
The exposure is how much light there is in your picture. It’s the same parameter that you adjust directly on your camera before you take a picture. This is good if the picture is too dark or too bright. Try play with the exposure, offset and gamma together to get the best result.
Vibrance
Vibrance is almost the same as saturation, but Vibrance only uses the colors closeby, and does not affect the whole image all at once. If you turn the Vibration all the way down, the image won’t become totally black & white. You’ll still see some tones in your image. A great way to tone down the colors in your image without crushing them too much (either way, up or down). The saturation doesn’t calculate the neary colors, so it affect the whole image all the way from original to black & white, or up to strong oversaturated colors.
Hue/Saturation

There is a distinct difference between hue and saturation. Hue is the version of the colors. If you drag the hue slider, the whole image will start to change colors into something that looks quite surrealistic. The saturation adjustment is the strength of the colors. On the negative site, you got totally weak and black & white colors, and on the positive site you got strong almost crushed colors. And of course you got the lightness slider that adjust the blacks and whites in your image.
The awsome thing about hue, is the ability to change the values of the colors separately. On default, the settings is set to «Master», but if you open the drop down menu, you can see there is many separate colors to choose from. Try choosing a separate color and adjust the hue... The perfect way to change colors for spesific objects. If you want to change the color of your jacket from yellow to green, just select yellow in the hue drop down menu and adjust the hue!
Color Balance
Color balance is the only adjustment layer that actually ads colors to the image. If you got a black & white image, and add a color balance to it, the image changes from black & white to the color you choose. This is a good way to for example add some yellow highligts to your picture, or some blue shadow tones.
You can choose between shadow, midtones and hilights. Those represents the types of colors you choose to adjust. If you select tha shadow tones, only the dark places in the image get affected, but if you choose the highlighs, only the bright colors changes.
Black & White
The best way to make your image black & white. Of course you could adjust the saturation all the way down, and play with the contrasts in the curves or levels adjustments, but in this adjustment layer, you got so much more control over each color. Just try it, and you’ll know what I talk about.
Selective Color
Selective color is the concept from the separate color tones in hue/saturation taken a step further. You can choose a main color and tweek it endlessly until you get the one you like. The relative button works almost like vibrance with nearby colors and the absolute button uses all colors in the image.
And these are actually the main adjustments you use for color correction! So now that you know a bit about each adjustment layer, let’s get to work!
Now, the image is a bit dark, so I’ll brighten it up with some exposure. Add a «exposure adjustment layer» and tweek the image to your liking. Remember that you really don’t have to make big changes for the picture to turn out great, so I’ll only turn the exposure up to +0,52, and I’ll leave the other parameters alone.
Now, I want a bit more contrast in the image, so I’ll add a «Curves», and start adjusting it to something like this (see illustration). And when I first work with the curves, why not adjust the colors a bit. Click the RGB dropdown menu, choose the colors and adjust to your liking. You may be able to see on my curve almost what I did with the different curves. The main tweek is the red channel that I’ve darkened a bit.
And actually finally I’ll just play around in the selective colors. There are so many parameters that I wont bother you with all the numbers. But just play around and see what fit your picture the best!
I’ve only used some few of the adjustment layers for this picture, but often you won’t need more than two or tree layers to make your picture look awsome. And it really doesn’t need to be so hard. The things I’ve done today is fairly easy steps, and if you start to play around, you’ll see how easy it is to lift your picture to the nexl level.
Here is the difference between the original and the photoshopped image:
So good luck, and I hope you have learned a little bit from this tutorial. If not, you can always go back and read about the different adjustment layers.
A big thank you to my wife who actually took this image, and helped me with this tutorial.
-Torgeir Bull
You may also download a PDF-version of this tutorial HERE.
There is mainly two ways to work with a photo (either in photoshop or elswhere); destructive and non-destructive. What is destructive and non-destructive, you ask? Well, the easiest way to understand this, is to think a real life situation. You got a picture in front of you and you want to paint a tree on it. The destructive way to make that tree happen, is to pain right on the photo. The non-destructive way is to lay a transparent paper on top of the photo and paint on that one, so if you make a mistake, the original picture is not destroyed. This is really how we want to work in photoshop too. We’ll apply «filters» on top of the photo, instead of just working directly with the original photo. Adobe have made this quite easy with their newest versions of Photoshop.
I’ll use a .jpg in this tutorial, as an example, but I recommend you use a raw image (16-bits) if you can. Read one of my previous guide why.
I’ll start off using a picture of my friend Kine. The original photo is not really interesting, but with a few easy touches, this image will be a totally new experience. This is what we'll try to achieve:
What I’ll focus on in this tutorial, is adding adjustment layers (filters) to the image. I will not use any tools, retouching or other things for now. That’ll be another tutorial. So ok, lets open the picture and get started!First, download the image of my friend Kine (click HERE). Open photoshop, and choose File > Open. Choose «Kine_Original» Now, in the layers window there is a button for adding different adjustment layers. Press this button, and a box will appear. These are the different adjustment layers we can use. I’ll cover the most used ones.
Brightness/ContrastThe brightness and contrast is quite self explainable. You adjust the brightness and contrast in the image. Usually, I don’t use this feature, because I think «Levels» and «Curves» do a better job, but sometimes the Brightness/Contrast is the best way to achieve the desired result.
Levels
The levels adjustment is a great way to work with the contrast and/or seperate colors in the image. You can work either with the blacks, the middle tone or the whites. If you take one slider and drag it, you see the contrast changes, and your chosen parameter starts to increase. This is a perfect way to for example remove grey areas.
At the bottom in the levels box, you’ll se the output colors. These are the main colors as a starting point. If you adjust these, the black will become lighter or the whites will become darker.
On top of the levels-box you can see a drop down bos that says «RGB». If this is correct, you’re working with all colors at the same time. In other words, you’re working with the contrasts. I you want to, you can choose either red, green or blue in stead, and now you can adjust only one color separately. This gives you enormous power, and is a feature we’ll use a lot in other adjustments layers too.
Curves
This is maybe my favourite adjustment layer. The possibilities is almost endless if you know how to handle it. The curves works a bit like the levels adjustment layer, but instead of just a line, you got both x and y to work with. To make this easy; down to the left is where the shadow (blacks) information lies, and on the top right the hilight (whites) lies. So if you make a point on the line by clicking on it, you may drag the point up or down to make the selected color range either darker or brighter. In example; if we make two points on the line, one a bit above the middle, and one under it. We drag the above upwards a bit, and the one below a bit down, the picture suddenly got a lot more contrast. The coolest thing about curves, is the ability to choose the different colors (Red, Green, Blue) and manipulate those separately. Try open the drop down menu that says «RGB» and choose either red, green or blue. If you now make a point and drag it up or down, you can see that the whole image is changing its colors. This is a great way to adjust your image. The Curves adjustment layer doesn’t add any color to the image, it simply adjusts the ones that’s already there.
Exposure
The exposure is how much light there is in your picture. It’s the same parameter that you adjust directly on your camera before you take a picture. This is good if the picture is too dark or too bright. Try play with the exposure, offset and gamma together to get the best result.Vibrance
Vibrance is almost the same as saturation, but Vibrance only uses the colors closeby, and does not affect the whole image all at once. If you turn the Vibration all the way down, the image won’t become totally black & white. You’ll still see some tones in your image. A great way to tone down the colors in your image without crushing them too much (either way, up or down). The saturation doesn’t calculate the neary colors, so it affect the whole image all the way from original to black & white, or up to strong oversaturated colors.Hue/Saturation

There is a distinct difference between hue and saturation. Hue is the version of the colors. If you drag the hue slider, the whole image will start to change colors into something that looks quite surrealistic. The saturation adjustment is the strength of the colors. On the negative site, you got totally weak and black & white colors, and on the positive site you got strong almost crushed colors. And of course you got the lightness slider that adjust the blacks and whites in your image. The awsome thing about hue, is the ability to change the values of the colors separately. On default, the settings is set to «Master», but if you open the drop down menu, you can see there is many separate colors to choose from. Try choosing a separate color and adjust the hue... The perfect way to change colors for spesific objects. If you want to change the color of your jacket from yellow to green, just select yellow in the hue drop down menu and adjust the hue!
Color Balance
Color balance is the only adjustment layer that actually ads colors to the image. If you got a black & white image, and add a color balance to it, the image changes from black & white to the color you choose. This is a good way to for example add some yellow highligts to your picture, or some blue shadow tones.
You can choose between shadow, midtones and hilights. Those represents the types of colors you choose to adjust. If you select tha shadow tones, only the dark places in the image get affected, but if you choose the highlighs, only the bright colors changes.
Black & WhiteThe best way to make your image black & white. Of course you could adjust the saturation all the way down, and play with the contrasts in the curves or levels adjustments, but in this adjustment layer, you got so much more control over each color. Just try it, and you’ll know what I talk about.
Selective Color
Selective color is the concept from the separate color tones in hue/saturation taken a step further. You can choose a main color and tweek it endlessly until you get the one you like. The relative button works almost like vibrance with nearby colors and the absolute button uses all colors in the image.And these are actually the main adjustments you use for color correction! So now that you know a bit about each adjustment layer, let’s get to work!
Now, the image is a bit dark, so I’ll brighten it up with some exposure. Add a «exposure adjustment layer» and tweek the image to your liking. Remember that you really don’t have to make big changes for the picture to turn out great, so I’ll only turn the exposure up to +0,52, and I’ll leave the other parameters alone.
Now, I want a bit more contrast in the image, so I’ll add a «Curves», and start adjusting it to something like this (see illustration). And when I first work with the curves, why not adjust the colors a bit. Click the RGB dropdown menu, choose the colors and adjust to your liking. You may be able to see on my curve almost what I did with the different curves. The main tweek is the red channel that I’ve darkened a bit.And actually finally I’ll just play around in the selective colors. There are so many parameters that I wont bother you with all the numbers. But just play around and see what fit your picture the best!
I’ve only used some few of the adjustment layers for this picture, but often you won’t need more than two or tree layers to make your picture look awsome. And it really doesn’t need to be so hard. The things I’ve done today is fairly easy steps, and if you start to play around, you’ll see how easy it is to lift your picture to the nexl level.
Here is the difference between the original and the photoshopped image:
So good luck, and I hope you have learned a little bit from this tutorial. If not, you can always go back and read about the different adjustment layers.
A big thank you to my wife who actually took this image, and helped me with this tutorial.
-Torgeir Bull
You may also download a PDF-version of this tutorial HERE.
29 oktober 2010
8-bits and 16-bits vs. 32-bits
The difference between 8-bits color and 32-bits.
Every photographer talks about taking pictures in a «raw-format» because the image has much more information to work with, but what are really the difference? Why is 32-bits images better than 8-bits or 16-bits?
I won’t use the time to explain what a bit is. It’s the computers way to calculate things. In images this is quite important. The bits is (in a nutshell) how much information there is in the picture.
An 8-bits image got a color range from 0 to 256. That means that each color (red, green, blue) has 256 variations. Roughly 16.8 million colors to work with. Not bad, and every .jpg-picture is actually 8-bits.
16-bits images has 65 536 possible values per color. In other words 281 trillion color possibilities! This is far mor colors than the eye is capable of seeing, but in the editing process, more information is never a bad thing.
So why is there 32-bits? 32 bits is actually 24-bits color and 8-bits of non color information. Instead of a range between 0-255 or 0-65 536, 32-bits images uses the values between 0 to 1. 0 is black and 1 is white. But (and this is the big but), the image can actually contain values lower than 0 and higher than 1.
If you try to adjust the colors in 8-bits or 16-bits images, the black cannot get any darker than 0 (black), and the whites cannot get any brighter than 256 or 65 535 (white). But in the real world, there may be things that is actually «brighter» than white or «darker» than black. For example a lightbulb, a flashlight or the sun. You cannot get this information from an 8-bits or 16-bits image.
This is where 32-bits comes into action. 32-bits images contains an extra 8 bits with this light-information. In other words, like a dream to work with compared to 8-bits images, and destined to give you great results!
The only downside, is that there is no camera to date that has the capability to take a picture in 32-bits. A single lens reflex camera has the maximum of 16-bits. But if you make a HDR (High Dynamic Range Image), this image has the possibility of 32-bits. And if you work in 3D or with user created content in a graphical software like Photoshop or After Effects, you may create images (or video) that has the 32-bits information.
And maybe, one day the real world is ready for real 32-bit images.
8-bits image vs. 32-bits HDR-image (see the difference around the sun):
Every photographer talks about taking pictures in a «raw-format» because the image has much more information to work with, but what are really the difference? Why is 32-bits images better than 8-bits or 16-bits?
I won’t use the time to explain what a bit is. It’s the computers way to calculate things. In images this is quite important. The bits is (in a nutshell) how much information there is in the picture.
An 8-bits image got a color range from 0 to 256. That means that each color (red, green, blue) has 256 variations. Roughly 16.8 million colors to work with. Not bad, and every .jpg-picture is actually 8-bits.
16-bits images has 65 536 possible values per color. In other words 281 trillion color possibilities! This is far mor colors than the eye is capable of seeing, but in the editing process, more information is never a bad thing.
So why is there 32-bits? 32 bits is actually 24-bits color and 8-bits of non color information. Instead of a range between 0-255 or 0-65 536, 32-bits images uses the values between 0 to 1. 0 is black and 1 is white. But (and this is the big but), the image can actually contain values lower than 0 and higher than 1.
If you try to adjust the colors in 8-bits or 16-bits images, the black cannot get any darker than 0 (black), and the whites cannot get any brighter than 256 or 65 535 (white). But in the real world, there may be things that is actually «brighter» than white or «darker» than black. For example a lightbulb, a flashlight or the sun. You cannot get this information from an 8-bits or 16-bits image.
This is where 32-bits comes into action. 32-bits images contains an extra 8 bits with this light-information. In other words, like a dream to work with compared to 8-bits images, and destined to give you great results!
The only downside, is that there is no camera to date that has the capability to take a picture in 32-bits. A single lens reflex camera has the maximum of 16-bits. But if you make a HDR (High Dynamic Range Image), this image has the possibility of 32-bits. And if you work in 3D or with user created content in a graphical software like Photoshop or After Effects, you may create images (or video) that has the 32-bits information.
And maybe, one day the real world is ready for real 32-bit images.
8-bits image vs. 32-bits HDR-image (see the difference around the sun):
04 september 2010
How to prank your friend with fake mails!
Warning: This should not be used for illegal purposes. I don't take any responsibility for anything you do after reading this guide. This is made for fun purposes only!
So you want to prank your friends? Well, what is better than sending them a fake mail? It's a bit complicated, but it's sure fun!
Well, to do so, we need to find the ISP (the ones distributing internet for you), log into their mail server, and send a mail manually, telling the server what adress and content.
So, how do you do so? You use Terminal in OSX (this is possible in MS-DOS also, but I'll cover OSX for now)! So let's start:
Start with opening Terminal on your mac (applications->Utilities).
Type: telnet
Now, we need to find out what ISP you are using. visit http://whatismyipaddress.com/ to find out. Then visit their webpage to find out their mail server adress (smtp/pop a.s.o.) Should look like this: mail.domain.ext in example; smtp.itfactory.com.
Type: o mail.domain.ext 25
(a small o, not a 0 (zero))
25 is the port all mail is shipped from, therefor we need to specify that after the mail-adress.
You should receive something like:
Trying xxx.xxx.xxx.xxx ... Connected to mail.domain.ext. Escape character is '^]'. 220 mail.domain.ext ESMTP Sendmail ?version-number?; ?date+time+gmtoffset?
Now, you're in but you have to say hello to "walk around".
Type: helo domain.ext
(or in this example; helo itfactory.com. Why only one L? I don't know)
You get:
250 mail.domain.ext Hello local.domain.name [loc.al.i.p], pleased to meet you
Now, we'll have to give the mail adress to send from, so
Type: mail from:desired@fakemail.com
(or net, org, biz or whatever)
(no space between : and desired mail adress)
Answer:
250 2.1.0 desired@fakemail.com... Sender ok
Now, we need the recipients adress, the frend you're going to prank.
Type: RCPT to:friends@mail.com
(no space between : and mail adress)
Answer:
250 2.1.0 friends@mail.com... Recipient ok
Type: data
Answer:
354 Enter mail, end with a single ".".
Now, you're ready to write the mail. Maybe you should start with a subject.
Type: Subject:Desired Subject
(no space between : and subject)
Then, press "enter" twice. The next thing you type will show as the text in the mail. You may press "enter" as much as you like now.
Type (example):
Hi friend!
This is a fake mail from a fake mail adress just to
show off how awsome I am, and ho show how easy it is to
use fake mail adresses to prank friends.
Cheers!
When you're finished writing the mail, just press enter twice, press . (dot)
and enter once again. The mail is finished, but not yet sent. The only thing left to do is to
Type: quit
You'll get the final answer:
221 2.0.0 mail.domain.ext closing connection Connection closed by foreign host.
And you're done.
Problem solutions:
501 nouser@nosuchplace.here...Sender domain must exist .
The domain that you are sending from must exist
550 mail@domain.ext... Relaying Denied
You're probably not uing the right ISP. Try to find out once more what ISP you're using.
So you want to prank your friends? Well, what is better than sending them a fake mail? It's a bit complicated, but it's sure fun!
Well, to do so, we need to find the ISP (the ones distributing internet for you), log into their mail server, and send a mail manually, telling the server what adress and content.
So, how do you do so? You use Terminal in OSX (this is possible in MS-DOS also, but I'll cover OSX for now)! So let's start:
Start with opening Terminal on your mac (applications->Utilities).
Type: telnet
Now, we need to find out what ISP you are using. visit http://whatismyipaddress.com/ to find out. Then visit their webpage to find out their mail server adress (smtp/pop a.s.o.) Should look like this: mail.domain.ext in example; smtp.itfactory.com.
Type: o mail.domain.ext 25
(a small o, not a 0 (zero))
25 is the port all mail is shipped from, therefor we need to specify that after the mail-adress.
You should receive something like:
Trying xxx.xxx.xxx.xxx ... Connected to mail.domain.ext. Escape character is '^]'. 220 mail.domain.ext ESMTP Sendmail ?version-number?; ?date+time+gmtoffset?
Now, you're in but you have to say hello to "walk around".
Type: helo domain.ext
(or in this example; helo itfactory.com. Why only one L? I don't know)
You get:
250 mail.domain.ext Hello local.domain.name [loc.al.i.p], pleased to meet you
Now, we'll have to give the mail adress to send from, so
Type: mail from:desired@fakemail.com
(or net, org, biz or whatever)
(no space between : and desired mail adress)
Answer:
250 2.1.0 desired@fakemail.com... Sender ok
Now, we need the recipients adress, the frend you're going to prank.
Type: RCPT to:friends@mail.com
(no space between : and mail adress)
Answer:
250 2.1.0 friends@mail.com... Recipient ok
Type: data
Answer:
354 Enter mail, end with a single ".".
Now, you're ready to write the mail. Maybe you should start with a subject.
Type: Subject:Desired Subject
(no space between : and subject)
Then, press "enter" twice. The next thing you type will show as the text in the mail. You may press "enter" as much as you like now.
Type (example):
Hi friend!
This is a fake mail from a fake mail adress just to
show off how awsome I am, and ho show how easy it is to
use fake mail adresses to prank friends.
Cheers!
When you're finished writing the mail, just press enter twice, press . (dot)
and enter once again. The mail is finished, but not yet sent. The only thing left to do is to
Type: quit
You'll get the final answer:
221 2.0.0 mail.domain.ext closing connection Connection closed by foreign host.
And you're done.
Problem solutions:
501 nouser@nosuchplace.here...Sender domain must exist .
The domain that you are sending from must exist
503 Need MAIL before RCPT
A recipient has been specified before a sender.
550 mail@domain.ext... Relaying Denied
You're probably not uing the right ISP. Try to find out once more what ISP you're using.
31 juli 2010
Beginners guide to advanced multi-pass render in C4D.
Everybody told me that the way to render out from Cinema 4D was in multi-passes. That you split the image up in different "passes", and put them together again in a compositing software for better adjustments in post production.
I tried to read about how to do this, but I couldn’t find anything good or a bit more advanced about how to. So I figured I’ll learn it by myself and make an "easy to understand" tutorial so that more people could learn this very useful technique in their workflow.
Multi-Pass?
So, what is a multi-pass render? Usually, when you render out from Cinema 4D, you render out a flat image. A so called "Backgound" og "Regular Image". There is no good way to adjust this picture in post production without crushing the colors or destroying the picture. So how do we adjust the intensity of the shadows, reflections, specular and other filters? Multi-passes. A pass is a layer with only the information of the desired layer, i.e. only the reflections or shadows.
I’ll use a scene with a mini-cooper in this tutorial. If you want to follow, you may download the free scene at oyonale.com.
Software used in this tutorial
Cinema 4D 11.5 and After Effects CS4.
To the case
So everything looks near perfect in the preview render with the right light, materials and so on. But you want to have a bit more control over the picture in the compositing.
Lets start with opening the render settings. Now check of the multi-pass box at the left side of the "Multi-Pass" name. This enables multipass, obviously. Now, click the "Multi-Pass..." button. This will enable you to choose whitch passes you want to include in your render. What I like to do, is to render a complete image using only multi-passes, and not use the background image at all. The background image is great for looking at your render, but it includes all the passes (whenever you have enabeled multi-pass or not) inside itself. So if you use the regular image beneath the passes, you’ll just apply information that is already included in the image, it wont look good at all and you won’t be able to turn of for instance the shadows or the reflection, since they’re already there.

Passes
To render at complete image with passes only, you may choose «Add image layers» in the «Multi-Pass... menu» (still in the render settings). You’ll get a quite long list of different passes on the left side: Ambient, Diffuse, Specular, Shadow, Reflection, Refraction, Ambient Occlusion, Global Illumination, Caustics, Atmosphere, Atmosphere (multiply) and Post Effects.
I don’t think you really need all of these to get a good image to work with, so I’ll try to point out whitch you’ll need:
Reflection - Yeah, the reflections.
Refraction - Also called deflections. The reflections and light passing through glass, windows etc.
Shadow - Render all the shadows in the scene. There are different ways to control the shadows (and lights). Usually don’t use this pass. I like to control the lights and shadows like this:
Go to the «render settings» and click the «Multi-Pass»-name. You’ll get som few choises; Under «Separate Lights» choose «All». This will enable you to adjust the lights in the compositing. Under «Mode» choose whatever you wish. I like to choose 3 (diffuse+specular+shadow) channels to get the most out of it, but 2 (diffuse+specular,shadow) works as well. Check the «shadow correction».
But since there is no lights in our scene to add, check the shadow pass to render the shadows as well.
Diffuse - The geometry without any light, shadows, reflections or anything. Just plain ugly geometry. If you enabeled the «seperate lights,» the diffuse pass will be rendered there, and there’s no need for this pass, but I tend to keep it even so. Just play with the opacity and you render could look a lot better. So keep this in mind; the diffuse pass may be handy, even if you got two of them in a composition.
Global Illumination - Probably one of the most important passes to make the scene look realistic. GI is how the light behave. How it bounces from one object to another and lights up more places than just where the light first hit the target (an easy example to understand GI is to rig a table, place a spotlight over it and render. Without GI, the floor under the table is dark, but with GI enabled, some light rays get under there and lighten ut just a bit)
To make the Global Illumination work, you need to enable it to render in the Background image. Press the «Effects...»-button and choose «Global Illumination». You can play around with the settings, but I’ll keep it on default for now.
I have to say, this may be the heaviest pass to render. Sometimes GI is the only way to make the picture work, other times things work just fine without. If you manage to make things work without GI, don’t use it ;)
Ambient Occlusion - If you’re working with a still image that won’t take weeks to render, check this pass. But if you’re working with a longer sequence with animation, this pass could almost double the rendertime. But don’t worry, I got a solution for you in a second.
Render a test image. In the picture viewer (if you use Cinema 11.5), you can play around with the different passes, look at them or put them together to see how they work. You can even adjust the opacity for every pass and find out how your composition could look like in the end.
Make Ambient Occlusion-pass render fast
The problem with Cinema 4D, is that it always have to render out the RGB Background image. So if you only need for instance just a pass, say Ambient Occlusion, C4D will render out the whole scene with materials, textures and so on. This will cause things to get slower than needed. I tried to render out a scene with AO as one of the layers in a multi-pass, and the whole picture took 19 minutes to render. Then I removed the AO-pass, and the picture only took 2 minutes to render. As you may understand, if you got thousands of images, the render will be many times longer with Ambien Occlusion added. But I got a work-around:
Take the c4d-file and duplicate it. Rename it «whatevername_AO.c4d». Open the file and delete every single texture both in the materials and in the objects-window (remove the texture tags). You need a clean scene. Then make a new material. Double click on it and only enable the «luminance» channel. The texture will become totally white making the render a white plane. Perfect. Now, go to the render settings and uncheck everything but the Ambient Occlusion effect applied on the background (not as a mult-pass). Be sure to check the «apply to scene» in the AO editor.
If you now render out the sequence, you’ll see that it won’t take as long as if it was a pass.
Afterwards, you just apply the AO-sequence over the rest of the composition (in AE or whatever). Use «multiply» as the blending mode to make the white colors transparent over the rest of the composition. The AO-render will works as a multi-pass even if it really isn’t.
Kind of wierd, but that’s how it works. In my test, I managed to render out only the AO in 5 minutes. So compared to 19 minutes to render with AO as a pass, to only 7 minutes (AO + the render without the AO(no GI-pass)) if you use this workaround. It will only take you five minutes to do, and may make the render time half as long. A really good tip if you’re working with longer sequences.
Delete the atmosphere, atmosphere (multiply) , the ambient, caustics, ambient and specular.

Pass-research
When you're working with other projects, try rendering all the passes, and browse through in the Picture Viewer to see whom you’ll need in the final rener. Just klick on the «layer»-tab, and choose multi-pass to turn on and off a pass. You should also check out the Single-pass-tab to use as research as well.
The saving and final compositing
Go inside «render settings» again and click on «Output». Find out what resolution you need, and choose how your frame range. I want my render to go from frame 0 to frame 500 in this scene.
Now check the «save» box in the left menu, and click on the name to open the «save menu.»
You don’t actually need to save the «background,» so no need to check the save box in the «Regular Image» menu, since we're doing everything with passes (yeah, thats how good we are!).
To save the multi-pass layers, you need to click «save» in the render settings, and check the save box in the «Multi-Pass Image» menu. Choose «Photoshop PSD» as file format and choose a path to save the passes. 8-bit depth will work fine, but 16 and 32 is also avalible if you’re into those things ;)
All the passes will be saved in one PSD-file as layers, and as one file per frame in the sequence. In other words; longer sequences, more files, smaller disc space.
And now for the life saving option: «Compositing Project File.» When working with multi-passes, you got many files and even more layers, all with different blend modes and opacity. Compositing this together could be a nightmare, but luckily Cinema 4D got a great resolution.

Under the «Muti-Pass Image» menu, you got a menu called «Compositing Project File». This menu may be opened, if not click on the small arrow at the left side of the name. Check save and choose your compositing software. I’ll use After Effects in this tutorial. Last, find a place to save this file. Cinema 4D makes a finished After Effects-file, ready to go! Every pass is pre sorted in layers with the right blending mode.
After Cinema 4D has finished the render, just open the .aec-file in After Effects. Here you can adjust every pass exactly as you wish. Turn down the shadow opacity, adjusting the colors, reflections and so on. It's your call! Now you've finished a sequence made of multi-passes only, the advanced way.
Ps: Remember to import and add the Ambient Occlusion sequence in After Effects if needed!
Post notes
I hope you learned something from this tutorial and hope you play around and try out whats best for your scene. Good luck!
So what do you think? Was this useful information?
-Torgeir Bull
If you want to, you can also download the tutorial as a PDF.
I tried to read about how to do this, but I couldn’t find anything good or a bit more advanced about how to. So I figured I’ll learn it by myself and make an "easy to understand" tutorial so that more people could learn this very useful technique in their workflow.
Multi-Pass?
So, what is a multi-pass render? Usually, when you render out from Cinema 4D, you render out a flat image. A so called "Backgound" og "Regular Image". There is no good way to adjust this picture in post production without crushing the colors or destroying the picture. So how do we adjust the intensity of the shadows, reflections, specular and other filters? Multi-passes. A pass is a layer with only the information of the desired layer, i.e. only the reflections or shadows.
I’ll use a scene with a mini-cooper in this tutorial. If you want to follow, you may download the free scene at oyonale.com.
Software used in this tutorial
Cinema 4D 11.5 and After Effects CS4.
To the case
So everything looks near perfect in the preview render with the right light, materials and so on. But you want to have a bit more control over the picture in the compositing.
Lets start with opening the render settings. Now check of the multi-pass box at the left side of the "Multi-Pass" name. This enables multipass, obviously. Now, click the "Multi-Pass..." button. This will enable you to choose whitch passes you want to include in your render. What I like to do, is to render a complete image using only multi-passes, and not use the background image at all. The background image is great for looking at your render, but it includes all the passes (whenever you have enabeled multi-pass or not) inside itself. So if you use the regular image beneath the passes, you’ll just apply information that is already included in the image, it wont look good at all and you won’t be able to turn of for instance the shadows or the reflection, since they’re already there.
Passes
To render at complete image with passes only, you may choose «Add image layers» in the «Multi-Pass... menu» (still in the render settings). You’ll get a quite long list of different passes on the left side: Ambient, Diffuse, Specular, Shadow, Reflection, Refraction, Ambient Occlusion, Global Illumination, Caustics, Atmosphere, Atmosphere (multiply) and Post Effects.
I don’t think you really need all of these to get a good image to work with, so I’ll try to point out whitch you’ll need:
Reflection - Yeah, the reflections.
Refraction - Also called deflections. The reflections and light passing through glass, windows etc.
Shadow - Render all the shadows in the scene. There are different ways to control the shadows (and lights). Usually don’t use this pass. I like to control the lights and shadows like this:
Go to the «render settings» and click the «Multi-Pass»-name. You’ll get som few choises; Under «Separate Lights» choose «All». This will enable you to adjust the lights in the compositing. Under «Mode» choose whatever you wish. I like to choose 3 (diffuse+specular+shadow) channels to get the most out of it, but 2 (diffuse+specular,shadow) works as well. Check the «shadow correction».
But since there is no lights in our scene to add, check the shadow pass to render the shadows as well.
Diffuse - The geometry without any light, shadows, reflections or anything. Just plain ugly geometry. If you enabeled the «seperate lights,» the diffuse pass will be rendered there, and there’s no need for this pass, but I tend to keep it even so. Just play with the opacity and you render could look a lot better. So keep this in mind; the diffuse pass may be handy, even if you got two of them in a composition.
Global Illumination - Probably one of the most important passes to make the scene look realistic. GI is how the light behave. How it bounces from one object to another and lights up more places than just where the light first hit the target (an easy example to understand GI is to rig a table, place a spotlight over it and render. Without GI, the floor under the table is dark, but with GI enabled, some light rays get under there and lighten ut just a bit)
To make the Global Illumination work, you need to enable it to render in the Background image. Press the «Effects...»-button and choose «Global Illumination». You can play around with the settings, but I’ll keep it on default for now.
I have to say, this may be the heaviest pass to render. Sometimes GI is the only way to make the picture work, other times things work just fine without. If you manage to make things work without GI, don’t use it ;)
Ambient Occlusion - If you’re working with a still image that won’t take weeks to render, check this pass. But if you’re working with a longer sequence with animation, this pass could almost double the rendertime. But don’t worry, I got a solution for you in a second.
Render a test image. In the picture viewer (if you use Cinema 11.5), you can play around with the different passes, look at them or put them together to see how they work. You can even adjust the opacity for every pass and find out how your composition could look like in the end.
Make Ambient Occlusion-pass render fastThe problem with Cinema 4D, is that it always have to render out the RGB Background image. So if you only need for instance just a pass, say Ambient Occlusion, C4D will render out the whole scene with materials, textures and so on. This will cause things to get slower than needed. I tried to render out a scene with AO as one of the layers in a multi-pass, and the whole picture took 19 minutes to render. Then I removed the AO-pass, and the picture only took 2 minutes to render. As you may understand, if you got thousands of images, the render will be many times longer with Ambien Occlusion added. But I got a work-around:
Take the c4d-file and duplicate it. Rename it «whatevername_AO.c4d». Open the file and delete every single texture both in the materials and in the objects-window (remove the texture tags). You need a clean scene. Then make a new material. Double click on it and only enable the «luminance» channel. The texture will become totally white making the render a white plane. Perfect. Now, go to the render settings and uncheck everything but the Ambient Occlusion effect applied on the background (not as a mult-pass). Be sure to check the «apply to scene» in the AO editor.
If you now render out the sequence, you’ll see that it won’t take as long as if it was a pass.
Afterwards, you just apply the AO-sequence over the rest of the composition (in AE or whatever). Use «multiply» as the blending mode to make the white colors transparent over the rest of the composition. The AO-render will works as a multi-pass even if it really isn’t.
Kind of wierd, but that’s how it works. In my test, I managed to render out only the AO in 5 minutes. So compared to 19 minutes to render with AO as a pass, to only 7 minutes (AO + the render without the AO(no GI-pass)) if you use this workaround. It will only take you five minutes to do, and may make the render time half as long. A really good tip if you’re working with longer sequences.
Delete the atmosphere, atmosphere (multiply) , the ambient, caustics, ambient and specular.

Pass-research
When you're working with other projects, try rendering all the passes, and browse through in the Picture Viewer to see whom you’ll need in the final rener. Just klick on the «layer»-tab, and choose multi-pass to turn on and off a pass. You should also check out the Single-pass-tab to use as research as well.
The saving and final compositing
Go inside «render settings» again and click on «Output». Find out what resolution you need, and choose how your frame range. I want my render to go from frame 0 to frame 500 in this scene.
Now check the «save» box in the left menu, and click on the name to open the «save menu.»
You don’t actually need to save the «background,» so no need to check the save box in the «Regular Image» menu, since we're doing everything with passes (yeah, thats how good we are!).
To save the multi-pass layers, you need to click «save» in the render settings, and check the save box in the «Multi-Pass Image» menu. Choose «Photoshop PSD» as file format and choose a path to save the passes. 8-bit depth will work fine, but 16 and 32 is also avalible if you’re into those things ;)
All the passes will be saved in one PSD-file as layers, and as one file per frame in the sequence. In other words; longer sequences, more files, smaller disc space.
And now for the life saving option: «Compositing Project File.» When working with multi-passes, you got many files and even more layers, all with different blend modes and opacity. Compositing this together could be a nightmare, but luckily Cinema 4D got a great resolution.

Under the «Muti-Pass Image» menu, you got a menu called «Compositing Project File». This menu may be opened, if not click on the small arrow at the left side of the name. Check save and choose your compositing software. I’ll use After Effects in this tutorial. Last, find a place to save this file. Cinema 4D makes a finished After Effects-file, ready to go! Every pass is pre sorted in layers with the right blending mode.
After Cinema 4D has finished the render, just open the .aec-file in After Effects. Here you can adjust every pass exactly as you wish. Turn down the shadow opacity, adjusting the colors, reflections and so on. It's your call! Now you've finished a sequence made of multi-passes only, the advanced way.
Ps: Remember to import and add the Ambient Occlusion sequence in After Effects if needed!
Post notes
I hope you learned something from this tutorial and hope you play around and try out whats best for your scene. Good luck!
So what do you think? Was this useful information?
-Torgeir Bull
If you want to, you can also download the tutorial as a PDF.
03 juni 2010
How to make your photo "monitor -like"
(jeg tar denne på engelsk, siden kanskje fler kan ha interesse for det)
Ok, I thought I'd try to explain how easy it is to make your pictures seem to appear on a monitor/television.
First you need this pattern (click to open, right-click and Save As)
Then you'll need a picture to "monitorize". I'll use this one:
And of course you'll need a software to use. I use Photoshop CS5 (any photoshop will do).
Ok; Open photoshop. Then start with opening the pattern-pic. Once opened, go to "Edit ->Define Pattern".
Name your pattern, and click ok.
Now you've told Photoshop how this pattern works. Now, let's apply it to the picture!
Open the picture. Make a new layer (button in layerwindow, or "Layer ->New ->Layer")
Select the new layer, and select the marquee tool (shortcut; M). Press Cmd+a to select all, and then
go to "Edit->Fill" Choose "pattern" and select the monitor-patten that vi made earlier.
Now you have applied the pattern on top of the picture (if the empty layer is the top one, as it should be). In the layer-window, choose "Multiply" (multiply tells photoshop that everything that close to white, is transparent. Blacks is not) in the layer styles (you may choose something else, but I prefer multiply on this one).
Now it starts to look like something, but we're not quite there yet...
Go to "Layer ->New Adjustmentlayer ->Curves". Make this curve (or play around yourself), and click ok.
At this point, the curve-adjustment inflicts all the layers. We want it only to inflict the pattern-layer. So hold the alt-key and click (option-click on Mac) between the curves-layer and the pattern-layer, in the layer-window. This will make the curves only work on the pattern.
And thats it! Quite easy really. You just make a new layer with the pattern applied, and blend it with the picture. Now you've monitorized the picture. Hope you liked it!
If you want to put your photo inside a monitor, just apply the pattern on after you scale it down. If you scale down the finished monitorized photo, you won't see the "pixels". So just apply the pattern after the scaling and transforming is finished. You can also scale up the pattern if you want it bigger.
Ok, I thought I'd try to explain how easy it is to make your pictures seem to appear on a monitor/television.
First you need this pattern (click to open, right-click and Save As)
Then you'll need a picture to "monitorize". I'll use this one:
And of course you'll need a software to use. I use Photoshop CS5 (any photoshop will do).
Ok; Open photoshop. Then start with opening the pattern-pic. Once opened, go to "Edit ->Define Pattern".
Name your pattern, and click ok.
Now you've told Photoshop how this pattern works. Now, let's apply it to the picture!
Open the picture. Make a new layer (button in layerwindow, or "Layer ->New ->Layer")
Select the new layer, and select the marquee tool (shortcut; M). Press Cmd+a to select all, and then
go to "Edit->Fill" Choose "pattern" and select the monitor-patten that vi made earlier.
Now you have applied the pattern on top of the picture (if the empty layer is the top one, as it should be). In the layer-window, choose "Multiply" (multiply tells photoshop that everything that close to white, is transparent. Blacks is not) in the layer styles (you may choose something else, but I prefer multiply on this one).
Now it starts to look like something, but we're not quite there yet...
Go to "Layer ->New Adjustmentlayer ->Curves". Make this curve (or play around yourself), and click ok.
At this point, the curve-adjustment inflicts all the layers. We want it only to inflict the pattern-layer. So hold the alt-key and click (option-click on Mac) between the curves-layer and the pattern-layer, in the layer-window. This will make the curves only work on the pattern.
And thats it! Quite easy really. You just make a new layer with the pattern applied, and blend it with the picture. Now you've monitorized the picture. Hope you liked it!
If you want to put your photo inside a monitor, just apply the pattern on after you scale it down. If you scale down the finished monitorized photo, you won't see the "pixels". So just apply the pattern after the scaling and transforming is finished. You can also scale up the pattern if you want it bigger.
22 november 2009
Proxy-hemmeligheten!
Har du hatt lyst til å høre på pandora radio igjen? Se film på hulu.com eller gå inn på sider man må være i utlandet for å se? Da skal jeg røpe en helt genial hemmelighet; Proxy!
Proxy er rett og slett en mellomstasjon. Altså en server du kobler deg opp mot for å "skifte identitet". Det er 100% lovlig, og ikke skadelig på noen måte. Greia er at hver datamaskin har en såkalt "IP-adresse". Dette er en tallrekke som beskriver hvor du befinner deg i verden (f.eks. 84.202.41.164 som er min)(sjekk hva din IP er). Altså en adresse på maskina di. Det er denne adressen som gjør at folk kan spore deg på internet om du har lastet ned en film eller lignende. Om man derimot kobler seg opp mot en proxy-server, endres IP-adressen til serverens, og man kan "komme fra hvorsomhelst!"
Så for å koble meg opp mot Pandora, trenger jeg å komme fra USA. Enkelt! Jeg finner bare en proxy-server som befinner seg i USA og har amerikansk IP. Et lite enkelt google-søk viser meg tusenvis av sider jeg kan finne gratis proxy-servere. Vil jeg se noe fra BBCs nett-tv, kan jeg koble meg opp mot en britisk proxy.
En proxy-adresse er en IP-adresse og en port. Ofte skrevet slik (eksempel): 216.157.31.58:80.
Porten er tallet etter : (kolon), i dette tilfellet 80.
Ok, jeg har funnet en proxy; hvordan setter jeg den opp?
Personlig bruker jeg Firefox. Bruker du noe annet, kan du sikkert finne det her.
Klikk HER for windows-versjonen.
For å sette på proxy i Firefox (mac) klikker du på;
Firefox(øverst i høyre hjørnet av skjermen) ->instillinger->Avansert->Nettverk (fanen)-> Instillinger. Deretter velger du "Manuelle mellomtjener-instillinger (proxy-settings).
Her skriver du inn proxy-adressa og porten.
Voila! Du har nå en utenlandsk IP-adresse! Da er det bare å se film, høre musikk og gjøre alt man ikke kan gjøre fra Norge!
Proxy er rett og slett en mellomstasjon. Altså en server du kobler deg opp mot for å "skifte identitet". Det er 100% lovlig, og ikke skadelig på noen måte. Greia er at hver datamaskin har en såkalt "IP-adresse". Dette er en tallrekke som beskriver hvor du befinner deg i verden (f.eks. 84.202.41.164 som er min)(sjekk hva din IP er). Altså en adresse på maskina di. Det er denne adressen som gjør at folk kan spore deg på internet om du har lastet ned en film eller lignende. Om man derimot kobler seg opp mot en proxy-server, endres IP-adressen til serverens, og man kan "komme fra hvorsomhelst!"
Så for å koble meg opp mot Pandora, trenger jeg å komme fra USA. Enkelt! Jeg finner bare en proxy-server som befinner seg i USA og har amerikansk IP. Et lite enkelt google-søk viser meg tusenvis av sider jeg kan finne gratis proxy-servere. Vil jeg se noe fra BBCs nett-tv, kan jeg koble meg opp mot en britisk proxy.
En proxy-adresse er en IP-adresse og en port. Ofte skrevet slik (eksempel): 216.157.31.58:80.
Porten er tallet etter : (kolon), i dette tilfellet 80.
Ok, jeg har funnet en proxy; hvordan setter jeg den opp?
Personlig bruker jeg Firefox. Bruker du noe annet, kan du sikkert finne det her.
Klikk HER for windows-versjonen.
For å sette på proxy i Firefox (mac) klikker du på;
Firefox(øverst i høyre hjørnet av skjermen) ->instillinger->Avansert->Nettverk (fanen)-> Instillinger. Deretter velger du "Manuelle mellomtjener-instillinger (proxy-settings).
Her skriver du inn proxy-adressa og porten.
Voila! Du har nå en utenlandsk IP-adresse! Da er det bare å se film, høre musikk og gjøre alt man ikke kan gjøre fra Norge!
Abonner på:
Innlegg (Atom)



















