Attached Link: http://www.clydesight.com/galvr/vrmovie.html

Hi All, I love Bryce and doing stuff with it, but I have one issue I've never been quite able to manage. A lot of times when I render an image (and I usually have to use the basic Anti-Aliasing) it isn't as sharp and crisp as I would like. I turn off haze and fog in the Sky Lab (as these effect any scene, indoor or out if turned on). Usually after I render a scene, I take it into Photoshop and sharpen it (using sharpen or unsharp mask, depending on which works better for me.) Often, I can get a pretty clear static image, but not a nice as some images I've seen in the galleries. The crisp and clarity issue gets to be a bigger problem when I want to make a VR scene. VR scenes (using QuickTime) are often compressed via JPEG to crunch the file size. The images that come out are really not that good, and even if I used no compression for a VR (which makes the file HUGE), there still are problems with clarity and crispness. Take a look at this VR I have done and you'll see what I mean: http://www.clydesight.com/galvr/vrmovie.html Maybe I'm expecting too much, but to me, some of this VR looks like it could have been crisper and clearer. What do you think? I work at 800 x 600 px resolution, and apparently Bryce renders at 72 dpi. I am not sure if changing these settings would make a difference. What do you all do to get really crisp and sharp images when rendering from Bryce? And if you make VRs from Bryce (either panos or cubics), got any tips to improve the final image quality? Thanks for any suggestions. I find the Bryce community is always helpful!

I've never seen a difference with changing the dpi, some say it should be at 96dpi, or higher. Can't hurt, I guess, but like I said, I've never seen a difference, ever, lol. I almost always use the default AA. Rarely do I use the higher AA. I plop render soft shadows whenever I can. I looked at your VR, not really seeing what you mean by sharpness/clarity? Although, I agree almost every pic I do that goes into Photoshop will get a 25% (or so) increase in an unsharp mask. I think you have the problem we ALL have, and that is we look at our own work WAY too close, and "see" far too many imperfections, lol. AS

One instance for me was when I tried doing some painting on a 1024x768 size image and could not get the results I did when I started with an image at 6000x5000. Once I scaled the larger image down to 1024 it did look sharper. I discovered this while browsing galleries of artists that use large scale images. Their resolutions seemed so much better. 11000x9000, which freaked me out because even my 1gig of ram has a tough time in photoshop trying to manuever in a file that size. Try a larger render, do all your post work and resize for your final posting. Also bare in mind that some lower end image editors use different compression for for JPEGs. Photoshop handles this well. Hope this helps. ICM

roflmao@icm. drac

AA is certainly part of the problem since it's really just a 2D blur of contrasting pixels. I never use AA except for render tests and scenes I want to be "soft". Best; Dan

DPI: This seems to confuse a lot of people. Please remember that DPI = "Dots Per Inch". This is used for output devices such as printers, and defines how many dots are used per inch; higher numbers = more resolution = cleaner picture. On a monitor, however, the Dots Per Inch is equal to Pixels Per Inch - and that's FIXED for the combination of monitor and screen resolution you are using (usually somewhere between 75-100 pixels per inch). Increasing DPI past this won't help at all :-)) Some useful links: http://en.wikipedia.org/wiki/Dots_per_inch http://www.sphoto.com/homedd/main.html http://desktoppub.about.com/cs/intermediate/a/measure_dpi.htm Hope this helps. Cheers, Diolma

right dio how many times i tried to explain this to my web design colleague with pronting background :( lol a higher dpi will just print the same info on a smaller size.

pronting = printing

If you're working at 800x600, do you mean your output image is that size, or do you mean your monitor is set to that ratio? Don't let the monitor's aspect ratio box you in too much. Go for some wide shots, or tall ones, you can evoke different emotions from each. Sharpness? Do you mean sharp like Rochr's images or sharp like MrDodoBird's? I'm not sure I'm following the sharp concept...

If you have Paintshop Pro, try Effects>Enhance Photo>Clarify. A setting of 1 adds a nice degree of sharpness, or clarity to most images. The setting can be upped in some instances to give an even sharper look.

render bigger. All web graphics are meant to be displayed at 72 pixels per inch. Print graphics are usually 300 dots per inch on paper, which is the same as 300 pixels per inch on screen. This means to get a higher resolution, the image must be a LOT larger on your screen (divide 300 by 72) For sharper VR images, use larger renders. Remember that for any typical VR, the resolution is decreased by a factor of four, because you only see about a fourth of the image at a time. For example, if you start with an 800x600, in the VR you are only seeing a 200 pixel width at once, but the size of the window will be much larger. Therefore, try rendering at 4 times the size for your VR images.

Attached Link: http://www.scantips.com/no72dpi.html

Actually, dpi or ppi or whatever you want to call it is meaningless on a computer screen. A computer screen displays x number of pixels in the width by x number of pixels in the height. Video cards and monitors don't know anything about dpi, just the width by the height. This is still one of the things that causes the most confusion among people doing computer graphics. Unfortunately it's not helped by some people who should know better (magazine writers, software reviewers, etc) saying things like Macs have a screen resolution of 96 dpi while PCs have a resolution of 72 dpi. That is just plain wrong. Change your renders from 72 dpi to 600 dpi. You won't notice any difference. This will only make a difference when you print them out. The 600 dpi will print a lot smaller than the 72 dpi image. Have a look at the attached link for an even more detailed explanation of why the whole dpi thing is just not relevant to a computer screen. On the sharpening side of things, one thing you can do in PSP is to split the image into channels (HSL) and do your sharpening (use unsharp mask)on the Lightness channel. This is better than using contrast and brightness or the plain sharpening filter, because it won't affect your colours. Then recombine the channels and you should have a much nicer looking result. Using this method seems to give much more latitude when sharpening, too. You seem to be able to be a bit more aggressive on the sharpening doing it this way.

TO All, Thanks for your insights and ideas! The DPI thing apparently is very confusing to many of us, and some of your tips are very interesting to try out! AgentSmith had a very interesting comment: "I think you have the problem we ALL have, and that is we look at our own work WAY too close, and "see" far too many imperfections, lol." Do you all run into this problem when considering your work as well? Do you know of any psychological studies done on this issue? One thing that I don't know how to get around is this: I create an image. I work very hard on it. I get to the point where I like it. I show it to someone else. They see it for the first time, and have the impact it creates because it is unknown. But to me, it can never be a "first time" because I created it. So I miss the impact with my own work that I get when I look at something in the galleries, which I have never seen before. Do you run into this problem, and how do you deal with it. Sort of a philosophical question, I admit. But Bryce is a very philsosphical program isn't it? Anyway, Thanks for ALL your comments and tips! I know I can always count on the Bryce community for wonderful lessons and ideas! Best to all.

Clyde, In response to your private email and your question, here's a clarification on the sharpening method I talked about. If you're using Photoshop, then change your image to LAB mode, make sure you have the channels palette open and select the Lightness channel. Do the unsharp mask operation on the Lightness channel. Then change back to RGB mode. For the Paint Shop Pro users, you have to split the image into HSL channels (Channel Splitting -> Split to HSL: this exact command may vary depenedent on your version), which will give you 3 images on screen - hue, saturation and lightness. Do the unsharp mask operation in the Lightness channel, then go Channel Combining ->Combine from HSL. Your new RGB image will be produced from the 3 channels that are on screen. Doing sharpening this way doesn't affect the colour information in the image, so is not prone to producing ugly colour shifts. I have found this to be an excellent way to sharpen my digital photographs and have also found it useful on 3d rendered images from time to time. I hope that clears up any possible confusion Clyde.

I think ICM has one of the most important pointers here, render it BIG, do your postwork and then chop it down to the size you want.

To All, especially Gog CAI and MadMax, The tip of rendering BIGGER and then shrinking for use in a VERY GOOD TIP, especially in a VR movie! I tried it, and saw a marked improvement in image quality (although the final file size of the VR was increased as well, but within tolerable limits.) Here's a tip I discovered using Bryce (after years of using Bryce, wouldn't you know it dawns on me NOW what the feature meant! ;>) When you go to Document Set-Up, you can set your workspace to any size you like. Down near the bottom of the window, you get choices for rendering "resolution", which really is the rendered image size. This has nothing to do with DPI (which as we NOW know, means nothing to monitors anyway). You can pick a very large rendering size, but still have your workspace in the preferred size that you like using. For example, if you choose the "standard" document size, your workspace will be 640 x 480. But you can choose any rendering size, and your workspace will remain 640 x 480. When you render, however, the image will be at the much larger size. So this makes it easy to do the "really big image" thing. What I found for a cubic VR (one that let's you pan 360 degrees) is to do this: You need six SQUARE images: four sides, a top and a bottom. You get these by positioning the angle of the camera. Render each image (at the BIG size) and number them 1,2, 3,4,5 and 6. Set the camera's focal view to 112.5 Take the images into Photoshop or any other image editing program. They will be huge, of course. You can sharpen or tweak as desired and re-save or save as a OICT file type (don't JPEG, that will introduce artifacts). Using a cubic VR program (I use Cubic Converter-- but there are others), import the images into the cube and convert it to a VR movie in the program. Last, you can set the compression and final movie size. Here is where I shrink it down to the size I want. Then I export. The resuting VR is much clearer than if I rendered the Bryce stuff at the size I was working with it on. So, THANKS to you all for the tips!