Arnold Gallardo

Or find a way to 'bake' your Global illumination (Monte-Carlo based) computation into textures. Not all 3d apps can do this very well. As much as Radiosity is viewed nowadays as an 'old' technology, hybrids that accelerate it to exist and most computations anyway deal with direct luminaire and indirect illumination environments with minimal directional diffuse situations unless you really dealing with exotic material that needs this capability. I think that your undertsanding of light quality plus photographic/cinematic eye would count more in making a scene believable as well as acuurate. You have to remember that ALL render engines are geared towards mimicking the camera/film combination when rendering out images. Special effects like bokeh and optical artifacts being incorporated inot render engines are part of this trend. You have to 'see' what you are 'looking at'.

That depends on your render out settings. Most render engines have the capbility to render out alpha channels along with the scene. This is probably the best course of action for you. If not you might need to render another pass with a background color that is not inlcuded int he scene like green or blue or even orange. Then just use a vidoe editing program that has chroma-keying capabilities aka 'green screen'. Something like this is really hard to give tips to without seeing what you have. Also you might wanbt to look into using a camera tracking tool for compositing the foreground and background together aince you already have the camera animation.

NT 4.0-heaven??? it was for me

Well for max to support other color spaces would need to rewrite the render engines. Most render engines claim spectral response computations but stores data as RGB crazy! Anyway that would require a render engine rewrite if the color space is change as well as a different kind of tone mapping routines.

I don't know I never relied on PS's built-in conversion stuff. Ive had weird colors happen when I directly convert from RGB to CMYK although this always happens with the 'metamers'. I wont take that risk. I used to be a pressman so I try to be careful when it comes to these things. Converting from RGB to cieLAB to CMYK has been very safe for me since I also do pre-press work. It is always a good idea to ask for a 'printer's spec' before you start any job that will go out to press. This would save you time as well as money. You get as much information for the 'end use' before you start any job BTW also don't convert from RGB to Grayscale directly. It is better off to convert to CIELAB first then remove the chroma channels (a and b channels). The 'Lightness' channel has all the luminance data you need. GO test it yourself, RGB to Grayscale will have less contrast and less dynamic range than RGB-LAB-remove a,b channels-Grayscale.

Nice tips Ernest on the RIP issue. That's something most people do not address When you do convert from RGB to CMYK never ever convert that directly IF you have CIELAB color space to convert to as an intermediate. In Photoshop this would be the LAB color space. Why? CIELAB has a wider gamut for that than the RGB, the indirect conversion would not have a drastic threshold chroma clipping. Always go from RGB-LAB-CMYK.

A good reference for this is also the Graphic Arts Guild handbook: http://www.amazon.com/Graphic-Artists-Guild-Handbook-Guidelines/dp/0932102115 since there isn't one for the cg-arch community. Ethical guidelines and such are also discussed. Yes, you need to be very specific with your client about havoing access to the 3D files. Most job's specify image output or video output only and have nothing to do with 'source file' access. That's always is negotiated and priced differently, normally computed based on future revenue possibilities. That's worse than signing up an 'All Rights' transfer because 'multiple derivative work' can be made form it!!!!

What you are talking about is called an "elevation" in architectural terms. Most of the time this is a 'half section' of the plans. I don't know though what you mean by acquiring a 'picture of a 3D house'. In most cases this is generated by a draftsman in a CAD program or an architect.

Very nice! I like the blenidng of the sky and the building in some of your photos. I went there a couple of years ago too and here's what I got: http://www.flickr.com/photos/zeroneg1/sets/72157594307359186/ My shots are more abstracted that yours though and yeah I used film back then.

you can try Darktree Simbiont: http://www.darksim.com/html/simbiontmax.html I have used Darktree for almost all my 3D applications.

This is fairly normal. Inside ACAD the data it handles is parametric meaning they exist only relative to each other and mostly composed of POLYLINES, SOLIDS and surfaces that created the 'data' still there just hidden or turned off in the layers. When you import that in Max everything is made 'concrete' and every surfaces is now triangulated to follow the 'CURVES' and lines of ACAD specially if u set the tolerance/curver deviation high during import.

Aside from what Ernest said what you really need to do is this: SEE everything not just LOOK. there is a difference in SEEING and looking at stuff. You need to be aware of the minute nuances that life has this means looking at the forest as well as the tress as well as the little pine needles and how light affects it. What Ernest said is wonderful because it is the ONLY WAY to learn how to SEE and REALLY SEE. You need to see negative and positive spaces, subtle tonal changes and where it comes from, be able to discern small color casts brought about by light source oeirntation changes as well as kelvin temp changes etc. All this will come. For me this came after being a photographer doing darkroom, work although I have been sketching since I was 5 years old. Good Luck and yes BE AWARE of EVERYTHING and ignore nothing

Technically the only people that can get away with something that is recognizable in an image is when it is used EDITORIALLY. If you look closely any commercial photography or videography (FILM SPECIALLY) either has 'generic' background elements (meaning props and yes props were made for this very reason as well) or that they used actual product but are very hard to decypher to make a claim against. Anything recognizable is a PRODUCT PLACEMENT and it was used with permission if not with PAY. Next time you watch a commercial or look at a print advertising scrutinize the background elements and see if they are generic or not. SOme will not be and most of the time these are 'street' shots which means they are PUBLIC DOMAIN however if there is a big billboard with surely recognizable product or corporate identy like the Coca-Cola logo or even the Nike one, those are either blurred or replaced altogether. Now compare this with editorial or newsworthy footage or a still photojournalistic work. They show things as they are 'IN SITU' with no changes. I hope this clarifies this issue for you.

This has nothing to do DIRECTLY with CODEC. Codec determines your BITRATE and quality of the video. How big your files are is determined by the CODEC. You should look into your DVD AUTHORING program more as well as your DVD BUNER software as well as the BURNER software itself. I tend to BURN with NERO and Author on DVD Architect. As for the widescreen this nomally means 16:9 format. The NTSC standard uses 720x480 size but the 'aspect ratio' is changed. This means the PIXELS are either wide or narrow depending on the aspect ratio settings. Most DVD players will recognize if the video is 19:6 or 4:3x and change the playback accordingly. You will want to do this because not all people have 16:9 displays YET so you want then to be also playable on 4:3 displays. You have to understand that the DVD itself is ONLY a STANDARD. That means not all IMPLEMENTATIONS of it will be identical they just have to follow certain standards, how they are directly implemnetd are up to the manufacturer. In most cases you will run into this problem if your client is using a DVD player that is MADE for commercially pressed DVD movies only. Those have problems playing DVD-Rs and more so with DUAL LAYER DVD-+Rs. For your DVD to be universally played you need to have it MASTERED and burned commercially.