Jump to content
Hash, Inc. - Animation:Master

heyvern

Hash Fellow
  • Posts

    5,205
  • Joined

  • Last visited

Everything posted by heyvern

  1. Yes but... in my opinion there are dedicated standalone applications that already have those file transfer functions and chat functions etc so why put it in AM? Just run them along side AM. Those other programs are BETTER at it than AM ever could be. I don't want the limited resources of Hash to be wasted on trivial "novelties" like "chatting" in the program. Even MS word doesn't have a "chat" feature does it? If Hash removed the chat feature I won't lose sleep over it (unless the headphone cord brushed my leg). Does anyone need Skype to have a 3D modeling/animation plugin? I mean... you may want to animate something while you are typing or talking and it would be so much easier than all that work to open up AM. Of course if AM has chat you can chat and do 3D in Skype AND AM at the same time... while modeling in 3D in skype you can chat in AM and never miss a second of productive time. Put an email application and a beer dispenser with a snack machine in AM and I would never leave the house. Put a game engine in there so you can play games while working in AM. A link to Hulu and watch TV as well. Make it an entertainment console... that also does 3D... Okay I went a bit far with this... forgive me. Obviously... I don't own an all in one printer/scanner/fax machine either. p.s. Hash never did put in my feature request for a coffee machine USB hook up that activates with long render times. -vern
  2. Yes, it was fun for a while... then the novelty wore off. I am prone to being frightened by simple things when I am deeply engrossed in work like this. It could be a tiny breeze causing a small scrap of paper to flutter on the desk and I will jump like a gun shot went off. One time I freaked out when the change in my pocket shifted slightly against my leg while I was in my "creative trance". My headphone wires are the worst. They will shift and fall funny and brush up against my neck, shoulder or arms and scare the bejesus out of me (I really need to get a wireless headphone). I had to turn the AM community thingy off. Too much stress. -vern
  3. Nice job so far... ... yikes... I would however, not buy a robot to work in my house with a but crack like that. This is just a personal preference. Others may like a good but crack on their robots... who am I to judge. -vern
  4. My trick with Thom being assembled by particles was done by reversing the rendered sequence. You can't "reverse" particles so the best way has been covered. Get both animations the way you like them. Render the second "equence". Import that sequence and reverse the time. You could just render both sequences and composite with another program or you could composite both in AM or just use the one sequence as a rotoscope and "hide" that model and particle sequence in the chor. You would only have to render TWICE instead of 3 times. So you render one sequence to go backwards with the other model hidden or not active. Import that sequence as a rotoscope and reversed the time. Make THAT model not active and turn on the OTHER one. You could also throw in camera movement this way as well but would have to reverse the keys of the camera movement for the reversed rendered sequence. See the rendered sequence used as a rotoscope will match the camera movement but it's backwards in time... ... oh my... I'm getting dizzy... is anyone else's head spinning or is it just me? -vern
  5. In my case there will be shots of stationary characters talking at a table facing each other, the camera locked. Behind one character will be a large set of windows overlooking a large production assembly line. I figure the "window" and frame of the window can be a non moving single image render. The background elements seen through the window can be a short looping animated cycle. The foreground elements and characters would be another layer in the composite. The character with no moving background elements is even easier. If I cut back and forth and keep the wide shots to a minimum this should save a lot of rendering time. -vern
  6. I find it very distracting. Also if I forget to "turn it off" it scares the hell out of me when someone chimes in. Literally I jump out of my skin. Rush of adrenaline. I had to deactivate it. When I have AM open I plan to work not chat. I don't see the purpose of "chat" software inside a program for getting something done. If I need to chat there are plenty of dedicated programs for that and my hair won't turn white and fall out when I'm contacted. -vern
  7. I plan to try something like this with my Terminator spoof to speed up render times. Since I plan to use an environment map for reflections it should work for the shiny stuff. I figure the shiny robots will require the most rendering time so if I can render them separately from the background elements it will speed things up a bit. The "background" elements will be animated but not as much. Don't many of the "big" studios (Pixar for example) use this technique? Rendering different elements in separate passes to composite? Is this done for rendering speed or other purposes? -vern
  8. I think the question of "what is cheaper" to do, real vs. computer isn't about the cost. In some cases it might be the same cost to film real objects as opposed to using 3D. The benefits of 3D come in for absolute and total 100% control over every aspect. PERFECT product shots, PERFECT lighting, PERFECT camera moves. Nothing left to chance. Also in many situations the product may not even exist yet. A product might only exist as "blue prints" or "cad" designs. Being able to get a jump on the advertising in advance of production is always good. They do that with cars all the time. Print ads of cars for the next years model are often not real, just 3D. I remember having to do some stuff with a product that technically existed but... the production had changed slightly. The plastic was a different color. The shape had changed slightly. They didn't have any prototypes for photographing at all... so... had to use 3D. Real photography was not even an option at all. -vern
  9. Uh... er... gravity? In what situation? Hair? Particles? Newton dynamics? Dynamic constraint? I believe hair and particles use the "force" setting in the chor which is on by default. Newton dynamics have the gravity set in the properties for each object. There is no "gravity" that you can "turn on or off" for plain objects. -vern
  10. I thought Yves shader was "broken" in 15e? I was trying to use it a while back and it didn't project correctly. Has it been updated somewhere? Or is this a different shader? The spherical mapping didn't seem to work right. Fantastic amazing mind blowing renders by the way. Good grief you would almost think AM is some sort of professional 3D application... uh... wait... it is.... never mind. p.s. Where can I get those genetically modified "coreless" apples? Those will sell like hotcakes. -vern
  11. Yes the fog image was removed per the client request. They wanted the logo on an entirely black background. I just made the camera color black so the fog is black. -vern
  12. Thanks John, That's normally what I do. In this case I didn't create the original extruded letter models. I noticed some creasing and bad spots and went through the whole thing very carefully... I missed that one tiny spot mostly due to not doing tests at high resolution (stupid vern). ---- On the subject of the QT Pro export. I can't understand why it's getting "washed out". Is that just the nature of the extreme compression of the codec (H.264)? The original targa files look great (same as the still frame in the first post). The problem is I'm using QT Pro on the Mac. The original files look the same in the mac before and after export but on windows the exported H.264 output is washed out compared to the originals. -vern
  13. Well it's done! I added some motion blur just at the end for the kettlebell falling. It's subtle, about 10 frames only 5 passes at 20% but enough to soften the motion. For some reason QT export "washes" it out. The "original" looks fine. I need to fix that. pkb_final_motion_blur_hd.mov I have to say not bad for a few days work. A few days work to complete a project in AM for me is like a nano second. -vern
  14. Thanks for the feedback! Animus, holds on the letters... I tried a few other things first but decided I really wanted that opening to zoooom in FAST. It's actually a teeny tiny bit slower than I wanted but close enough. I used a lot of bias tweaking on the key frame curves and if I tightened them up it makes them go wonky and I have to tweak them again. It looks good. I wanted those letters to really zap in and SLAM into place quickly. I think they will be using this for the intro to training and demonstration videos. (I realize now I could have created just one action file for all the letters and offset them. It would have saved a lot of time but I thought of it too late.) Nancy and John, The fog image... I've never used fog before really. So then I was sitting there with this logo trying to figure out how to have the letters "appear" out of "nowhere"... hmm... how the heck... mist? Dust? Particles? Gradient combiner with animated transparency?.... ... hold on a second... fog? Will fog work? YES FOG! Then the problem was the request for a black background. A big black kettlebell on a black background... .... it nearly drove me nuts so then I figured out that image trick in the fog. I never used fog so it was all pretty new to me. Of course after playing around with the image in the fog... the client wanted to take out the color. At that point I was arguing that the black kettlebell would "vanish" against an all black background... but just to be cooperative I did a quick test render (the one I posted above). It actually looks BETTER than the color fog image. It works. It took about two hours to render the final... and DAGNABBT! Have to render again. Look closely at the image I posted. Look at the left side of the letter "C". There's a black "smudge". An obvious "dent" in the model. The AM AI import is always a bit... hit or miss when it comes to this stuff. I thought I had fixed all the wonky connected splines but I missed a spot. I also rendered with an alpha channel which I thought was even worse because it removes the background until I realized the background is entirely black anyway... I took out the orange fog image. . Anyway the black smudge on the C might be fixable in photoshop... but do I spend two hours painting out the spot on 180 frames or do I go to bed and let the computer do it? Easy choice. -vern
  15. So this is the first time I've done an animated logo with AM. Turned out pretty good. The company sells Kettlebells which are used for building muscles. For those of you who don't know, muscles are things in your body that are slowly shriveling up as you sit in front of the computer for hours and hours with out moving anything buy your hands and fingers. This isn't the final version. It's a low res sample. The final is HD and doesn't have the orange glow.... man it looks SWEET. I'm watching the final render right now... still about 90 minutes to go. pkb.mov Here's what the final will look like: The zoom was just straight forward key framing with that little "bounce" at the end. I used fog on the camera to have the letters zoom out from. Worked great. The orange glow is actually a fog image. That is COOL. Fog images. Neat. The fog takes on the color of the image... but of course not using that image for the final but still a neat trick. For the last bit I used Newton physics for the letter "bounce" when the kettelbell hits the ground. I needed the letters to bounce but end up in the correct spot afterwards. First I made the letters "dynamic box objects". I did this so a letter like the "P" wouldn't just fall over... actually a lot of them would just fall over because they are slanted. A box object defines the shape as a box so letters like the "P" won't be top heavy and fall over. After simulating the bounce of the letters I just went in and copied the keys for each letter from before the bounce to after the bounce. This created the illusion that they bounced a little and ended up in the same spot. To bounce the letters (you can't really "bounce" things from an impact that easily) I placed two temporary ground planes under each row of letters. I keyed the letter objects "up" a bit before the simulation so they had an upward force. They go up and then down onto temporary ground planes under them to stop the movement. After the simulation I just had tp clean up the key frames a bit, modified the keys so the upward motion was smooth and didn't go to high. I also removed most of the rotation keys. They are like a billion miles long to get that trailing 3D extrusion and some of the excessive rotations looked bad. Overall this was still WAY easier than trying to create the bounce effect by hand. I don't know if this is normal but I did some wacky things with lights to get that black kettlebell to "stand out" from the black background (that was what the orange glow was for originally. It looks good now against black so the orange glow is history). I have lights around the kettlebell with huge intensities... like... 1400%. I don't know why or how I ended up having to do that but there was just no way to get those rim highlights to show up otherwise. I ended up putting in a reflection panel (on the upper right) to give it more shape. Almost all of the lights have intensities well above "normal". The back of that thing glows like the sun but you hardly see it from the front. Anyway... whatever works right? -vern
  16. So Nancy, Goddess of "Front Projection" (that makes the mind wander ), Front projection doesn't work with alpha channels on images? I was trying to put an animated "shimmer" across some surfaces using a blurred white streak with an alpha channel. Unfortunately the models with front projection render "black" where the alpha is instead of what I expected... the white line to render over the unmodified surface of the shape. Maybe I'm expecting too much. I did turn on "Include in Alpha Buffer" so that isn't it. -vern
  17. I started a new thread for the model/project and download here: http://www.hash.com/forums/index.php?s=&am...st&p=300404 -vern
  18. There is a recent topic about this I just can't find it. In a nutshell AM only renders 72dpi. You don't increase the dpi in AM to make it "bigger" for print use (AM doesn't have "inches" for image size, only pixels) you increase the pixel width and height of the render. How can changing 640 x 480 to 300 dpi make it "higher resolution"? It doesn't. 640 px is going to be 640 pixels regardless of the dpi. 72 or 300 it's still 640 pixels. To make it higher resolution for print you change how many pixels... 640 to 1280 doubles the resolution. It's still 72 dpi. -vern
  19. Are you doing the render like... the real time render? I've never tried the real time render. Is that like AM plays back and is recorded to a movie file? Have you tried a regular type render? Where each frame is "rendered"? If you are using the real time playback... a chor as complex as that one appears to be even on a huge fast machine might experience occasional skipping of frame. Maybe you should turn off "skip frames". Just guessing. -vern
  20. Wracking my brain on how to make the contrail model work with any model in a chor. I added a "target" bone. That target bone is the target for the translate and orient constraints for the other trail bones. Then when you drag the contrail model into a chor you only constrain the target bone to your model. The other bones follow along. So you can then drag any number of contrail models and put them anywhere you want. A few other details to work out and then I will post the project. -vern
  21. When I use the arrow keys to step through this in QT frame by frame... it is as if some frames are in multiple times. It's weird. I hit the forward arrow several times but the frames only change once. Could there be some issue with frame rates when rendering? For instance in the beginning when the tank first comes into view there are several frames of the tank not moving... then it jumps to it further than it should. Almost as if some frames are "missing" and others repeated. The "stuttering" isn't a playback issue. It appears to be in the mpg itself. -vern
  22. Thanks Meowx! Great job. Just to be clear this "idea" was inspired by my memory of someone elses brilliant idea from a long time ago. I hope to find that original sword swoosh file on a back up or old hard drive and hopefully determine the source of it. I'm glad I remembered as much as I did... uh... actually I'm SURPRISED I remembered as much as I did. The other exciting potential would be having the "pre rigged" or constrained "trail" model with poses. You could drop in as many as you want to create bunches of overlapping streaky trails. Just line them up with whatever part of the object you want a streak coming from. Adjust the poses to change the length and "duration" of the streaks for variety. -vern
  23. That looks great! I really wasn't sure how you would use this for a tail. I kept having images of a poor mouse running and running because if it stops its tail would zoom up inside it like a retractable cord on a vacuum cleaner. -vern
  24. Oh heck... (In this animation you can see the start and end of the trail. In the other project the end is fine but the start wasn't right so I cropped it out of the render.) contrailB.mov contrail4.zip -vern
  25. Ha! Figured out how to have the tail "stream out" at the beginning. Very simple. In the above sample the max lag on the last bone is 10 frames. Go to frame 10 and key ALL the bones lag value at the amount that is already there. You just want that key. Go back to frame 0 and key the lag for all bones to 0. Now as each bone is pulled along the tail stretches out as the ship moves. When the ship stops moving the tail collapses. Works great. If you want the project I could post it but... it only takes about a minute to key those bones... it's a good learning experience. This whole thing could probably be set up with a pose so you only need to key one pose of the contrail model. Maybe have the lag value for each bone linked to an expression to a pose slider and you could stretch or shorten the trailing effect by keying a pose. This is pretty cool. It could be used to create engine exhaust/flames that cut in and out... add some noise to the pose keys controlling the lag and the jet stream would "vibrate" etc. -vern
×
×
  • Create New...