Import of high-res PSD, PNG all blurry

This is my first time importing .PSD or .PNG files with color into Animate. I’ve done this many times over in TBS 4.5 without problems with the colors/definitions on the images. Animate, for some reason, blurs everything I import. There is no clarity on anything, but if I do the import into TBS4.5 the object is as sharp as the original. Am I missing some settings or something in Animate that are blurring my imports? TBS4.5 doesn’t behave like this.

what is your Operating System (if a Windows OS, 32 or 64 bit?)and which graphic card? -Romi

Did you look at those bitmaps in the Render view because it’s normal that the OpenGL view is blurry since it creates automatically a small version of your bitmap so that you can play the scene faster.

Select the resolution you want for the projet, NTSC, Pal, HDTV and then look at a rendered image. You should get the same quality as the original if your output resolution is the same. For example if the psd was 2000x1000 and you output is HDTV 1920x1080 you will have the same image quality. If you zoom into you bitmap that’s another thing because then you are using a section of the image of course.

I think that TBS doesn’t create a proxy version of the bitmap in the display so that is the difference.



First, thanks for answering, guys, I appreciate it. I have Dell 3ghz, 1.5GB ram, WXP Pro with an Ati Radeon X1300. Red Alert 3 plays wonderfully on it, btw. ;D

Initially, I figured that would be an issue, and I didn’t want to work with the original files. I resized the file to be much smaller (even saved as PNG), and in photoshop it still looks crisp as a 2 x 4 inch size character, yet if I make the NTSC settings change you suggested (ideally trying to fit this in a 480x360) then go smaller and smaller in Animate, the result is still the same blurry view. My character’s not as crisp. I tried the different views and the render looks the same, except the background is now black rather than Gray (which is the OpenGL view). It’s like I lost ‘definition’. I tried to export the file as a PNG from each program, and even a movie, and the picture is always blurry when it comes out of Animate (also a big bigger), each time. Check these out:

Here is the ORIGINAL file RESIZED to be smaller to fit in smaller resolution–> http://img.photobucket.com/albums/v324/nylinux/fromtbs4.png

Here is ANIMATE’s view (exactly as outputted is how it looks inside Animate)–> http://img.photobucket.com/albums/v324/nylinux/fromanimate.png

Here is the file outputted from TBS4.5 → http://img.photobucket.com/albums/v324/nylinux/fromtbs4.png

Here’s the file outputted from Swishmax 2 → http://img.photobucket.com/albums/v324/nylinux/fromSwish-000001.png

The view/output is great from all programs BUT Animate- or am I smoking crack or something? Does that Animate version look like the rest?

We are investigating this now and effectively there is a slight softening of the image if you have a bitmap 480x360 and import it into a scene with exactly the same resolution of 480x360.

Now in practice you most often will not have the same resolution for the scene as the bitmap so some resampling will happen. For example if you move/rotate even a 1/2 pixel your camera or bg or zoom in slightly the render has to recalculate every pixel of the bitmap and make them fit in the resolution matrix.

In your case if it was for a character cutout animation in bitmaps there will always be the necessity to modify the bitmap pixels because arms are moved, the character is scale up or down depending on the scene, etc. Your output resolution will most certainly be different than your parts resolution.

If you did your character in vector then of course you won’t have this issue. It’s only if you want to use bitmap images for your character’s part.

What resolution will be your projet?

Other point also, when work in HDTV or film, is that if you have a bg that is very sharp and you start doing pans and zooms that are too fast you will start having strobbing issues (not just in Animate) so often people will add some softness to balance this effect.

We’ve investigated a bit more and found that it’s only the view in Animate, when you render out to a sequence of image and compare your original with the rendered image they are identical.

So this is the test we did:
Import the png 480x360 into a scene set to resolution 480x360. Output to TGA file format. No camera move or any change in scale of the image in Animate. In the Render view we see that it’s a bit blurry but when we compare the original and the output render in Photoshop the quality is the same.



These were all PSDs that came from Photoshop CS4, originally, then I imported them into Animate to check the output. I also converted them to PNG and tried the various products I own, results posted in the earlier email note. I did read the forum’s complaints about Bitmaps, so I stay away from those. I was aiming for a smaller sized file so that was the smallest screen resolution I could stand 480 x 360. I could always try a bigger size but I was aiming for Web animations, initially. and keeping the overall size smaller. I will play with the bigger sizes and get back to you. I’ll also try the example you suggested in the email below this. Thank you very much for replying.

I know the Animate ‘viewing’ blurs a bit and you definitely need that soft look when things are moving, but this is just straight import/conversion that I was testing, on static images. I normally work in ‘Compositing’ Workspace view and the Perspective cam shows the real image as it is and it’s just like the original (http://img.photobucket.com/albums/v324/nylinux/viewincompositingworkspace.png) as opposed to the rendered view, so I know TBA didn’t strip it down on import. You can see the image views side by side there. The export, however, no matter what I do, looks more like the blurry rendered view and less like the original or Perspective view. No matter what I do, too much softening/blurring happens. Here is the test you asked me to try (I did TGA output but Photobucket won’t display .tga files, so I outputted .png because there is no difference in view once leaves Animate)–>



TEST1

Original 480x360 image->
http://img.photobucket.com/albums/v324/nylinux/Test1_Original480x360.png

Animate 480x360 outputted image–>
http://img.photobucket.com/albums/v324/nylinux/Test1Animate_Outputted.png

As you can see, the mouth, glasses, pockets, shoes, all lose definition once it’s outputted from Toon Boom Animate. I removed transparency so there are no optical illusions for myself, there is clearly a difference in this files. I also tried reimporting the TBA outputted file and checking the perspective and it also shows the output is blurry and not like when I view the original in the Perspective view. Colors maintain themselves well, but any definitions with lines are eroded it seems and sharpness is lost.

Here’s another example, on another image that is bigger done in full NTSC not 480x360. My small character example might not be big enough to see differences:



TEST 2:

Original 480x360 image-> http://img.photobucket.com/albums/v324/nylinux/Test2_Originalfile.png

Animate 480x360 outputted image–> http://img.photobucket.com/albums/v324/nylinux/Test2Animate_Outputted.png

Again, blurry output although I notice that TBA is blowing up every image no matter what size it is, which is what happened in Test2. so if I had an image that was 400 x 300 (not exactly 480 X 360), TBA will blow that up to fit in the window and I bet that has something to do with the views looking funny. I don’t normally include the borders of pictures in what I’m importing, I’m normally concerned with the actual image itself, so I’ll trim down a picture to only be the just what the image is and delete unnecessary space. I didn’t do that, however, for Test 1. It’s the pic, background and all.

What am I not configuring correctly in TBA, or not doing properly? I never get this if I create a file inside of TBA and export in these formats, mind you.

I imported your image you had sent and outputed it and here are the results in Photoshop. As you can see they look identical.


http://img72.imageshack.us/img72/937/screenshot001ri2.jpg
http://img72.imageshack.us/img72/screenshot001ri2.jpg/1/w924.png


I tried with your other picture and it works also. I had to put my scene resolution to 527x436 of course so that no interpolation happens. You can’t go from 527x436 to 720x540 because then you will introduce some interpolation of pixels.
http://img120.imageshack.us/img120/3296/screenshot002mf1.jpg
http://img120.imageshack.us/img120/screenshot002mf1.jpg/1/w1066.png

Maybe there is another thing you are doing.

I agree that in the Camera view they look different but in the output it’s the same.




Now regarding this point. If you want to work without the fitting you have to change the import setting. Import drawings> Vector unchecked> Alignement rule >Projet resolution instead of Fit. This will “scale” your image in the view in the proper pixel per pixel relationship. For example, my scene is 400x300, my image is 200x150, if I leave it to Fit the image will be “scaled” to fit 400x300 in my camera view. If I choose Projet Resolution you will see the image takes and area 1/2 the camera view.



Ah, you solved it again. I did this part here and now it looks good. See, I told you it was a dumb newbie error. :slight_smile: In my messing around I did play with that setting but noticed that I can’t do cutout if I don’t vectorize. Am I mistaken? Do I have to import each piece of the puppet independently and use it if I don’t vectorize the whole image on import?

I would say that cutout style can be done in either technique, vector or pure bitmap, it depends on the style of the animation.

What’s nice with vectorized bitmap (color as texture) is that you can set pivot point different on each drawing of a layer, for example let’s say that you were bringing in 10 photos of hands to put in your hand layer. Vectorizing them would allow you to tweak the pivot point for the rotation of the different hands. Or it could be a shoulder layer that would contain the front, 3/4, side, back shoulder images, each one of them you would assign pivot at different place.

Vectorized bitmap allow you do do basic editing also, like triming, cutting parts away. This is because the bitmap is encapsulated in a vector “box” that you can cut. For example you could, I don’t say it’s the best way, vectorize your whole character and in the Drawing view use the cutter to remove the head and put it in a different layer or drawing. If you did it this way make sure you have a much higher resolution for your initial image because you might be doing truck ins of the camera or scaling the character later on.

So you’re saying to work with high res bitmaps rather than PSD or PNG? I’ve basically been doing what you mentioned about importing the whole character and cutting out pieces/placing them in new layers, but with PNG files. I’ll try this with Bitmap now because I would like eventually do like to be able to animate my character like that TBA Ballerina sample puppet does. That thing has a million pieces for each position to put it in. Thanks Steve!

PSD and PNG are bitmap images. I’m just saying to make sure you have a big enough resolution for the PNGs that you are importing.

Personally if the characters you want to create are like the one in the example I would draw them in vector. I don’t see any complicated shading or texture in it so it would give you more options to adjust, cut, modify in Animate.

If you were doing something like this with photos:

http://www.teletoon.com/teletoon3/detour.php?language=Fr&func=php|templates/video.php|1-punch-639-rtmpe-u-manlist-fr&xVar=

I would understand to use bitmaps.

To be honest with you, with the exception of that character and a handful of others, I’ve been working strictly in Animate and not been importing unless it’s my Swishmax flash work. To your point, there are many more options in Animate, that I previously used Swishmax to do, with regards creating character/ still images. I’m satisfied with this product. Now if I could stop being such a MAN, and actually read a frakin’ manual every now and then, I might be able to get somewhere. lol Can’t help evolution.

That was a hot link! I had to bookmark that because that gave me some great ideas! We should have a LINKS page with links just like this to view other peoples’ work. I know we have Showcase but I’ve seen those already. Do you have more links?

I don’t really have a bunch. I just knew this production that I helped to train with Harmony. You have these ones also.


http://www.rubygloomtv.com/media.asp

http://fablefolk.blogspot.com/2008/06/wobbly-land.html

http://www.rubygloomtv.com/media.asp

Thanks very much for this. I have a billion ideas of things to do now. All that is possible with Animate?