[Japong]

DISCLAIMER: None of the images in this post are mine. All credit goes to Highside and his site, Real Trace(translated). I'm hosting these image on my webspace because direct-linking is useful, but impolite if you're using someone else's bandwidth.

Now, I'm sure a lot of you have probably been to Highside's personal gallery,, and seen the amazing adobe illustrator art this guy has done. If not, go and take a look - this is a pretty good example of what to expect:





Now, I've never seen anything like this done in Illustrator before. If anyone
else knows of illustrator work of similar quality, please, drop a link for me!

However, the work he's done on Real Trace surpasses even that, and this is where the point of this post comes in:




Can anyone explain how *that* is possible in illustrator? As far as I can tell from looking at the gallery page, the furry cats are not a reference, but the final image. I can't read highside's explanation, as the accompanying text is written directly on the jpeg files, and thus untranslatable.

Finally, has anyone been able to get ahold of/translate the Real Trace book? The samples of it once again are untranslatable for an aglophone such as myself, so if anyone knows of an english edition that explains these techniques, please reply!

[balistic]

http://en.wikipedia.org/wiki/OCD
_________________
brian.prince|light.comp.paint

[nafa]

I could be way out of my league (and out of line) here. However, I suspect the images are imports of scanned photoes. They are just way too photo-realistic, life-like and detailed to be created vector by vector from scratch.

[Mikko K]

What's the purpose of tracing like this?

[Gort]

[Mikko K]

quote from the book page

[Japong]

The purpose is that this is now vector art - it can be expanded to any size you want - as large as the side of a building - without a loss of quality or pixelization - as well as the more obvious answer: because it's a challenge - why paint out photorealistic paintings, right?

[Mikko K]

I understand the challenge thing. But if you think a little further, does this add any value to the original photo? If there was a large commercial demand for this kind of thing, there would be a lot more stuff like this around (I don't care if they're fakes on not, not my problem).

It must take a while to do such a thing in Illustrator. Now, all the decisions have been made in the photo itself. Photos can be easily color corrected etc. in PS, but once you create the vector image out of it, can it still be changed easily? Plus when you think about outdoor ads, the raster sizes are often so big that a high-resolution bitmap is enough. They're meant to be viewed from distance anyway.

[B0b]

it looks like textures have been placed into illustrator and manipulated using the 3D tool, as i can't see any paths for each of the textures within the image..

so if they did make the image huge, tho' the outlines would be smooth, the texure would loose res..

can't remember the guys name, but he did some hot babes totally within illustrator, and they looked damn nice

[Japong]

From the website:

[B0b]

...

[Capt. Fred]

There's a gradient mesh tool in illustrator.
Draw a shape, and within that shape you can manipulate a grid. Not a linear grid but one made of paths you can manipulate like all paths in illustrator. You create gids of any complexity you like. The points where the grid lines meet can each be given a colour and illustrator interpolates between the coloured point(intersection) and the neighbouring point - 'gradient mesh'.

It appears by having so many lines in the mesh, he effectively has a lot of 'pixels' like a bitmap-style matrix, except these are points of no size, that fade into the colour of surrounding points. Also, these warped pixel meshes' edges are obviously vector as well, so you can be dealing with a grid of limited detail but maintain 'infinite' vector edge-quality. I'm not sure if I'm making sense.

Like an intelligent way of interpreting image data to make the image scale up well. I guess the human element comes in in choosing how to break down the image into separate gradient-mesh objects in order to keep edge quality. Within each mesh-object I can't imagine there's anything more to it than clicking each node and colour-picking the corresponding colour from the correct spot on the original image.

I could be wrong though. No reason this method couldn't be used without reference to create trullly original artwork, but I guess the method of picking a colour for each location in a mesh is not as human-friendly as laying down generalisation and strokes. A little bit like pixel-art I suppose.

You can also use bitmaps in illustrator, with blending modes and clipping masks, though I don't see why it's not possible to create that stuff without them I guess.

[spline]

Ofcourse there are advantages to doing it as illustrations, that way you can allways make it look more perfect than the photo, and you don't have to be a very good photographer.

The guy is obviosly very skilled at photorealistic art and probably knows how to draw very realistic. What he lacks in imagination he is compensating in skill...
However some of those could just as well have been done in a 3d program and could that way even have been animated...
Though his techniq is probably sometimes faster...

[Ian Jones]

"just as well have been done in a 3d program and could that way even have been animated... Though his techniq is probably sometimes faster..."

Make that a definite... 3D does take longer. Especially if you are considering the cats as the project.

[stacy]

He has his place cut out.
His clients can pay one artist a couple grand one time
to do an illustration that can scale down to less than
an inch for web or print catalogs or they can scale the
exact same piece to fit a billboard or even bigger.
3D modeling would be more expensive for the
client and in most cases probably not necessary.
He probably makes more than a photographer
would for the same piece, but the client doesn't
have to go to the wall for 3D.

Smart kid. Smart niche.

[sedgemonkey]

I'm a little puzzled by this... this level of work seems like it would be more suited to modeling than Illustrator. Why not go the extra mile and make the cat come to life?

The whole idea of basically tracing a photograph and producing a vector piece that looks just like the photo takes supreme dedication -- it just seems rather boring.
_________________
HEY YOU! We're looking for a few good pixel artists

[B0b]

wasn't it all created with the plug-in anyway? looks like an intelligent tracing tool to me.. is a bitmap image applied to the mesh? if so that'll bottom out after a 140% increase in size unless the line screen drops..

[stacy]

B0b

Capt. Fred Said..."There's a gradient mesh tool in illustrator.
Draw a shape, and within that shape you can manipulate a grid. Not a linear grid but one made of paths you can manipulate like all paths in illustrator. You create gids of any complexity you like. The points where the grid lines meet can each be given a colour and illustrator interpolates between the coloured point(intersection) and the neighbouring point - 'gradient mesh'."

I take that to mean the mesh is all vector. There's no bitmap anywere except to use it to pick colors from to place on the corresponding intersections in the grid.
It's jillions of little vector color gradiants down to almost the pixel size.