YAY! Delighted to hear it, Jim.

Hey Jim, any chance of a few screenshots of the new fish models being built and deisgned?

Just for a different style view about the development of MA3

There's seldom anything interesting to see until it's all put together. Just a bunch of hexidecimal numbers on the screen with an occasional header. Here's a grab of the screen I happen to be lookiing at right now from the Yellow Tang (the models are about 3000 lines each).

4;332,330,326,328;,
4;333,329,327,331;,
4;342,344,332,328;,
4;343,329,333,345;,
4;346,336,334,348;,
4;335,338,340,337;,
4;347,349,335,337;,
4;334,336,341,339;,
6;212,217,216,215,214,213;,
6;230,231,232,233,234,235;;

MeshNormals {
358;
0.099731;-0.115414;0.988298;,
-0.365069;-0.912424;-0.184954;,
-0.335409;-0.643249;-0.688281;,
-0.307242;-0.238864;-0.921166;,
-0.273106;-0.033498;-0.961400;,
-0.320791;0.263465;-0.909769;,
-0.514972;0.531775;-0.672324;,
-0.664222;0.719191;-0.203894;,
-0.870897;0.491466;-0.000127;,
-0.396798;-0.917906;0.000000;,
-0.348900;-0.768329;-0.536600;,
-0.279926;-0.439617;-0.853451;,
-0.211887;-0.021547;-0.977057;,
-0.240648;0.153382;-0.958417;,
-0.491076;0.314360;-0.812417;,
-0.710836;0.519615;-0.474038;,
-0.595284;-0.803515;0.000000;,
-0.430241;-0.583266;-0.688980;,
-0.228348;-0.218423;-0.948762;,
-0.134088;-0.028223;-0.990568;,
-0.664714;0.564253;-0.489667;,
-0.704678;0.540173;-0.460046;,
-0.411423;-0.687044;-0.598916;,
-0.255544;-0.481640;-0.838284;,
-0.160338;-0.275724;-0.947770;,
-0.007050;-0.169816;-0.985451;,
-0.326309;0.621407;-0.712303;,
-0.306784;0.542491;-0.782040;,
-0.066073;-0.635483;-0.769283;,
-0.069226;-0.573083;-0.816568;,
-0.110896;-0.444920;-0.888678;,
-0.166169;-0.064063;-0.984014;,

Also, bear in mind that even when the new fish models are finished, they probably won't look any different in a still shot. It's their behavior which will be changing dramatically.

Thanks :D

I really can't wait to see the new behavior, how far along is it?

Will this new behavior include slightly different shapes, sizes and textures for one species when you have multiple of them in a tank

yum!! when's the next feeding? ;)

This is still a long way off. I only recently made a breakthrough which makes it possible at all, and now I'm starting to move from mostly research and experimentation into the WORK (the most dreaded part). Some items are a close call. For instance, after viewing many videos and live examples of Yellow Tangs, I'm not sure if I should spend time on movable gills for the Tangs. I'm wondering if anyone would really notice.

The shapes probably won't change from one individual to the next, but the size and texture will.

Going from 162 vertices to 358. Nice!

Post the .X file and I can load it up in DirectX 9 viewer. ;)

Well, now that you've pointed it out (the gills), that's probably the first thing I'll look for... ;) I think overall, after you're done, you'll be happy you did it. It's these little nuances that has always made a difference in your work from others.:TU:

I agree with Surferminn. Unless there are performance problems the more subtle animations the better. No one sees all of the motions at once anyway but this illusion is meant to be contemplated at length and I think that most people that buy it will do just that. I have certainly spent hours at it already and it isn't even complete! Even if people don't have the vocabulary to describe the subtlest anatomical motions or know the word operculum they will *see* the motions as more authentic no matter how subtle they are. If you listen to a piece of music rendered with cathedral organ voices exactly as sampled at the source, for example, you will hear it as very "churchy" sounding. But if just the slightest hint of vibrato, wah or flange effect is added to those same voices the same piece will be perceived in a completely different way by everyone who hears it even if they don't have the knowledge to describe the differences that they hear.

Well, I guess I have to move the gills, now that everyone will be watching for it.

All together now - DOH! :)

I agree about the detail, though. I never would have thought to look for the tail muscle rippling before I saw it v1 of the aquarium. Now I miss it. And the specks of debris in the Prolific Goldfish saver really add something to the feeling of realism, though I know we won't get that here. My point is, it's the little details which bring the whole piece to life.

Fish Detail
 

I couldn't agree more. The thing I love about Jim's product is the amount of time and effort he puts into his product. I have seen alot of other "Fish" screensavers and they are not up to the quality and this is why I keep coming back to MA3.:TU:

I also agree!
"Tail muscle rippling" - Ahhhh!!! ..... If only!
Thinks! .... If we apply a little pressure, do you think we could get the 'N' key (or an equivalent effect), resurrected!!! :)
MA2.6 is still way ahead of MA3 with respect to fluid lifelike movement, IMHO!
Of course, Mark and I both know that is not true!... ;)

Sounds like years from now. :(

OMGoodness. When I came on and saw the NCalc in Cliff's signature, I thought it had something to do with him being some kind of obsessive math nut or something, I had no idea it had to do with the aquarium. :)

Well, you got the, obsessive nut, bit right! :D

I'm just gonna sit here and keep quiet! ;)

A much shorter wait in calender time I'm sure but it may *feel* that long to our favorite developer :D

If I stare at that text long enough will it be like one of those paintings that were popular in the nineties that looked just like a random pattern but if you could get the focus just right there was Elvis on a bicycle or whatever ?

Okay silliness aside , the yellow tang is going to be the first remodelled fish , or was that just a sample you are working on , or should I just mind my own beeswax ;) ?

No disrepect to Elvis intended . He was/is the king .

Yes, the Yellow Tang is the one I'm working on. Then the other Tangs, because the movement will be the same. Then the Butterflyfish, because their movement is similar. Then the Triggerfish, then the Lionfish. Then the weird little guys like the Percula Clownfish, because they are the hardest (wiggliest).

Yellow Tang - First and Last and Always! - Best in show since 2000...

I just hope all the work I'm putting into it actually shows when I get it in the tank. The texture is twice the dimensions of the old one (4X the number of pixels). Part of the realism of the old version came from the false shading around the belly area (the shading was simply part of the texture). I can't do that any more, because the fish will now be able to bank, and even turn completely upside-down if they wish. It would look unrealistic if the shadows remained on the lower side during such maneuvers.

Well I wouldn't expect to see loops or barrel rolls, but it sounds like it should pay off. :TU:

WHAT!!! no loops and barrel rolls :)

Nope, but a hammerhead might fit with the genre nicely. ;)

I like the tail slide part before they kick it over. :)

Well said! :D

Well said, YT. - I didn't think it would be long before we had a comment like that! :TU:

I've got a rudimentary version of the new Yellow Tang model in the tank and have been doing a lot of experiments with it. There's no movement yet, but I can rotate the model. The first thing I found is that my new 256x256 texture is completely wasted on a fish this size. I resized the texture back to 128x128, and if anything, it actually seems to show MORE detail. No reason to have a 65k texture for any fish that can do just as well with a 17k pic.

As some have pointed out, the Yellow Tang was already probably the most realistic fish, so the differences in the new model are not dramatic. The shape is more accurate, with a thinning of the face between the eyes and the mouth. The spines in the fins show up better. The side fins are now attached to a mound on the body, and are not just "free-floating" as they were before. The "tang" (the sharp whte razor near the tail) now actually protrudes instead of just residing in the texture.

But on an object this size the visual changes are subtle, just setting the stage for the new movement algorithms. That's when the little guy should really come to life.

Sounds like you've really got the detail you wanted to.

Have you tried running this new prototype at 1920 x 1080? I picked up a 24" Acer that runs this resolution for $250.

My resolution is always 1920x1200.

Hi Jim,

Would it be possible to have multiple color selection on the percula clownfish since you're remaking the fish models?

The different species of Clownfish really have different shapes, so they will require several models. Some day.

Hey Jim, Didn't want to start a new topic but just wanted to see how you were getting on with thr new models and behaviors? See how your doing getting over the problems you were experiencing?

Not doing well at all. Stuck.

Sorry to hear you are stuck, its been such a long time since any updates. Keep the faith!

Its a shame you can't team up with a third party software development company or take on some staff with you in charge of a development team. I'm sure it would help get the product moving quicker and the experience and knowledge would be there amongst a team to allow the aquarium to benefit from the latest technologies that are out there like DX10 and DX11 which are supposed to make things easier to program and better, more lifelike in appearance. You'd have people to consult with and be able to offload sticky areas, owners of the latest PC's would find the screensaver makes better use of their computers, the screensaver would look better and would have a final release out a lot quicker.

I'm sure though you must have considered it many times and there must be real world obstacles that have prevented you going this route ie, if it were so easy you would have gone that route already.

But are you still trying? Any thought of taking a little pause in the movement, and perhaps let something in the background come to live? Sometimes a small pause helps. :)

The only reason Marine Aquarium has been so successful is that it's a one-man operation. Over the years, major software companies have assembled teams to try to duplicate it. The two largest companies abandoned the project after a few months (one ended up licensing mine). I have worked on such teams - Cinemaware, for example - and in my opinion, it's difficult or impossible for a collection of individuals to really produce anything great. They are far too concerned with politics and power-trips. It always comes down to "Who are the monkeys?" The programmers see themselves as the geniuses, creating elegant algorithms which only need some monkeys to throw some pixels together so that their creation can be displayed on a computer screen. The artists have the opposite viewpoint. They see themselves as the Creators of magnificent visuals, and only need a few monkey code-writers to move their animations around.

Each group battles for power, and is frustrated by the stupidity of the other, so very little real communication takes place. Only when the two types live in the same brain do they respect each other enough to accomplish much.

Jim, it's absolutely clear that you are the recognized expert in this area. It looks like you've spent 6 months or so on actual implementation.

Are you totally convinced that what you envision, is practical with today's off-the-shelf PC?

If so, why?

If not, then what would be necessary to prove whether it is or is not practical?

I believe the success of Marine Aquarium is based on the quality of the artwork and attention to detail. Do I think quality is exclusive to 1 person working alone? No.

There are just too many examples of extremely high quality movies, TV shows, software, games, etc. that were done by a small team of like-minded professionals with an artistic vision.

After 9 years, which is most likely?

A) It is impossible for a creative team built on mutual respect to produce a great product
B) Jim has simply had rotten luck in the teams he's been involved in.


And to Dale, I suggest looking at the best games released in the last 5 years. They are photorealistic, with incredible lifelike animation, motion, lighting. They are almost indistinguishable from film. Today's computers can handle anything we want to throw at them.

Hi everybody.

So, one way to move forward (possibly), is to sacrifice compatibility with older computers and raise the system requirements to something like 2 GHz processor, 128MB graphics, 1GB RAM? Just a thought, and might not be part of the vision of Mr Sachs.