Inside: SereneScreen Fan Forum - My personal opinion

Inside: SereneScreen Fan Forum (https://www.feldoncentral.com/forums/index.php)

- Marine Aquarium 3 for Windows (https://www.feldoncentral.com/forums/forumdisplay.php?f=46)

- - My personal opinion - it's time. (https://www.feldoncentral.com/forums/showthread.php?t=5144)

Quote:

Originally Posted by snook (Post 117446)

Where is Commander Data when you need him? I'm sure he could find all of the collision points in a few seconds.:lol:

Here he is...and this is a true fan:loco::loco::loco::lol:
.
.
.
.
http://farm4.static.flickr.com/3474/...0d6b9659a9.jpg

Kudos to the artist as it's amazing work, but for the person carrying the art? Run away... run away... !

i might be talking completely out of my butt here, i'm certainly out of my depth anyhow. however, jim what you are explaining about your problem sounds massively complex and almost infinite of avoidance of collisions.

you may have already been down this route, and please do excuse my lack of knowledge in this field but, is it at all possible to break these functions down.

based on object going from a to b, could something be achieved by looking at the objects current position and looking at it's immediate available movement and positions possible for maybe a certain amount of moves/time? basically in steps breaking the whole process down into more managable segments. so maybe at a certain time object 'a' could only be a certain place while object 'b' could be a certain other place avoiding all possibilities of collision of object 'a' within this certain time frame/amount of moves?

did that make any sense to anyone? sorry for my basic talk, as u can see i'm not technical in this side of things lol.

The problem with breaking down an infinitely complex problem is that each step is also infinitely complex (or at least that's how it seems to me).

I'm convinced that the solution is not to think of objects avoiding each other in a nearly infinitely complex 3D space, but to find a whole new paradigm - a trick that APPEARS to work the way 3D objects would, but is actually run by a little man behind the curtain.

Let's say each fish has a "preferred" vector or direction it wants to go. That vector is extrapolated about 3 inches ahead of each fish. If that vector crosses any other fish, then both fish need to be "steered" slightly off the vector line. If you are starting to steer around another fish with fully 3 inches to go, it's unlikely they'll conflict.

Also, by having spheres figure your avoidance, you're not taking advantage of the shape of each fish, which is larger in the front, smaller in the back, and each fish is mostly flat so several fish can slip past each other in a confined space if they are side-by-side. I think calculus will be required, or at least 3D geometry/trigonometry.

For 9 years, we've been hearing that outside help cannot be trusted to maintain the quality. I'm sorry that Jim has had so many bad business partners in the past (including Prolific), but there are people out there who probably would do a fantastic job helping on this issue. But maybe it will take another 9 years before help is sought. Game engines are getting so advanced now that things like hair, cloth, collisions, etc. are all "free" and just part of the engine.

Jim, I hope you had a great holiday season like our family did. Any updates on your progress. I'm guessing not since we haven't heard anything, but I figured it doesn't hurt to ask. Thanks!

Nope. I've been dealing with website issues lately. Trying to get my mind around Swish.

Quote:

Originally Posted by Jim Sachs (Post 117986)

Nope. I've been dealing with website issues lately. Trying to get my mind around Swish.

With all due respect - what happened to ...

Quote:

Originally Posted by Jim Sachs (Post 117212)

To paraphrase Thomas Edison - Though I haven't found algorithms for the fish movement or collision avoidance, I have found a thousand things that DON'T work. If I divert my attention now, I'll quickly forget what I've already tried.

I got talked into diverting my attention and went down the rabbit hole.

Don't waste your time learning Swish it builds bad sites with too much flashy content, go for plain and simple or let someone els build your site.

Nobody else can create the Aquarium, anyone can build your site.

Yeah, but they get money to do it. Anyway, I already bought Swish - it's about 1/5 the price of Flash, and their examples look really good.

Again this reminds me how lucky we are as end users to just click on a link or a shortcut and enjoy.

Quote:

Originally Posted by Jim Sachs (Post 117991)

I got talked into diverting my attention and went down the rabbit hole.

While you're down there, would it be reasonable to take care of a few other issues that have come up in the last 8 or 9 months?

[For a partial list, see the starting note in this thread]

Jim, - PLEASE!... just keep on going down that rabbit hole and don't stop untill you get to the mad tea party, - then enjoy yourself....

..... Don't come back, - until sanity has returned to this forum! ;)

Quote:

Originally Posted by cjmaddy (Post 118009)

..... Don't come back, - until sanity has returned to this forum! ;)

Your presence on this forum will be missed Jim.. it has been a pleasure.:D

I'm well aware that 6 weeks ago, Jim announced he was going to work on some simple additions.

And I'm aware that 5 days ago, Jim said: "A couple more days, and Beta 11 should be ready for testing by the general Forum members."

Four months ago, I started this thread.

I think the following question, edited from the quote below, is still a reasonable question.

So, a question - if 10 months between feature releases isn't very long, then how long in your opinion would be "too long"?

Quote:

Originally Posted by Dale (Post 117196)

I appreciate that it has been unexpectedly difficult to accomplish the major improvements in fish movement. We've almost all been patient, and we're quite willing to wait longer while Jim gets this done to his satisfaction - and then for the optional Creature Packs.

...however, Beta 10d was released over 6 months ago. No visible progress since then.

That's a long time for a single Beta release.

Personal opinion: it's time for another Beta release.

Assuming that there's nothing to release in terms of fish movements, there are still a lot of bugs and product features that have been covered in this forum - that could be fixed. Yeah, I know it would take time away from the work on "New realistic fish behavior, including schooling, chasing, and pecking at corals."

A partial list of things that (in my opinion) should be done before waiting another 6 months:

o Bug fixes and feature improvements that have already been discussed here.

o (quote) "Automatic lights. There will soon be a greatly expanded interface for customizing the background colors"

o The crystal improvements that have been promised.

o The various documentation, help-file, and UI improvements that have already been "promised" here.

o Perhaps even the starfish...

Now, I'm rather sure that Jim would rather keep focused on the realistic fish behavior. And I'm sure that expressing my personal opinion will draw some angry responses.

So, a question - if 6 months between Beta releases isn't very long, then how long in your opinion would be "too long"?

My own personal opinion is that someone not a million miles away must be a bit of a sucker for punishment to ask that question. Given numerous statements along the lines of "it'll take as long as it takes", and given that the whole thing is driven by someone with the heart of an artist rather than the heart of a production-line software developer, it seems to me that the above post and the terms in which it is couched are just asking for trouble.

But that's just me.

Quote:

Originally Posted by Dale (Post 117196)

Now, I'm rather sure that Jim would rather keep focused on the realistic fish behavior. And I'm sure that expressing my personal opinion will draw some angry responses.

Correct.

Quote:

Originally Posted by Dale (Post 120496)

So, a question - if 10 months between feature releases isn't very long, then how long in your opinion would be "too long"?

It will be done when it's done.

Both responses by Mark and Cliff sound right to me.

My take on "Too Long" would be when all work on the MA3 has been abandoned.
I guess the problem comes from defining what "beta" means. On our side of the fence we would like something new every other day.. On a practical level we get something new.. once in a while.
If that interval is too long... well, there is nothing in our contract with Jim or the this forum that demands we stay poised for testing. Enjoy the rest of the forum.. or not