Human players generally design better ships than the AI does in the sense of balancing attack and defence, creating balanced fleets, etc. The AI updates its designs less often as well. The AI rarely seems to "max out" its designs by cramming in the maximum possible weapons and armour in there. Let alone engines... They seem to value cheaper designs, but sadly, numerous weak ships are easy prey to faster, more powerful designs which then gain experience and become even more superior.

Also, wonders like Eyes Of The Universe mean that you don't have to waste any space on sensors, for instance.

There's no intrinsic difference between Laser-1 and Doom-Ray damage, just that the Doom Ray does a lot more of it.

They seem to value cheaper designs

Usually they just seem to research less military techs, in one recent game I was at war with the iconians and they had only reseached to lasers 2 for their ships while I had phasors, and I was playing on challenging.

Trick is to look at your opponent's ships before going to war with them. It doesn't cost that much to get advanced espionage so start putting money into it as soon as you can afford it. Look at how your opponent is outfitting his ships. The AI tends to specialize in one type of weapon and defense. Initially. build your ships weak with respect to theirs, don't defend against their weapon specialty and build weapons to counter their defense specialty. That will trick them into staying on their course. Assuming you've been fattening up your treasury, upgrade all your war ships to the opposite before going to war. The opponent won't be able to react in time. You can beat an enemy with a higher overall military rating that way. In some cases, the advantage is huge and you can wipe them all over the map with little expense on your end. What really kicks butt and is super fun to do is set up your ships in a perimeter and take out all their planets in one turn. They don't even get to fire a shot...hehe, cool.

Something I am missing from Master of Orion is big fleets of hundreds of ships. In Gal Civ even with the latests logisitic techs we just have a few ships in a fleet. For me a true fleet is like 30 to 40 ships at the least, not 5 to 10, that is called a squadron not a fleet   

hat works unless they are the one to declare war and then won't end it

Yea, true, but I usually put a lot of effort into diplomatic tactics so I don't have to go to war until I'm ready. There's a point in the game where it gets sketchy, but I can usually keep them at bay.

I read the one about the ship that shaped the galaxy and followed up on the replies. What the heck are you people talking about? Towards the end of your research ( I have played where I researched everything but the last two techs on the research branch, which gives a tech victory)you should have the same weapons and defenses available as in any other game in which you pretty much research everything, true? So what are you talking about when you say creating space ships that have weaponry superior to anything the other civs can throw at you. Is there a difference between the last laser you can research vs. the second or third along the tree, a difference besides damage and size at least. This is all very vague to me.

The game I wrote that particularly story about was a "Normal" research game, played in a "Small" galaxy. There wasn't anything like enough time to reach the end of tech branch, which meant that my early use of beam weapons & defense influenced the AI's choice of tech research (i.e., they kept on with the beams & beam defenses). That, in turn, left them more vulnerable to ships armed with mass drivers.

True, if all of your games go long enough to let you research everything, then yes - you (and likely the computer, if difficulty is set high enough) have access to all the techs, and the only question is if you have enough money to keep shuffling between designs, rock-paper-scissors fashion. But with slower tech advances, smaller maps, or early-game wars, shifting the opponent's attention to one particular weapon/defense combo can be advantageous (provided you're using another).

Slow down your tech research setting, if you want a more interesting early/mid-game.