BTW, According to Avault, MOO 3 has a better AI than GalCiv. It rated its "intelligence" higher than GalCiv.

Here's the Avault forum if you want to read what I wrote there:

http://forums.avault.com/cgi-bin/ultimatebb.cgi?ubb=get_topic;f=3;t=001071

I too think the review was unjust. You have to spend time with this game to learn the nuances."Onions have layers" and Galciv is one huge onion! A delicious one I might add

This review really sucks and is especially hurtful due to its gross misrepresentation of a piece to this game that is beloved by its designer and extremely valuable to those of us who play this game. They really need to reconsider what they published.

~SDC~

Ouch, not a very accurate review, and not inkeeping with the praise GC is getting in General. Saw your post brad and i agree, once i get home from work i will add my piece in that forum too

~SDC~

You have got the be kidding here. The Heroes IV was known to be one of the biggest AI disasters in gaming history. When the game came out the AI hardly knew how to even move. Goes to show how reviews are 'bought' or written by reviewers who didn't play the game properly these days.

The MOO3 AI rated higher? Not bad for an AI that doesn't know how to invade planets. That same AI was easily beatable at the highest difficulty level where it cheated like hell too!

~SDC~

Yes i was shocked they said Moo3's AI was better..its impossible to lose a game in Moo3 i found. The AI only ever seemed to make Troop tranports, they NEVER attacked me, NEVER invade a planet and destoying their 400 Troop transports was rediculously easy even on the impossible level.

Oh yea and then there is that Diplomacy system where the AI's will repeatdly tell you to cease your actions..yet you have no idea what they mean, then the declar war one turn, peace the next etc.

~SDC~

I'm sorry this upset Brad so much. But at the same time I am glad he is so impassioned about it. If he wasn't we wouldn't see this sequel from so long ago fleshed out to such detail. I have stated I have bought GalCiv primarily for the AI, and this is quite true. And as critical as I have been. Impassioned and cutting to the bone in my suggestions, you can tell it is because I have something new in my life that is worth nurturing and helping to blossom, season after season. It was obvious to me he was mistaken as well. One of the reasons I have read www.gametspot.com for so long is that they to touch bases with the developers about any major issues and state what was discussed and whether changes are in the works or not. I can count on one hand PC stetegy games that play this fair.

~SDC~

It is especially odd coming from Avault, for 2 reasons. First off I respect their reviews.. they are usually dead on. But here is the interesting thing. Typically, they come out with a review a rather long time after release. Even if they had received the game well before it was released to the public, you don't see reviews there until well after it has been out. I've seen reviews crop up there for a popular game well after a month has gone by. It seems to me that this reviewer did NOT spend enough time playing this game to make an accurate review, not the kind I normally expect from Avault.

I hope they do make changes as they said. He made a huge assumption about the AI cheating, when this was one of the selling points of the game. Hope an apology is forthcoming as well.

fleabitfox

Have to say, I think Brad is on the money re: the AI being scored so low. What a crock. This reviewer seemed to be upset that it actually challenged him. Maybe that's understandable, considering how we've come to expect dumb AIs from strategy games. But come on--considering how totally braindead MOO3's AI is, and that got a 3, the 2.5 score for GalCiv is crap.

And to just flat out accuse the game of cheating in a review, without any evidence but instead extremely circumstantial evidence (which was shown to be flawed) is just unprofessional. If I were the owner of Avault and was made aware of this, I'd be laying the smackdown on Brian pretty quick for his shoddy research into the game.

I just double checked the source code with regards to constructors and the Yor AI personality. Simply put, the scenario he claims cannot possibly have happened. If the AI has no money, it cannot literally lease anything.

But that seems to be the central thing in the whole review -- leases. He didn't know how to lease. At normal AI and beyond, the AI starts using leases. The net result, the AI is cranking out units left and right. The whole review is thus tainted since from his poitn of view, teh AI is just cheating like mad.



~SDC~

test

~SDC~

I have just posted a lengthy comment on said forum as have others. feel free to comment.

~SDC~

In the beginning, I thought the AI HAD to be cheating. What I realized later was that my whole game strategy was the same strategy I used for all of the other Space TBS in the past. All I really did was make minor modifications and apply it to the new game, and it worked because the game wasn't really essentially different in terms of AI.

But in Galciv the reality is that the approach of the AI is so radically different that my MOO1/MOO2/STARS/CIV1/CIV2/CIV3 concept of how I expect the AI to behave gets me killed. I had to fundamentally change my idea of how to play against the computer.....and that's the idea....

What I am finding is that people accuse the computer of cheating because they don't realize that their basic concept of how the AI will respond is based on the other tbs games and the according strategy gets them easily killed.

I had expectations of how things would turn out and the AI exploited them, just like any good enemy should.

Brian Clair (Publisher) - "AVault reviews games out of the box."
Richard Leader (Reviewer) - "Stardock's first patch for the product..."

Graphics gripe about the video cutscene when starting a new game is also inaccurate. Click the mouse and voila, no more cutscene... Or is that just me?

Gameplay gripe about social improvements taking a long time is... misleading. They do take a long time when your economy is weak. When you control dozens of worlds you can build some things 'next month'.

"even after applying the latest patch"

Stardock *has plans* for a Metaverse? What is it that I've been submitting my scores to since it (GalCiv) was released, then?

A mention of the BonusPak's MP3 player. Was that in anyone else's copy "out of the box"?

Just on the AI stuff as well - the 'evil' races he's talking about are probably picking military production bonuses, thus giving them cheap ships, including Constructors. Easily explains how they get more of them, earlier.

And he closes on a note about not being able to recommend it "out of the box" when he's mentioned features that aren't even in the box in a couple of places. Possibly more, since I don't remember specifically what came with the BonusPak.

To tout this on the front page of their site as an "in-depth review" is ludicrous. While he's obviously done more research than some of his "Game Reviewer" brethren I could name, he's not looked deeply enough into certain aspects of the game, instead choosing to come to his own conclusion about a feature.

I don't blame Brad for getting annoyed at this... But sadly this sort of gaming "journalism" is becoming far more common these days.

*clambers down off his soapbox*

~SDC~

Lets not get too wrapped up in this review. After all a 3.5 = 70%, versus MOO3's 3.0 = 60.

See this quote from their MOO3 Review:

Overall: Master of Orion 3 has proven to be one of the most difficult reviews I've ever had to write. The elements for a great game are present, but the focus and execution of those elements is not. Combined with an underwhelming and questionable AI, this puts MOO3 at a disadvantage right out of the gate. In the end, it's difficult to recommend this title to anyone but diehard MOO fans.

So post something in their forum if you want but lets keep it rational then move on. It's a complex game some people are going to get, and some people are not. Now I am off to work on my GalCiv Mod!

~SDC~

I'm not wrapped up in it, I just read it out of curiousity (it's 5:30am over here and I can't sleep ) and noted the discrepencies as I found them.

Comparing it to MOO3 isn't so much the point. The review contains at least one glaring error that the reviewer doesn't seem to have investigated.
I would post something in their forum, but I've found that writing to such places doesn't actually lead to any kind of result.

Maybe I'm just too cynical. Or drunk.

~SDC~

True, but if they give the AI a fair shot, it should score at least 4,5 stars. That should up the total score to AT LEAST 4 stars. I can live with 4 stars but not with 3,5 stars because of a crumby AI.

~SDC~

I really want to jump in and point out that this kind of reviewing is most definitely the exception and NOT the rule at Avault. An amendment will be made to this review if deemed necessary, I'm sure.

~SDC~

Good lord! I just read that review. I won't reiterate everything everyone else has said, but that was without a doubt a hatchet job. Geeze.

~SDC~

I wonder why his improvements are taking so long (the reviewer's) I mean in this game I'm currently playing right now, between Rubicon, and Sol, I'm cranking the MAJOR civ, trade, racial wonders at the rate of one every 3 to 6 turns. Most other improvements on those two planets take about 15 to 20. The big problem I'm running into is that I got too many planets kicking out stuff too quickly. Not to mention I'm making money.

I didn't realize until recently how huge of a difference it makes to do the following things....
Set propeganda for most planets
Set your initial tax higher
Lease from time to time
Adjust your budget sliders
And stage multiple starbases in a sector

Now I've got this goaliath of an empire that's registering a 130 military, 306 research (!!!), and about 180 on the economy.

So it's quite possible to out-produce/expand the AI if you play your cards right. Now if this idea of downloading AI tweaks based on Metaverse games sees the light of day, I'm sure things will get really interesting.

SeanB

The reviewer has responded to the thread, for better or for worse.

~SDC~

Well to those of us who actually "played" the game we know what's up. I've spent 5-10 hours a day with this game since release and feel confident in my/our opinion. The AI is real sneaky and intelligent but does NOT cheat!

I've been a columnist for Avault for 2 years now. I've written most of their developer articles. I'm very dismayed at some of the arguments coming back from them.

The old "Game X was reviewed by Reviewer A and Game Y by Reviewer B so you can't compare them.." is not the kind of thing one should expect from a professional game magazine/webzine.

No one expects a reviewer to look back at every similar game over the past 2 years. But looking back in the past few months at other reviews for similar games wouldn't be unreasonable in my opinion.

~SDC~

Something tells me Brad won't be writing too many more Avault developer articles from now on... (that'd be a shame - there's some interesting stuff in those articles about the gaming industry, ideas about non-copyprotection ways of combating piracy, etc)