An hour long trailer clip for the upcoming Firaxis X-Com game.  Some really good looks at the tactical combat and other stuff.                    http://www.rockpapershotgun.com/2012/08/29/enemy-yours-watch-an-hour-of-the-xcom-remake/#more-121893

I want to be excited but I'm not sure if I like the style.   It looks a little cartoony for me but if the substance is faithful to the original, then I'd still likely get it.

Still looking like a first day must buy for me.

I'd really be surprised if this doesn't end up being a huge dissapointment. I do not think a console game can pull off a strategy game that is home on a PC. The memory limitations alone almost garantee poor A.I.- heck their last title Civ5 can't even beat you with cheating aside from a tech race. It's the wrong platform aimed at the wrong user base. What Railroads was to Railroad tychoon this will be to XCOM.

I'd love to be wrong.

Originally Posted By ghostryder (30 AUG 2012 8:53pm)

I'd really be surprised if this doesn't end up being a huge dissapointment. I do not think a console game can pull off a strategy game that is home on a PC. The memory limitations alone almost garantee poor A.I.- heck their last title Civ5 can't even beat you with cheating aside from a tech race. It's the wrong platform aimed at the wrong user base. What Railroads was to Railroad tychoon this will be to XCOM.

 

I'd love to be wrong.

Ummm............you do know this is developed for the PC don't you?  I also have concerns, but mine are based on the fact that every Firaxis title since Civ IV has had problems at release and needed copious patching.

It is being made for the console and being ported to PC. That has already been announced. Like DA or any other title-being developed towards the lowest common denominator- which is the memory restricted 256M PS3. Do you know that isn't even enough memory for your computer to browse the JAVA based website of amazon.com?

That tells me two things very important we all need to recognize.

1. It will be impossible to code an A.I. with a full featured language with libaries. They will have to use something else with a very low overhead- in almost most cases something like LUA. An old assembly line industrial computer language designed for configuration on an assembly line- not databases, speed, nor really anything you could base any A.I. on beyond basic pathfinding in a game. (and why FPS's dominate consoles). XCOM had a lot going on on the field and strategically off the field as well that absolutely requires the best most devloped full featured language possible.

2. No designer would happily restrict themselves to the confines of a system so behind the hardware curve. This was a bean counter driven decision and anytime they get that involved in the development of a game the results are likewise compromised. It'll take some real impressive guru to wrangle out anything in that tiny box- and given what we see in Civ 5 I don't think anyone at Firaxis is close to being that good.

Fire up baldur's gate 2-- mod in a lua based script for party managment and let me know just how much your impressed? Heck use the default ones. Fact is your be hitting pause like every other player and taking control because of the awful job the scripts do. That's exactly what they are working with here- for the console dollar by order of a guy in a suit that doesn't have a clue what XCOM was or a new follow up could be.

It would be very wise and smart to approach this one with a lot of doubt.

I have been waiting for this game for close to 2 decades now.

As for the console port issues, the game will be moddable. I've never seen a game with mod support that the modding community couldn't fix. I am very, very saddened to hear that this is a console port, but hopefully they've been doing a lot of work to actually make it a "good" PC game as well as a console game. Only time will tell.

Originally Posted By ghostryder (30 AUG 2012 8:53pm)
What Railroads was to Railroad tychoon this will be to XCOM.

 

I'd love to be wrong.

Sadly, this is exactly what I fear from Firaxis.

I share Ghostryders fear - and his hope that he is wrong.

Fortunately, an independent producer is trying to do an updated, modern X-com.  I have more hope for it.

Has there ever been confirmation that this was devloped for consoles and then ported to the PC as opposed to the other way around?

straight from firaxis:

NEW YORK&mdash

It'll be the same game on each- you might get a texture pack later for the PC to replace the low rez ones that are default console-or someone will mod them. We're see how it turns out in 5 weeks.

That just says it's being released on both formats.  It doesn't address which one it was designed for.

Gambryo-like Unity is cross platform. There's no need to dev to each one then do ports. You dev to all at the same time. then you simply compile to the intended format. Same code for all. Mono auto translates. In Unity the engine even automatically scales down the textures. (not posative on Gambryo never used it). Therefore because the PS3 and 360 are in the mix- you restrict your coding to meet those constraints. The code still runs on the PC once mono translates. Typical .net stuff.

I'm using Unity so i could easily target the console...but I want good A.I. that's dynamic and not simply triggered-which this game technically would have to do to play on a PS3. You can absolutely forget anything alone the lines of the aliens being able to make decisions depending on the current situation or to formulate plans in order to accomplish a task.  A*, stack based FSMs and Goal Oriented Action Planning algorithms ain't going to happen ever on a console. Your getting the very same code on the PC.

I'm not crazy about the camera jumping all around like it does with close ups and such...besides if I was truly interested in this game then I would probably need to buy a console, because I'm uncertain that my pc can handle the graphics...

Since the original was near perfect- in terms of difficulty, balance and fun combat I agree this will never be close to the original. I don't think any commercially produced follow ups currently can compete with games that were made when creativity was king- and there was no 'corporate' bosses thinking up ways to make more money. In those days sales gimics didn't bring you financial success but rather the actual quality and innovation of the game you produced.

It was still a buisness especially as the companies grew but overall the game designer was willing to throw the dice on new ideas. That produced some great gems- but it also was the downfall of some good design houses. Overreaching doomed houses like Interplay dispite having great games in their line up. On the flopside houses that were once hailed as leaders released disasters as soon as the corporate bosses entered the picture. Who can forget Papayrus, front Page Sports, Dynamix and all the adventure lines under the Sierra umbrella? As soon as Ken and Roberta stepped aside and the company went corporate every franchise line under it exploded into doom. the last Frontpage Football game was so bad it was totally recalled and no attempt to 'fix' it ever insued. Papyrus--still to this day the only company that ever offered such realistic detail in racing games-was dissolved- every adventure game flopped-in fact the intire gendre was wiped off the map. With the exception of the warehouse and distribution department, the entire studio was shut down. Sierra then terminated employees at the previously unaffected Dynamix, Impressions and sublogic. By 2000 the company itself didn't produce games at all- it now centered on casual games and home productivity software.

Today if it publishes games at all it's from third party developers like Relic (Homeworld).That model extends to EA, Strategy First,UbiSoft and Sega.  In my mind not one of these houses has produced a game that could be called innovative or an all time great. CA simply builds on it's default Total War Series with added features and new time periods- underneath we've played that same game to death. And that's the best line up any of them offer. Firaxis hasn't released anything new since Alpha Centuri and not even that had any A.I. to speak about. Even IF it were possible to do XCOM on a console this company wouldn't allot the time neccessary to pull it off. At best your going to get standard FPS logic. Get within range and the enemy then belines toward you. I'd be surpised to even see take cover tatics at all let alone flanking or any other intelligent abilities.

It's why there's a lot of cut scene footage and barely any gameplay footage. Nothing whatsoever on how the game actually plays just 4 and half weeks from gold?

Twitch had a 1 hour gameplay video.So watch and start hating.

http://www.twitch.tv/2k/b/330155062

I am looking forward.

The original doesnt hold up well at all when I played it back in 2011.

yup as I assumed. node based set patrol paths- as once spotted weak beline logic - no tatics whatsoever using terrain or position and enemies seem to totally lack the 3 senses of the original and 4 required of today's game. To be aware of the environment, sight, sound and memory.

The scariest part dispite the player's frequent and many mistakes (not reloading. risky unguarded run and moves) and the fact the game was set to the absolute highest difficulty level he succeeded with only one death because the enemy A.I. was too dumb to follow through on multiple opportunities.

It may be significant that, while including multiplayer skirmish, they are not implementing singleplayer skirmish.  Could just be a priorities thing but it does nicely support the points that GR makes.

Cheers,

Andy

People tend to see what they want to see it seems.He was playing at normal level too.

What am i missing. please point it out. Dispite all the noise and mayhem not one enemy presented itself from other portions of the map. they were placed and didn't come alive until the player (who was awful) stumbled across them. That means sound has no meaning. From what i could also see the enemies moved in preset paths- instead of turning to a now blown hole in a wall or open door--hense node based patrol and no awareness of the environment. I never saw a single one take cover. (the sectoids were placed in cover and remained stationary dispite player advances--that was scripted)All just made the typical Beeline to the player. And at the end of the mission the discussion clearly between the players clearly mentioned the difficulty setting, which you correct was on normal. Ironman was mentioned as the typical can't save feature.

These were also presaved missions- obviously set up with triggers and not random maps which will be the bulk of the missions in the game-and even in the scripted maps and all the player mistakes he never paid for a single mistake zans one.

Then as mention the console'ish graphics and forced camera angle anoyness aside the true measure of the game will come from what occurs as missions progress and the global strategic elements take place- and again given the engine, the system and language used will likewise not be dynamic but scripted and linear.

Very interesting. Can you give an example of a game which uses advanced ai logic to create a credible enemy threat on the same level?

Originally Posted By ghostryder (31 AUG 2012 6:13am)

 

 

1. It will be impossible to code an A.I. with a full featured language with libaries...

 

2. No designer would happily restrict themselves to the confines of a system so behind the hardware curve. This was a bean counter driven decision...

http://pl.twitch.tv/2k/b/330605570

This video starting at 3:48:38 showcases the PC interface in combat. I probably said "oh wow" about 15 times(this is the first time I've watched a gameplay video, and I watched it because this thread gave me some concerns and made me nervous). My favorite part was using the mousewheel to zoom up and down and the fact that it pops the building levels up and down. Something you can't do on the consoles.

I don't think we're going to be disappointed.

F.E.A.R.

The enemies communicated with each other and assisted. Cover tactics were abundant. All senses- sight, sound, awareness of the environment were active and used=

it does a lot more. It's a FPS but what's important here is the algorithms used-- here's a pdf explaining them in detail:

http://web.media.mit.edu/jorkin/gdc2006_orkin_jeff_fear.pdf

“In F.E.A.R., A.I. use cover more tactically, coordinating with squad members to lay suppression fire while others advance. A.I. only leave cover when threatened, and blind fire if they have no better position.”

Since the states in F.E.A.R. have so many parameters, and the editing process is becoming increasingly more difficult, a planner is used to automate the solution. A planner works by analysing dependencies of each action, and figuring out how to realize them:

“We can satisfy the goal of eliminating a threat by firing a gun at the threat, but the gun needs to be loaded, or we can use a melee attack, but we have to move close enough.”

By spending more time to develop a planner, less work is required to implement specific behaviors (compared to a FSM for example).

 Don't let level designers script A.I. behavior!

“The designer’s job is to create interesting spaces for combat, packed with opportunities for the A.I. to exploit. Designers are not responsible for scripting the behavior of individuals, other than for story elements.”

The overall architecture inspired by MIT’s C4 architecture, which emphasizes the use of a blackboard to store facts about the world. Then, separate sensing operations are used to populate it with information, which is used by the decision making process. This provides a form of decoupling to increase the modularity of the implementation.

F.E.A.R. in particular uses a dynamic blackboard where multiple facts can be added dynamically, based on events that happen in the world around each AI actor. (This provides more flexibility but less efficiency than a static blackboard.)

[size= small]That last bit BTWis totally impossible to impliment in a language such as LUA because of the way it does databases so poorly. When your pulling data from the blackboard it needs to be available immediately as the A.I. loop gets to it. In LUA you cannot access any data until the table fills up. (hense the very jerky movement of the NPCs in Baldur's Gate 2 when a custom script is being used-while doing rather simple things like when to cast a heal or an offessive spell has to wait on the data)

Orkin release the source code to this A.I. fully documented and runs his site solely aimed at game A.I.-- it's an excellent place to start if your interested in learning it.

And no- none of that is happening here. As the players repeatedly tried to convince us of how 'difficult' this game was they forgot to forego the Kill screen at the end of the mission-where they clearly stated the mission took place at about a 3rd of the way into the game. There was but 7 loses....all but the one you saw die rookies. Your not get through the original-even on easy-with such small loses at that game point.

Watch the video. As soldiers advanced the sectoids stood motionless as they were approached and surrounded making no effort to reposition. The big guys on the right-who were 'suppressed' refused to take any cover at all. I'm glad what's called 'suppression' is modeled but who among us really believes 'suppression' means petrified like a deer caught in headlights? The only real reponse was the saucer taking out the soldier that was standing right in front of it by the players bonehead move- and though it was hit by another Xcom agent-and turned back from the spider shape it made no attempt to reposition. Far from diving through windows, knocking over barrels or any common A.I. behavior in F.E.A.R.

A shooter 8 years old that has superior A.I. found in almost all Strategy games. What's wrong with this picture? Why are gamers who purport to love the gendre settle for ho-hum? Would they buy a chess game that has A.I. uncapable of beating them? They buy strategy games that can't- and lay praise to them as the best thing to come along since porn.

I just do not get it. Good A.I. is very doable and it's high time we expect, demand and anticipate it. Those who think ho-hum is okay are doomed to get a shovelful of more ho-hum dumped on them. In our backwoods gendre of wargaming and strategy the developer is like that old lady in front of you at the checkout counter with a pile of coupons to save a quarter holding up the line. They'll cut every corner they can get away with and if you don't care much for quality A.I. in your games you can bet your joystick your get the bare minimum effort from them. It's been going on so long people are beginning to accept and expect poor A.I. and they even will argue excuses for it. Fact is there are no excuses for being lazy and cutting time off the most important aspect of strategy games. The time of the developer  tacking on a half-butt'd attempt of A.I. at the final 10 days or two weeks of development is long past due to end. It's totally unacceptable and I'll keep beating that drum until this low acceptable standard changes to what it can and should be.

full presentation of the A.I. that should be the base standard in every game:

http://www.media.mit.edu/jorkin/aiide05OrkinJ.pdf

I based the A.I. of MJ12 on this- also a squad based combat game. prototyping that really showed a contrast to what I see in this video. I'll be modifying it to drive EB:G as well.

GR is right on several levels. It think it boils down to this- the reason so many games made at big studios are ho-hum for people over the age of 18 is because the guys pulling the strings (funding the project) don't really care if they are making computer games or roller skates. Thats an over-statement, but games are not made for the purpose of making a great game, they are made to make money. There are exceptions to the rule, but I doubt this game is it. That is why these type of games give you flashy cut-scenes and fancy special effects, because that is what the younger crowd looks for.

Ive said this before, but I will repeat again: What made X-COM special was the little things. I think one of the biggest things that made X-COM neat was the story. It made clever use of existing UFO lore, it played off real fears that some had. Its a given any UFO plot isn't going to grounded in reality, after all, but the plot was not ridiculous and outlandish like so many clones since. X-COM tugged the threads of real UFO lore and played on the subtle fears that people have of the unknown. Every clone I have seen started with an outlandish story from the beginning, a story that lack any nuiance or plausibilty in any sense.

It certainly help that its tactical and strategic layer was fairly deep, but I contend that story is the real thing that made it special.