Who's preordering X-COM?

[Ashenor]

Thanks for the info. Game does look fun. In the demo my sniper died first alien round sitting in the middle of a building roof not under cover was not a good idea lmao.

Wonder if there will be a sale anytime soon on it.

 

[thespyder]

there's a lot of good advise in the post above. I'd add that, although the tutorial is annoying to play, it does cover some of the basics like squad positioning, cover and the use of the individual classes. Once you have played through it, you may decide to go back and start again, but that isn't 100% necessary.

Playing the game at higher levels meaning balancing the costs on base building with the needs of your soldiers.

Spoiler

Satellites are very important to keeping Panic down.

And absolutely don't chince on outfitting your soldiers. Plus, make sure that you have several backups as life is fragile as an Xcom Soldier.

I didn't worry to much about changing the names of my soldiers. I actually liked the "Nick Name" feature of the game and found that I often referred to my favorites by those names instead. Cobra was my Psi-Ops Assassin and Deadbolt was my Sniper. And Tectonic was my Heavy for bringing on the Pain. So even if you can't customize the soldiers that much, they still remain memorable (to me at least).

Spoiler

I do also agree that you stay away from the ****PRIORITY***** missions until you are fully prepared. Doing so doesn't have any kind of an impact (though it should) and they are always things that move you to the next "Chapter" of the game. I think that could have been made more clear in the game. But there it is.

Anyway, yeah. there is a fair amount of depth to the game. And it is horribly addicting.

 

[Howard]

Don't build the Hyperwave Decoder until you have Firestorms. Do research the Hyperwave tech ASAP since it unlocks Ghost Armor.

 

[Ashenor]

Is it better to go firepower or armor? From the demo it seemed you could build either i went firearms/firepower first.

 

[thespyder]

Also, the Ark-Thrower is a PITA, but it is vital. Get used to using it. And get used to losing soldiers who use it unsuccessfully.

 

[Madia]

Thanks for the info. Game does look fun. In the demo my sniper died first alien round sitting in the middle of a building roof not under cover was not a good idea lmao.

Wonder if there will be a sale anytime soon on it.

I forgot Rule #1: Always stay in cover! That's the most important thing to learn in the field.

Green man gaming has it for $45 with either $5 back or $10 in credit. That's probably the cheapest you'll find it for a while.

 

[thespyder]

Is it better to go firepower or armor? From the demo it seemed you could build either i went firearms/firepower first.

it really depends on your approach to combat. And the level you are playing. On Normal, cover does a decent amount of protection. And combine that with Nano-Vests and that will do you for armor in the early game. Later on, you need the armor, But early on, you are probably better off by keeping your distance and taking out your opponents as quickly as possible.

However, on Classic and higher, I understand that cover doesn't help that much. And then you really need more help in staying alive.

In the end, I would say that i would put a tiny bit more focus on weapons than on armor. There are more levels of weapons. And Laser and Plasma weapons are longer range than normal weapons. So....

"Go not to the Elves for advice for they will say both no and yes."

 

[Martimus]

I say research the first armor level, then research laser weapons. The first armor is extremely useful, because it keeps you from being one shotted from all lower level aliens. Laser weapons are quite useful as the initial weapons are pretty crappy. But between the two, I would say armor is more important in the beginning, since you can always shoot that alien with one more gun, but getting shot and killed has no low tech solution.

 

[acheron]

Anyone?

You can turn on text. I had trouble hearing things at first too and then turned on all the subtitle options.

Doesn't help you now, I assume you want to read a transcript without replaying the level, but for next time.

 

[kevinsbane]

Is it better to go firepower or armor? From the demo it seemed you could build either i went firearms/firepower first.

If you're doing Ironman... I'd do armour first. Vets with guns > rookies with lasers.

 

[Ashenor]

Doubt i will be doing Ironman on first play of it

 

[Martimus]

Ironman is best on your first playthrough. It really makes the game more immersive and increases tension since you really don't know what's coming. After that, it is less important, in my opinion.

 

[zanemoseley]

Just got my first major glitch. I actually think I'm on the final ship (final assault) and I just cleared the room with two sectiods and I get my guys in the one hall exiting and its all black. I finally pan and zoom enough to get it to show the door but can't open it. Most of the time all it shows is black and no door/frame. I got pissed and closed it, guess I'll try again later.

 

[Madia]

Just got my first major glitch. I actually think I'm on the final ship (final assault) and I just cleared the room with two sectiods and I get my guys in the one hall exiting and its all black. I finally pan and zoom enough to get it to show the door but can't open it. Most of the time all it shows is black and no door/frame. I got pissed and closed it, guess I'll try again later.

The same thing happened to me on my first try. Did you take along a shiv? That was what glitched it for me.

 

[Howard]

Just got my first major glitch. I actually think I'm on the final ship (final assault) and I just cleared the room with two sectiods and I get my guys in the one hall exiting and its all black. I finally pan and zoom enough to get it to show the door but can't open it. Most of the time all it shows is black and no door/frame. I got pissed and closed it, guess I'll try again later.

Can you walk through it?

Also, you can't move on the tiles with Cyberdisc wrecks for 2-3 turns after killing the CD.

 

[thespyder]

Just got my first major glitch. I actually think I'm on the final ship (final assault) and I just cleared the room with two sectiods and I get my guys in the one hall exiting and its all black. I finally pan and zoom enough to get it to show the door but can't open it. Most of the time all it shows is black and no door/frame. I got pissed and closed it, guess I'll try again later.

That's your first major glitch? I have seen a couple.

To date I have the one where, on a triple decker ship, the ceiling keeps on flipping on when it shouldn't. And consequently the program doesn't seem to understand that I want to move the soldier across that same level, not to the upper or lower one.

I have encountered where either the program will move to the next turn even before all of my soldiers have taken their full action, or sometimes it will give my soldiers two moves with no Alien move in-between.

I have encountered where, after my sniper fires In the Zone, that taking an action, I can't end her turn with overwatch.

There is the annoying stuff where the percent above the red (or yellow) alien heads in the lower right hand corner don't register properly. And if you get a research completion while you are scanning, when it goes back to scanning the "Back" button no longer shows.

there may be one or two others. And I would dearly like to see the 'To Hit' rolls because I am not convinced that they are registering in accord with the listed percentages. I have seen 3 misses in a row when my soldier had greater than a 75% chance to hit. And I have seen Entire squads miss reaction shots on overwatch despite distance or cover. These are "Possible" but not very likely events. And they happen too often to be reasonable for my taste.

Not that it isn't a great game. Just saying there are some fixes that I hope get put in place.

 

[zanemoseley]

Not using a SHIV, actually I haven't even messed with the SHIV's, I'm probably missing out.

I've see a lot of glitches but none that I would consider "major" which result in not being able to finish playing a mission. The messed up levels glitch is the worst especially when you end up moving someone to a spot that is far away from where you intended them to. This alone makes me not want to play iron man mode, the game is hard enough without it screwing you in the process by dropping you in the middle of a bunch of aliens with no turns left.

 

[Howard]

there may be one or two others. And I would dearly like to see the 'To Hit' rolls because I am not convinced that they are registering in accord with the listed percentages. I have seen 3 misses in a row when my soldier had greater than a 75% chance to hit. And I have seen Entire squads miss reaction shots on overwatch despite distance or cover. These are "Possible" but not very likely events. And they happen too often to be reasonable for my taste.

Not that it isn't a great game. Just saying there are some fixes that I hope get put in place.

I find overwatch % to be fairly true to form. My sniper almost always hits them while the others are so-so. Remember that overwatch gets an accuracy penalty compared to regular shooting and that overwatch is almost always at maximum range.

 

[thespyder]

I find overwatch % to be fairly true to form. My sniper almost always hits them while the others are so-so. Remember that overwatch gets an accuracy penalty compared to regular shooting and that overwatch is almost always at maximum range.

Conversely, my sniper (who has an accuracy of 105 AND a scope) rarely hits on reaction shot. She is absolutely deadly in almost any other situation, rarely missing a shot. But none of the rest of my guys can shoot for crap during reaction either.

But it is more than that, as stated, missing 3 consecutive shots with more than a 75% chance to hit is a little odd. And on another scenario, I had a 98% chance to hit and missed. Again, statistically possible. But those types of things happen more often than I think are reasonable in game. Plus the insane number of times I have hit with 10% chance. And since the percent to hit on the lower right hand corner is borked, I just wonder what other 'To Hit" percentages are working as advertised.

So I would like to either see the rolls, or have someone examine the 'To Hit' mechanics more closely.

 

[kevinsbane]

Conversely, my sniper (who has an accuracy of 105 AND a scope) rarely hits on reaction shot. She is absolutely deadly in almost any other situation, rarely missing a shot. But none of the rest of my guys can shoot for crap during reaction either.

But it is more than that, as stated, missing 3 consecutive shots with more than a 75% chance to hit is a little odd. And on another scenario, I had a 98% chance to hit and missed. Again, statistically possible. But those types of things happen more often than I think are reasonable in game. Plus the insane number of times I have hit with 10% chance. And since the percent to hit on the lower right hand corner is borked, I just wonder what other 'To Hit" percentages are working as advertised.

So I would like to either see the rolls, or have someone examine the 'To Hit' mechanics more closely.

Classic and Ironman, the percentages are pretty accurate over time. Apparently all 95-99.6% shots are treated as 95% shots. 100% shots are actually 100% shots. You get varying hidden bonuses in normal and easy to to-hit percentages, based on number of missed shots (missing shots in a row gets you a hidden accuracy bonus), number of squad members alive (each member <4 gets you an accuracy bonus). Aliens also have a hidden penalty to hit you with squad members < 4.

Think of it this way... missing 3 75% shots in a row is about 1 chance in 4x4x4 = 64. The odds are actaully very reasonable that you might miss 3 75% shots in a row - only 1 in 64. Even missing 2 95% shots in a row is only 1 in 400 - unlikely, but very possible. Even 3 95% shots in a row is only 1 in 8000 - you will have killed hundreds, if not thousands of aliens in a single campaign, and taken many thousands of shots.

There was an analysis of the prng used in XCOM - it's a decent, though not perfect one. It suffices for XCOM. Aside from the hidden bonuses above, any perceived deviation from the accuracy percentages is likely due to observer bias and/or individual variation.

 

[thespyder]

Think of it this way... missing 3 75% shots in a row is about 1 chance in 4x4x4 = 64. The odds are actaully very reasonable that you might miss 3 75% shots in a row - only 1 in 64. Even missing 2 95% shots in a row is only 1 in 400 - unlikely, but very possible. Even 3 95% shots in a row is only 1 in 8000 - you will have killed hundreds, if not thousands of aliens in a single campaign, and taken many thousands of shots.

There was an analysis of the prng used in XCOM - it's a decent, though not perfect one. It suffices for XCOM. Aside from the hidden bonuses above, any perceived deviation from the accuracy percentages is likely due to observer bias and/or individual variation.

I guess maybe I didn't make my thoughts clear. Yes, the individual instances described (3 shots at 75% being a miss or 4 hits at 10% in a row etc....) are statistically possible. My point was, that while these are statistically possible, the frequency that I have encountered them in game seem suspect at best. yes, I am playing "Normal" and no, I didn't know about the unseen adjustments, so maybe that is it. But it still seems a bit higher than I would have thought unless the modifiers are significantly higher than you indicate.

I used to play D&D way back when stats were generated randomly (3D6). One guy joined our group and wanted to include his Paladin who had all 18 stats. he claimed that he had randomly rolled the character. Then, when we doubted him, he proved it. He rolled something like Fifty-two thousand (this is a total guess, but it was a whole WACKING lot) characters before he came up with that one. He had written a program to do it. it was POSSIBLE, just highly unlikely.

And yes, I am aware of the number of times my soldiers fire. And the number of seemingly 'Normal' hits and misses. Plus the standard deviation of outliers. I am just saying that the number of outliers in the game I have played seems inordinately higher than the average.

I'd like to see the analysis you indicate. Do you have a link?

 

[kevinsbane]

I guess maybe I didn't make my thoughts clear. Yes, the individual instances described (3 shots at 75% being a miss or 4 hits at 10% in a row etc....) are statistically possible. My point was, that while these are statistically possible, the frequency that I have encountered them in game seem suspect at best. yes, I am playing "Normal" and no, I didn't know about the unseen adjustments, so maybe that is it. But it still seems a bit higher than I would have thought unless the modifiers are significantly higher than you indicate.

I used to play D&D way back when stats were generated randomly (3D6). One guy joined our group and wanted to include his Paladin who had all 18 stats. he claimed that he had randomly rolled the character. Then, when we doubted him, he proved it. He rolled something like Fifty-two thousand (this is a total guess, but it was a whole WACKING lot) characters before he came up with that one. He had written a program to do it. it was POSSIBLE, just highly unlikely.

And yes, I am aware of the number of times my soldiers fire. And the number of seemingly 'Normal' hits and misses. Plus the standard deviation of outliers. I am just saying that the number of outliers in the game I have played seems inordinately higher than the average.

I'd like to see the analysis you indicate. Do you have a link?

See this.

Conclusions

The tactical portion of the game does what is expected, correctly. The method the game uses for seed generation is fine. The underlying PRNG is not great and could be replaced, but the deficiencies in the algorithm should have minimal impact on game play. When you miss 3 90% shots in a row, you just got unlucky, the game is not cheating you out of your kills.
The (rather messy) C source code that I used to generate the data for the graphs is available on request.

I'm still going to go with observer bias + individual variation. Humans being humans, we look for patterns even when patterns don't exist.

 

[thespyder]

See this.

Thanks. I will read through it when I have a chance. I am always fascinated by the mechanics of any game. I guess that is why I am a programmer in real life (not games though).

I'm still going to go with observer bias + individual variation. Humans being humans, we look for patterns even when patterns don't exist.

Very possibly. I am in no way saying that my observations are not biased and subject to individual perception. I am also very much enjoying the game. I was merely making note of my observations. And in light of the disparities with the Percent to hit reporting issue with the Red/Yellow Alien Heads in the lower right hand corner, I admit that I was skeptical. But am happily prepared to be wrong.

 

[Daverino]

I think it has more to do with selective memory and the frequency of choices.

If you have a 90% shot, you will take it 100% of the time. You will be surprised 10% of the time and that surprise will remain with you.

If you have a 10% shot you will take it (close to) 0% of the time. You are never surprised when it does hit because you never tried.

Therefore we're more likely to experience events where we have a high probability of success and fail rather than a low probability of success and succeed. This skews are perception of the likelihood of all events and hence the 'cheating' feeling.

 

[thespyder]

I think it has more to do with selective memory and the frequency of choices.

If you have a 90% shot, you will take it 100% of the time. You will be surprised 10% of the time and that surprise will remain with you.

If you have a 10% shot you will take it (close to) 0% of the time. You are never surprised when it does hit because you never tried.

Therefore we're more likely to experience events where we have a high probability of success and fail rather than a low probability of success and succeed. This skews are perception of the likelihood of all events and hence the 'cheating' feeling.

I could take the reverse tack and claim that you both are 'Drinking the Kool-aid' far to readily in just assuming that the machine is always right and that the human is in error.

At the end of the day, it's still a fun game. And although I don't believe that selective memory colors my judgement to the degree you are both claiming, I am willing to accept the possibility. or at least suspend my disbelief while enjoying myself.