4 ms·
The context is that the author wrote a negative review of a space strategy game because he had poor experiences with it. This article seems to be a way of sayin
by 1123581321 10y ago
The context is that the author wrote a negative review of a space strategy game because he had poor experiences with it. This article seems to be a way of saying that he was wrong to say the game is objectively bad without denying his own bad experiences are an objective problem.
The solution is to tune the game and add additional mid-game content (the part of the game with the most room for dynamism), which the author points out is on its way.
One of the more interesting comments said that the game would be well served if its developers collected play through data and started prioritizing the game seeds that tended to cause the most interesting games, arguing that there would be enough seeds to hide the lack of randomness.
- pjc50 10y agoThis article seems to be a way of saying that he was wrong to say the game is objectively bad Where did he say that? People seem to keep mistaking reviews for impossibly objective analyses of games.
- wastedhours 10y ago"Seems" - it's not explicit, but the implication is that objectively saying a randomised game is bad when it's only been bad from a subjective perspective isn't the best approach.
- pjc50 10y agoBut I'm not seeing where the reviewer made the claim that the game was objectively bad?
- wastedhours 10y agoIt's in the linked review[0] that he wrote, and the subsequent follow up by the developer[1]. [0] http://www.ign.com/articles/2016/05/09/stellaris-review http://www.ign.com/articles/2016/05/09/stellaris-review [1] http://www.twitlonger.com/show/n_1solvon http://www.twitlonger.com/show/n_1solvon
- jsnell 10y agoIt'd be really strange for any publication to claim their reviews are "objective" (there's objectivegamereviews.com, but it is intended as humor). The usual view is something along the lines of a review being just an indication of one person's experience with a game at a specific time. And as far as I can see, the latter is the case here as well. Neither page you link to contains the word "objective". The only mention of the words "bad" is in the phrase "bad luck", in a sentence specifically talking about how other people might have a better experience and Rowan might have been just unlucky. There is a score, but to interpret that you'd look at the review guidelines for IGN. To quote: Under no circumstances are review scores influenced by anything other than _our own opinions_ on the quality of the product in question. (Emphasis mine).
- pjc50 10y agoNeither of which contain the word "objective", and the second one says "lower than liked score" and "opinion".
- fao_ 10y agoEven better, they might be able to find a pattern in the seeds that are more interesting (They could be prime, or a power of two), and generate or manipulate seeds to conform to that pattern.
- vertex-four 10y agoAssuming that the seeds are actually seeds for any sort of half-working random number generator, and given the amount of different subsystems which will depend on that seed, that sounds incredibly unlikely.
- fao_ 10y agoI didn't mean that they would find a pattern in the random number generator, rather a pattern in the interesting seeds, that they can then abuse to make random seeds interesting. An example of this is that perhaps numbers that are even produce more interesting games than those that are odd, so they could then write something like the following: int new_seed(void) { int seed = random(); if ((seed % 2) != 0) seed++; return seed; } which makes seeds less random, but is more likely to produce interesting games.
- vertex-four 10y agoFinding an obvious pattern in the seeds implies that the random number generator's output has an obvious pattern. Given the point of RNGs, that seems unlikely.
- lmm 10y agoPreserving a seed like that wouldn't lead to a similar game without a lot more work. If there's one more or less combat, or you meet the other factions in a different order, you'd "desync" the stream of random numbers and have an entirely different sequence of random events later on.
- y7 10y agoAssuming the random events are also based on how the game progresses, and not just from the starting seed (if they were, then at the start of the game you would be able to predict all random events for the entire game to come, which sounds implausible for a complex strategy game), then the seed won't help you much in predicting game events. Pseudorandom number generators are chaotic, otherwise they wouldn't be called pseudorandom. This means that if there is, say, an extra random number generated in the middle of your game, it would likely completely change the random events in the remainder of the game. What would help is, instead of prioritizing seeds, looking at optimal distributions of random events that lead to fun gameplay. This might mean for instance making your next random event dependent on previous ones (e.g. to compensate previous bad events with more fortunate ones, and vice versa).