These turn times are killer

[Ruanek]

I will personally guarantee that you will find more of an increase in your game if you get a solid state drive and install it on it than you will by using 8 gigs of ram.

What if I only have 1 GB of RAM?

In any case while I'm not an expert on RAM usage in games I would think that generally more RAM helps if a game is running slowly and a computer doesn't have a large amount of RAM to begin with. Especially considering how much of the RAM could be taken up by the OS, other programs, etc.

I would love a solid state drive, though. I'll probably just wait a few years for the prices to go down first.

 

[unmerged(402700)]

They already have, they aren't decreasing anymore, it's still a new gadget.

 

[Ruanek]

They already have, they aren't decreasing anymore, it's still a new gadget.

The prices will keep decreasing. Just look at how cheap terabyte hard drives are now compared to even 5 years ago. Or most things in the tech field, really. The fact that it's new just means that it has farther down to go.

 

[unmerged(402700)]

That's a very different range of technology though, this is the concept of no moving parts, it's a bit more expensive.

 

[Alucius]

Terrabyte drives are actualy quite expensive now due to the floods in Thailand. What you get for putting most of the factories at the same place. When it comes to SSDs and gaming. Sure you get a noticiable improvement by using an SSD, mostly in loading times. But that depends on how the gamecontent is stored. if its compressed then it has to be decompressed by your CPU while it loads so that will bottleneck the improvement. You wont see much of a FPS type of improvement as one myth says.

In some games this can be highly annoying as it removes immersion. For example in Mass Effect 2 while moving between floors in the starship you see the elevator movement as loading screen. However when running it on my SSD only system loading transition times are instant so you dont see the elevator movement picture with the map of the starship. Guess thats an example of hardware improving to a point not imagined by the game designers

RAM is cheap nowadays. So its never a waste to get more ram. I use a page file of 200 meg on my gaming rig for example (for those rare apps that require one) as i have enough ram not to need to burden my SSD disks with a page file.

 

[unmerged(402320)]

I have a I7 too running at 4ghz and 16gig of ram and with 3 AI's each turn takes 3-4 minutes, on larger maps, and about 2 on others, and menu changes can take just as long.

ALOT will have to be done both with the menu's and how stuff is done in the game perhaps with simultanious turns with most of the AI stuff for the next turn being done while the player is doing his, and I think it's needs multi-threading too, because there's no hope in hell of this ever being able to run 7 AI's as is without the player smashing his display up in frustration.

The games running like it was written in basic

 

[Ordo Impavidus]

I will personally guarantee that you will find more of an increase in your game if you get a solid state drive and install it on it than you will by using 8 gigs of ram.

Yes of course.

But I won't spend ~800.00 USD to improve this lappy in particular. :mellow:

I'd rather purchase a SSD (or 2) for a true gaming lappy....like for the ASUS G74 with Quad CPU and a dedicated 3GB DDR5 Vram!!

 

[Ruanek]

That's a very different range of technology though, this is the concept of no moving parts, it's a bit more expensive.

I never said it wasn't more expensive. Only that it will get cheaper in a few years (and probably by a significant margin, if other related tech products are anything to go by).

 

[eRe4s3r]

...it was adressed in the upcoming patch according to the official forums

Actually all they did was move the processing in the background , which in many ways is even worse and not "addressing the issue" at all. Though they removed the loading screens and probably optimized the processing a bit, in the end these processing times are already bad in turn 1 (with lots of random encounters) and are nightmare'ish indeed in turn 250, and by turn 250 you are often not even half-way through a game.

This simply needs a fix beyond "hiding it"

 

[Bacus]

Actually all they did was move the processing in the background , which in many ways is even worse and not "addressing the issue" at all. Though they removed the loading screens and probably optimized the processing a bit, in the end these processing times are already bad in turn 1 (with lots of random encounters) and are nightmare'ish indeed in turn 250, and by turn 250 you are often not even half-way through a game.

This simply needs a fix beyond "hiding it"

we have a winner!!!
yayyyy!!!
hope Mr. Kerberos reads this

 

[s1234567890m]

Actually all they did was move the processing in the background , which in many ways is even worse and not "addressing the issue" at all. Though they removed the loading screens and probably optimized the processing a bit, in the end these processing times are already bad in turn 1 (with lots of random encounters) and are nightmare'ish indeed in turn 250, and by turn 250 you are often not even half-way through a game.

This simply needs a fix beyond "hiding it"

source?

 

[ETF]

Just wow. I mean, I'm playing with a single AI player and turn processing is taking upwards of 30 seconds and I'm running an i7-950 at 4.06 Ghz with 6 GB of memory.

I don't know what the hell is going on but it needs to be fixed in the Wednesday patch. There's something not quite right there for it take that long. This game isn't terribly complex when you break it down, so it shouldn't require near that much time nor that much horsepower to process a single turn. I haven't even made it past turn 200 because I get so frustrated with it.

Anyone else encountering this? I imagine it's only worse on lower end rigs.

Try Pride of Nations.........3-5 minutes. In our LAN game we turn on the DVR to watch a movie in between turns

 

[eRe4s3r]

source?

I supposed you are joking

Before patch end-turn was instant + the combat calculations grew exponentially
After patch end-turn has a exponentially growing delay

So may source is "obvious" correlation

Only possibly answer for that is that "on end-turn" it calculates all combats, slightly faster (well, no loading screens ,p) but still as annoying as ever

 

[Tairneanach]

The game does have to calculate combat some time, though, and do it in a meaningful manner. In SotS Prime, certain factors like distance were simply ignored when combat was autoresolved, leading to situations where it was better to autoresolve - like raiders attacking CR freighters near a planet, where a fleet could intercept. If you fought that combat manually, especially early in the game, the freighter was lost, as the raiders deployed between it and the planet with the other forces. With autoresolve, the freighter usually survived.

Also, point defence wasn't factored in correctly.

So, in the end, there have to be some concessions.

 

[s1234567890m]

I supposed you are joking

Before patch end-turn was instant + the combat calculations grew exponentially
After patch end-turn has a exponentially growing delay

So may source is "obvious" correlation

Only possibly answer for that is that "on end-turn" it calculates all combats, slightly faster (well, no loading screens ,p) but still as annoying as ever

I thought the patch hadnt gone live though steam yet, :S

 

[eRe4s3r]

Unless i missed what patch we are talking all patches with a changelog listed in the forums here are active on steam. Unless you talk about the one coming out today sometime.. and who knows what that fixes.

In the end its going to come down to an option to abstract combat like it was in SOTS1 because at some point in SOTS2, the turn times will become excessive. And I am used to "excessive" - the problem is that the only game where i can stand excessive turn times particularly when you often just wait for something to happen.. is Shogun 2.

And SOTS2 does not even come close to the complexity of Shogun 2's turns. (Where you can often spend upwards of an hour on -1- turn)

Also it isn't rocket science to make an abstract calculation that is 99% accurate to a full simulation. So what, distance has to be taken account for each ship, but the fact is that it has to be abstracted. Kerberus just needs to do it properly, as I fully agree. And it was NOT properly in SOTS1 (in fact, that was downright lazy broken and glaring bugs have never been fixed like the PD and distance calculations......)

 

[Alucius]

it hasnt.

 

[unmerged(400937)]

Also it isn't rocket science to make an abstract calculation that is 99% accurate to a full simulation.

nope; it's maths.

a simulation that is 99% accurate essentially requires 99% of the processing power, ie, no noticable difference. you save some on outputting graphics, but it will take even MORE processing power if you want to do it FASTER than the real combat. the real combat is already a mathematical simulation, you see.

you know what the biggest problem with SOTS 1's autoresolve was? projectiles. try a stormer in autoresolve and in manual sometime. want to correct that bug? simulate every projectile in the same way as you do in real time. anything less and you'll still have inaccuracies.

 

[Tairneanach]

Also it isn't rocket science to make an abstract calculation that is 99% accurate to a full simulation. So what, distance has to be taken account for each ship, but the fact is that it has to be abstracted. Kerberus just needs to do it properly, as I fully agree. And it was NOT properly in SOTS1 (in fact, that was downright lazy broken and glaring bugs have never been fixed like the PD and distance calculations......)

Well, I'm no expert, but I don't know how one would abstract the whole thing without simulating it.

Actually, Anaris has said what I wanted to say, only he done better.

 

[eRe4s3r]

Actually abstraction means a close approximation to a full simulation. So inaccuracies are not just side effects, but the whole point. Stormers are a problem but the distance thing can be easily fixed as well as damage over time stuff...

Personally i'd rather take inaccuracies than a 30m turn - because i am not gonna wait for that turn to compute, i am going to play something else, which is by far the best abstraction ;P

Also, anaris, math would tell you that the last 1% is by far the greatest problem when comes to "accuracy" of anything ^^