Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[[NO_THOUGHT]]

I follow CES, and as soon as I saw this I knew that one day we shall all be free of FPS limits (and on tablets to boot!)

http://www.engadget.com/2012/01/09/onlive-launches-desktop-app-streams-start-menus-and-office/#continued

touch might be difficult to use but a bluetooth keyboard and mouse would fix everything

[peskyninja]

Infinite FPS is impossible as computing power, size and other stuff are impossible too.

[Sheb]

You probably don't want to run at more than a few hundreds FPS anyway. But running a huge fort without lag would be awesome.

[[NO_THOUGHT]]

This is Onlive we're talking about, they have massive servers running all of these simulations simultaneously. Infinite is an impossible dream but you could get way wayyy up there and not have to worry about pathing or cats or HFS or magma floods or elephants. Seriously, it would be fantastic!

But running a huge fort without lag would be awesome.

Dwarf fortress could probably give us a standard by which we can tell how good the onlive servers are...

[Jake]

I'll be hedgehogged if I'll play DF on a tablet even if I wanted one, and I'm old-fashioned enough to think that for upwards of two hundred quid I'd rather have a couple of USB ports and my own choice of OS. If the basic free version supports running uploaded .exe files I'd be very interested to see how it performs versus DFterm on a good gaming rig, though.

[Kogut]

This is Onlive we're talking about, they have massive servers running all of these simulations simultaneously.

DF is NOT multithreaded.

[Telgin]

This is Onlive we're talking about, they have massive servers running all of these simulations simultaneously.

DF is NOT multithreaded.

^^This.  Running DF on a super computer is likely to provide no benefit in the best case and a detriment in the worst as the architectures aren't designed to run this sort of application efficiently.

So distributed computing isn't going to help DF.  Not for a long time yet, as I assume Toady will eventually multithread parts of the game.  Eventually.

[Jake]

Running DF on a super computer is likely to provide no benefit in the best case and a detriment in the worst as the architectures aren't designed to run this sort of application efficiently.

So distributed computing isn't going to help DF.  Not for a long time yet, as I assume Toady will eventually multithread parts of the game.  Eventually.

I'd be surprised if it didn't run DF a bit better than my little Intel Atom-powered netbook with 1GB of RAM, though.

[Telgin]

Perhaps, or perhaps not, depending on the exact architecture used.

Some super computer platforms use many thousands of very weak individual processor cores.  Each would perform very poorly on a single threaded application like DF.  Imagine a single stream processor on a GPU, for example.  Only a few hundred MGhz clock and very limited in how it handles general purpose code, but when you string a few thousand together you get impressive results.  DF wastes that.

Other platforms would indeed use more powerful general purpose processors that would probably out perform an Atom, but it's still a criminal waste of resources to try and run a primarily single threaded application on pretty much any modern super computer or distributed computing network. 

[Ripolix]

Hahahaha

OnLive's inout lag is absolutely terrible. I can't imagine myself navigating through all the menus like this. It'll be close to being as bad as playing FPS on Onlive

[Kogut]

Not for a long time yet, as I assume Toady will eventually multithread parts of the game.  Eventually.

"It's complicated, and from what I gather it would be a really difficult long project and it's probably not going to happen."

[No1]

The problem with DF is that it's turned based, each creature moves when it's its turn. The only thing that could potentially be done on more then one thread would be path finding for one specific creature, it could use thread pools to dump each expanded node to do its search in its own direction. The problem is that it needs to be done thread safe, any world data etc needs to be accessible by more then one thread at any time. I've not done it in C or C++ so I do not know the level of programming required but I could imagine that it's not that easy.

[Telgin]

Not for a long time yet, as I assume Toady will eventually multithread parts of the game.  Eventually.

"It's complicated, and from what I gather it would be a really difficult long project and it's probably not going to happen."

This is a statement from Toady himself?

While I don't for a second deny the claim that it's a long and difficult project, I think Toady understands that it's probably an unavoidable necessity and will probably get around to it eventually, in some form.  The game's processor demands are probably going to scale faster than single threaded processor speed.  Actually, can anyone comment on how much slower the current version runs than 40d?  Any slower?

I know there are a thousand intricacies I can't appreciate without looking at the code, but the update step has to be parallelizable in some fashion.  Whether it's worth spending time instead of adding features, well, that's Toady's call.  Then there's the whole issue of whether or not to do it now and save difficulty later, or put it off...

[[NO_THOUGHT]]

DOH! 
I keep forgetting that DF isn't multithreaded, but I'm sure that Onlive has higher end servers for higher end tasks...

On topic (of where this thread went to) from what I've gathered from programmers is that multithreading is a completly different mindset than regular programming, it's like trying to solve a 4x4x4 rubix cube with only 3x3x3 methods if you don't think about it right. Tarn is eventually going to be forced to do something about lag, but I don't imagine that would be for a while yet. As an example though, the new random weather effects in evil regions, those are amazing but will require weather turned ON no? Which means that one way of maxing fps with minimal sacrifice is gone, and those weather effects might even make their way to neutral or even good biomes. The amount of processing power for those is minimal, but effects stack.

There will come a time when adding fps gains will outweigh adding features... but who's to know when that time will come?

[Telgin]

On topic (of where this thread went to) from what I've gathered from programmers is that multithreading is a completly different mindset than regular programming

Yes and no.

The programming itself is still mostly the same, it's just you have to decompose your program into sections that can be run at the same time without interfering with each other.  Then you have to make sure that they don't interfere with each other.

It sounds easy.  It isn't.

That's why I have no doubt Toady has been putting it off for as long as possible.  I agree though that he's almost certainly going to have to do so at some point.  Single threaded performance continues to improve, but the trend is slowing and if DF continues to demand more from computers (which I have no doubt it will), then it's about the only solution.

Really, I think DF is a testament to how a well written and optimized program can perform amazingly well despite its tremendous complexity.  Just think about all of the stuff the game has to keep track of and marvel at how it could ever achieve >100FPS.  I know I do.

But optimization can only get you so far.