Pathfinder
Member
Was bullshit marketing speak then, is bullshit marketing speak now.
Posts like these will be fun to look back on in a couple years.
Was bullshit marketing speak then, is bullshit marketing speak now.
Lets see what crackdown can do before throwing it all out as BS marketing speak.
Posts like these will be fun to look back on in a couple years.
Posts like these will be fun to look back on in a couple years.
The past year has affirmed all the things the skeptics were saying. So you're betting the coming 2 years will change that? What has changed? Why would the next 2 be different from the last year?
In multi-player it exists because the time for server to player is better than player to player and for syncronization. It would really depend on what AI because a optimistic 30ms ping means you are 1-2 frames behind. At 100ms you are 3-7 frames behind and that can be significant. Doing it for a single player game seems less sensible. You're accepting more lag for 'better' AI. It just doesn't seem like a better solution.
Any sort of cloud calculated 'effect' would need something like that to hide the transaction. Like a huge dust cloud then playing the calculated animation. But couldn't you just can 7 animations and run one at random? You could even can it based on a few variables. You wouldn't have to worry about the latency of the transaction exceeding the 'hiding' effects runtime.
The key though, is one is just piping up controller input piping down video directly. A easier conceptual problem than finding a computing problem that can be broken up in a way which is agreeable to small outgoing pipes, latency, network traffic variability, and peak load. The set of problems that are agreeable are small and not super impressive. So the results will be small and not super impressive.
If cloud computing will be as important and brilliant as we are told I pity the people with poor internet connection. There should be a warning on the boxes.
From what I can tell it was ninety percent bullshit from the start. A bunch of knowledgeable people here and across the internet gave what i thought were good explanations for why most of what Microsoft was promising was bs, but people kept believing the MS hype machine.
Posts like these will be fun to look back on in a couple years.
Hey, you were saying it was BS that they "could" do these things.
Lets see what crackdown can do before throwing it all out as BS marketing speak.
because we have an actual game being developed around it. let's wait for Crackdown.
Why? You enjoy confirming that other peoples statements are correct post tense or something? Weird hobby, but whatever floats your boat, I guess.
the new 'have you seen titanfall'?...
I'd be very interested to know what sort of response times you think typically exist in games today, to create AI that you don't scream "that's bullshit!" to. You think believable AI reacts to your input in 3-5 frames? That would come across as an instant "button reading" response in pretty much any game. A single light jab in Street Fighter is on screen for about 12 frames, how often do you see people react to that?
AI would have PLENTY of time online to process the situation, and respond long before even the fastest of comparable humans could.
It wouldn't need anything to hide it. Why would that be the case? The calculation would be completely invisible to you. Your screen only draws the result. It's no different from how the same sort of scenario would happen locally, only that the player would need to not be essentially kissing the target (it's gonna take a bit of time to get the result ready).
The reason you may not want to precompute thisis simply that it may just have too many different variables. Maybe the structure is player created in the first place. Maybe after shooting it once, you can shoot what remains again, thus multiplying all the scenarios you need to precompute. Maybe there's 100 different structures. Maybe one structure can collapse onto another. Etc.
Yea, they're different solutions to different problems. The PSNow stuff is a lot easier to conceptualise, because it's simply "what we do offline, but online". I don't think games should be designed with cloud utilisation as a targeted bullet point (and I'm worried about Crackdown for this reason), but there are probably some very interesting uses waiting to be discovered. My point in that paragraph though, is that you can't compare latency and bandwidth requirements to something like PSNow, as what needs to be transferred for a cloud enhanced game is likely to be much smaller than a constant 720p video stream, what needs to be updated will almost certainly be less time critical than character movement, and what needs to be processed won't have the constant overhead of running an entire game at all times. It's completely incomparable.
Collectively it's been so laughed off for gaming that I'm back to being annoyed with it as a business buzzword. None of this is magical.
I love how almost precisely 2 years ago at the XB1 reveal that everyone said "just wait a few years!". And here we are again. Can we agree to just check back in another 2? Let's talk about this again when your average internet user has access to a gigabit connection and a tiny ping. Wondrous things can surely be done if you don't have to limit your data and your latency is only equivalent to a frame or two, and there's so much free processing time around the globe that it can all be done on demand.
I'm sure it has uses for multiplayer titles but it's...
mostly just marketing fluff.
Do you just hate buzzwords in general, or do you simply demand that they only be used to describe things that are actually magical?
Buzzwords are typically used so that end users without a proper understanding or familiarity of a concept, can still differentiate it from other similar concepts. Uploading something to a cloud such as Azure, isn't the same as uploading something to a privately rented dedicated server. The "where" is largely unimportant for cloud implementations, but very important for standard dedicated server solutions. Would you prefer that we just say "servers" to describe all different types, and then explain the differences between with a detailed paragraph each time, rather than simply use the term "cloud" which immediately makes some sense (although not necessarily a lot, based on many of the responses in threads like this) to most people?
the new 'have you seen titanfall'?...
What?I saw what crackdown 2 could do ....... about as much as a cloud but twice as ugly.
100ms vs 30ms delay has a perceivable impact on game play for SF4 to SC2 to LoL.
As for SF, you do know people react to jabs. so all the time? Right, I'm not on crazy pills here? being half of a jab behind would have game play consequences?
The key benefit is that unlike canned animation; a cloud one could render something that accounts for player variables. But you have a lag sending the data up and receiving the data. You would absolutely need to hide that latency with something. If the player isn't interacting with it; then why not use canned animation?
Plenty of games use rough approximations calculate locally and have done fine. Halo, Crackdown, GTA, etc... How would better accuracy be worth the problems of inherent in splitting it up this way?
The comparison is that they both limited by latency. You are wrong in your assumption that time isn't an issue with graphics, AI or physics. In most scenarios time is a huge issue, all of it is tied to rendering.
Otherwise why not just wait for the CPU to take more time to make that calc?
Fine to be skeptical, and all (as we all should be), but it's funny how many minds are already made up on this stuff.
To me, at least, it seems a sensible person would first see what they have to show off after a few years of work with their "could" stuff before just assuming it's all complete and utter BS. I'm not saying you have to buy in to any PR/marketing speak, but you also don't have to assume everything is a lie. Maybe that's just me, though.
RE: Crackdown being mentioned...Seems to make plenty of sense why. Crackdown is probably going to be a the best example for MS of the type of stuff they think their cloud infrastructure will be able to help with. And based on their demo they showed off last year to the public that was early-crackdown footage, I'm intrigued to see what they'll come up with. I'm sure E3 will be interesting.
What?
It's managers vs implementer. When my supervisor says cloud. It's a magical thing somewhere doing stuff for us for free. When I say cloud it's a time slicse of a set of machines we rent which is worse than a dedicated machine but with less maintenance required by me.
I think that's where I fall on this topic. I'm seeing it for it's implantation warts and draw backs; and you're seeing it from the point of view of what it could do in ideal condition physics and logistics be damned.
You probably shouldn't assume where my point of view comes from. It's not like you know where I work, and with what technologies.
EDIT: And no it's not Microsoft, before anyone starts throwing out the "shill" accusations.
it's obviously cloudgine.
In the early day MS emphasized the cloud as something that could improve the gaming experience. It seems like this has yet to manifest itself.
Has MS given up on this approach or is it being used but not being talked about?
From what I can tell it was ninety percent bullshit from the start. A bunch of knowledgeable people here and across the internet gave what i thought were good explanations for why most of what Microsoft was promising was bs, but people kept believing the MS hype machine.
Yes it would have gameplay consequences.. but not for the reason you seem to think. It would have gameplay consequences before you would "perceive" the jab 12 frames later, which would then sit on top of your own reaction time, not because your reaction time is sub-12 frames. The AI doesn't need to give a shit about this, because it perceives your jab on frame 1. What typically happens in games like Street Fighter is that the AI just sits twiddling its thumbs until enough time has past that its response doesn't come across as cheating. It could be using all that time to think of horrible things to do to you instead.
As I said in my previous post, the calculation wouldn't have to start when the rocket impacts. It could start almost instantly after you pull the trigger, whilst the rocket is in transit, because the game know ahead of time what it's destination will be, when it'll get there, and anything that could possibly intersect with it. So you wouldn't have the rocket collide, and then start requesting a result. The result would have been sent to you whilst the rocket was in transit.
Depends. I can't really say that I've seen much in the way of dynamic world interaction in any of the listed games. It's not only the accuracy that can be changed, the scope and complexity can be dramatically changed to. Even for simple stuff like persistence, you could have the state of various areas in the game world stored in the cloud temporarily, preventing common situations like when you destroy some object in an open world game, return 3 minutes later, and it's all good as new, because the console needed that memory back for wherever else you where off visiting in that time. Things like this may actually make certain gameplay concepts viable, where they weren't before.
I didn't say time isn't an issue with graphics, AI or physics... especially graphics. What I am saying though is that some implementations of physics, and many many implementations of AI aren't very time sensitive. There's obviously still an operating window you need to adhere to (getting an AI response back after 45 seconds would be useless for most games), but these aren't typically single frame dependencies. There's a lot more time to work with before you get to the point where it'd be reasonable to expect a response from a human.
You probably shouldn't assume where my point of view comes from. It's not like you know where I work, and with what technologies.
EDIT: And no it's not Microsoft, before anyone starts throwing out the "shill" accusations.
Stuff doesn't occur in a vaccum. Lots of people know loads a stuff about servers, cloud, and the math of videogames.
You can make guesses at how things will go. Like when Sony came around and said 'hey, you know this Cell. It will fucking change the world.' Well we know what CPU can do. they describe to us what the Cell can do. We shrug and think well based on what we know the Cell probably won't change the world. Optimistically it might do some things alright enough to compete with the ever growing compute power of PC GPU's for a while. But we doubt it will change the fucking world.
For cloud. We can say it has some application but nothing like what they sold it as. Based on what we know.
In SF4 the AI reads your controller input because it had trouble keeping up otherwise. It's why it feels like it cheats compared to other players. To keep up with the better players it actually has to cheat and start it's decision making process earlier than another player would. It's why offloading the AI would make it terrible in a twitch game like that not better. You might have more compute power but you have lag between reading the game state which is significant.
IWhich boxes the potential in to scenarios of low player interaction. So why not just divide the building in 6 and where ever the rockets lands in plays the corresponding animation. 80% of the benifit, 2% of the work.
you're capped by bandwidth. How much of the game state can you send back and forth? You get texture/object pop right now due to the lag time between disk and memory. Imagine the problem of maintaining the game state online instead. Games like WoW that do this have lots of pop in. It's a acceptable trade off for a online game but is it for single player? Do you mask it with bloodborne style 45s loads?
That's what I am pointing out; all of the problems which this could apply to don't benefit that much from this solution but they make massive trades off for it. It's it more plausible that they'll just keep using 'server' power for traditionally 'server' things? Like syncronization, as you said persistence, and storage. All of which are seen everywhere on PC and isn't remarkable.
In the early day MS emphasized the cloud as something that could improve the gaming experience. It seems like this has yet to manifest itself.
Has MS given up on this approach or is it being used but not being talked about?
. . .The tradeoffs may not actually be massive though. . .
Posts like these will be fun to look back on in a couple years.
Pretty sure I was seeing posts like this a couple years ago too, but here we are, still seeing nothing of substance from the initial horse shit.
If Crackdown is shown off and doesn't have impressive physics that we haven't seen before then you can safely assume the cloud has dissolved.
Shhhhhhh... it's magic and we can't be convinced otherwise.
That's really the core of it. The trade offs are well known and the possible upsides less known; but so far tiny. This is the reason for all the cynicism.
I think they would make a much better case of azure leading to a more solid online experience than trying to shoe horn cloud compute in as a feature.
Just as kinect did something very well; but MS kept trying to mandate things it didn't do well.
Seriously, why would I put more faith in him than an entire corporation when they have a game in development? I know he knows a lot of technical stuff, but at this point, it's his word against a game in development. it would be dumb as fuck to take his word as faith right now. all we're saying is wait until Crackdown since it's confirmed to use it (as of right now of course), if it fails, then yeah, ya'll are right.
Crackdown is coming out after a significant amount of the One's lifespan has passed. And it's the only major use of it so far.
That means there is no chance Xbox One will be redefined on impressive new footing just because of the cloud. Not enough time even if Crackdown works well.
He is describing general principles familiar to me from work. They are hand having at something they are trying selling.
Like a stranger describing salfate based detergents vs a corporation extolling the virtues of pro vitimins. Just cuz they have money doesn't mean they aren't lying to you.
Crackdown is coming out after a significant amount of the One's lifespan has passed. And it's the only major use of it so far.
That means there is no chance Xbox One will be redefined on impressive new footing just because of the cloud. Not enough time even if Crackdown works well.
Saying something is "more powerful" doesn't always have to equate to "better graphics", even though in this case, it very well may. You'd never know though, as you'd never see what the game would have looked like if tasked with computing everything locally.
Won't it be really easy to tell? Just play the game offline.
But to this extent? There's exaggerating and blatant lying
Why would a game like this have an offline mode?
Did we both experience the same preceeding 2 years? Always online is absolutely necessary? Kinect is an integral component and can't be turned off? Windows 8/vista/ME is a huge step forward? Windows server is more stable than linux? We are Comitted to pc gaming? MCC is now fixed? Etc....
Because why not? Most games have an offline/single player mode save for MMOs. Games that aren't MMOs that have eschewed a way to play offline tend not to do too well.
At least with Crackdown it'll be really easy to tell exactly what benefit the Cloud is giving it. They'll undoubtedly make that playable offline.