True it won't be easy, but with extra power it should have at least some more scope.
Gaming PlayStation 4 (Console)
- Thread starter Biscuit1871
- Start date
-
In the latest episode of the United Hour Podcast, Nik (@Rood) is joined by CEO Duncan Drasdo and Financial Director Andy Green from the Manchester United Supporters Trust for a special episode on all things behind the scenes at Old Trafford including a brief history of MUST and their aims today, views on INEOS and Sir Jim Ratcliffe so far, and the Old Trafford regeneration.
It's a software problem rather than a hardware one.
I got a mark of 95 and 96 out of 100 in my two Artificial Neural Network modules at university, and 89 out of 100 in Declarative Programming - it was sort of a speciality of mine. It's complicated as hell.
Would you not say that's more reason to invest heavily in it?
It is something that still needs a lot of research. Plebs like Redlambs and other developers don't really understand it. Graphics is basic mathematics.
True, but with more powerful hardware it should make it more possible to expand the current software side in terms of AI and such as things can be run through quicker.
How does such a potentially interesting and exciting thread about the next generation of games consoles turn into a complete fecking borefest
The declarative programming approach (eg. Prolog - but you can do it in any language) is rule based - so it's scripted. You feed it information, and then it responds to the rules set down. The ANN approach makes it learn from its mistakes, and you need to train it to get to a particular level - it's far more suited to an MMO where it has the time to learn.
Sounds interesting enough. But I don't have the background or knowledge to actually discuss it in that sort of detail
This is the worst thread of all time. Only Weaste can make a next-generation games console sound completely uninteresting.
Lest we forget Redlambs coming up the rear, either.
What do you want to be discussed when we know very little so far?
Yes, however, there are a number of people encouraging him in his way.
Can you not get a grip of this thread?
Of course, you can mix the two approaches together.
The problem is, not many people know what they are doing in this regard.
Anything other than how you reverse the polarity of the neutron flow by wanking into its hard drive.
I know loads.
PM me?
The docs won't fit into a PM.
Shut up!
That's all you need to know!
I didn't quite catch what this chap is saying?
http://kotaku.com/5984963/check-out-final-fantasy-x-hd-version-for-the-ps-vita?utm_campaign
Not quite related but this has gotten me excited.
Not quite related but this has gotten me excited.
I've actually found what has been said so far interesting.
With the greater engine and capacity could they not go down the road of uniquely programming each NPC to at least give the impression of random behaviour? They already do something similar in that the "Bosses" tend to have one "personality" while the "footsoldiers" have another albeit every "footsoldier" acts the same.
That's what I mean by investing more in AI. You already have huge teams of artists and coders working on making a game like Skyrim have a huge and detailed Landscape, why can't you have a similar size team working on each and every NPC having their own personality and tendencies.
The leaked rumours so far have pretty much confirmed that with the architecture of the new consoles, any mid-high end PC you buy at the end of the year will be a very capable gaming machine for at least 5 years. With Publishers selling DD PC games at steep discounts within a year it's where I'll be at, unless MS/Sony can convince me otherwise with enough compelling first-party games.
As I said, it's a software problem, not a hardware one.
But you remove the software level crap! This box will perform like a high-end PC running Windows.
And unlike the MS Box, it does have some special sauce.
Some of the worlds in today's big games are graphically overly intricate. Many times you wonder about the practicality of making such overly detailed environments that do not serve any purpose to the interactivity of a game.
What's the point of incredibly detailed work desks in a game like Dead Space when more could have been invested in better AI or objects that serve a purpose to enhance the gamer's interaction within the Dead Space world. Assassin's Creed 1 is a perfect example of this. My purpose of playing a game is NOT to stare and critique the graphic artists' work. I want to be wow-ed by things that can be done in-game.
I know some people hate the convoluted story in the MGS series and mock the game for its cut scenes, but it's a series that utilises creative, inventive gameplay and interactivity so well to make its world, characters and enemies so darn interesting and fun.
Have we not been over this? How exactly do you suggest it is improved?
But you remove the software level crap! This box will perform like a high-end PC running Windows.
And unlike the MS Box, it does have some special sauce.
Well, since most third-party games will also be released on the next Xbox, they will almost definitely run on your i7 setup 5 years later as long as you don't mind the games looking slightly poorer at the tail-end of the generation.
There is no i7 set-up in either box!
Cell smokes an i7 at many things anyway. And guess what, the fecker's in!
With regards to AI, I agree with Poxbo's recommendations, and I'd like to see more done next-gen.
Is it a software problem or just the fact that not enough resources have been invested to improve it, unlike say, grahpics?
For a typical AAA game, would it be realistic to transfer some headcount and resources from the grahpics department to work on other stuff like AI and gameplay instead.
Not in the next Xbox surely. Thought the latest rumour was that the Cell isn't in the PS4? Even if it were, wasn't it only to run PS3 games? Also, aren't Sony & MS using some 8-core AMD notebook CPU instead?
By investing in more research and more teams working on AI per title. Maybe it isn't for this thread as it's not about the new hardware and it's capabilities as you have pointed out, but it's clear that AI is what lets down even the best of games.
Take GT5 for example. Beautiful game, great physics, very slick UI and all round great production.
But it's a racing game and the AI sucked. If you have a car to match the best car in the field, then if you are a half decent driver you should win every time because outside of following the racing lines, the AI had no tactics to fall back on. What sort of racing "Simulation" is that? You don't feel like you are racing against anyone, you just feel like you are trying to go faster than cars doing laps around a track.
As I've said, it's a software problem, more powerful hardware doesn't help with it!
I'm not talking about hardware! For a smart guy you don't half miss the point sometimes!
I don't miss the point, it's you that doesn't actually understand the point in the first place.
Rumours from The Times that it might be sold for £300 in the UK. That's got to be a load of bollocks surely, eh Weaste?
They might price it more aggresively this time around.
So how would you go about improving the AI in a game?