Scalability of ND's Uncharted engine *spawn

Does anyone know what is current [or at least last known] memory footprint of the XMB OS in PS3? I remember reading a while ago that its modular, devs can use only the "plugins" that they want [with spelling dictionary being >20mb :D ].

Who told u this?

(who ever told you was lying..)
 
Who told u this?

(who ever told you was lying..)

There's another thread where this was discussed and it seemed to be the consensus that:

A) The XMB now only reserves ~24 MB system Ram and no VRAM any more

B) Some of these "modules" numbers seemd way too high for what they were
 
1) ND Uncharted 2 engine is a beast!

Indeed! Since reading this thread I've played with this in mind and I have to say the game is simply brilliantly designed.

What I don't seem to get is why it's almost being critisised "it's only this good because they cheat" - the bottom line is anyone else could cheat the same but don't and where they do (Gears IMHO is a great example) it's just simply not as good.

FTR I'd love ND to do a J&D game - we know how talanted they are and so a cartoon style game would require less fidelity and offer a vey impressive open world (from pants memory J&D was open world)...
 
What I don't seem to get is why it's almost being critisised "it's only this good because they cheat"

Noone has criticised it. It's merely used as an explanation.
In fact as much as I see everyone pretty much agrees that it is one of the best looking games in this generation...
 
Noone has criticised it. It's merely used as an explanation.
In fact as much as I see everyone pretty much agrees that it is one of the best looking games in this generation...

well I did say 'almost' - but in your example...

"it is one of the best looking games in this generation...because the player is restricted in what he can do and where he can go"
 
People aren't supposed to talk about that though?

"On rails" games always have a big advantage compared to "open world" games in terms of how good any one screen looks. There's no reason to think that developer X could make an "open world" version of their "on rails" game without making compromises and changes.

I don't see any harm in taking into account what a game is doing when talking about how good it looks, and it does relate directly to the thread's title.

The amount of memory you could allocate for detailed or unique in game objects would be reduced, as would manpower available to generate unique content and a high density of scripted events. You'd also have to play it safe with the average rendering load, simulation complexity and density of in-game objects because you can't be so sure about what the player will line up or cause to happen in the game world / simulation.

The average ... everything ... that the player experiences would drop.
 
I would place UC2 as "others".
On rails would be like Sin and Punishment, RE: DC, Panzer Dragoon.
Sand box being GTA, Fallout, Farcry
Others being every others not above. :D
 
Yes. In a 'on-rails' game with narrow predefiend path you could use sprites very very close in the parts that are not accessible. However would it be more open-style it wouldn't work as you would have this in-your-face LOD swap when closing up on vegetation/objects. So then the solution would have to be to up draw distance for vegetation/objects to a distance where the LOD swap is less visible. Doesn't mather if it is a plain LOD swap or by dissolvement. This ofcourse would greatly increase load on the system having a large impact on perfomance and memory use. So solution would need to be reduced detail on obejcts in average to ease the load on system and memory.
 
well I did say 'almost' - but in your example...

"it is one of the best looking games in this generation...because the player is restricted in what he can do and where he can go"

And where did he say "cheat" there then? He's stating his opinion, backed up by his rational and the fact the player is mostly restricted on where he can go/do. If that is incorrect or you disagree with his analysis by all means discuss it, however we're all adults here and understand that all games (' engines) compromise leveraging smoke & mirrors.

I also agree with obonicus when bringing in "* with modifications" just invalidates any discussion. With enough modifications, you can make Wolf3D look as good as Uncharted 2 afterall. Grab the source code and a compiler and come back in 2 years.

The spirit of the thread, as I understand scalability of an engine, is "if I were to license this engine and had 0 hours on my budget for engine-programming could I make an Oblivion-type game with the same level of visual fidelity of Uncharted 2?". This isn't an academic exercise: many engine licensing decisions are strongly based on the (lack of) human/material resources to modify the chosen engine.

That is why you rarely get an engine license that deviates a lot from the engine's debut title either in genre or gameplay, unless that engine was specifically made to be as generic as possible thereby sacrificing some other aspect.

Everyone talks cpu, memory, threads, streaming but something as simple as your data structs used in the engine can have drastic implications on the applicability of that engine in some genres (off the top of my head, racing games' engines applied to RPGs).
 
And where did he say "cheat" there then? He's stating his opinion, backed up by his rational and the fact the player is mostly restricted on where he can go/do. If that is incorrect or you disagree with his analysis by all means discuss it, however we're all adults here and understand that all games (' engines) compromise leveraging smoke & mirrors.

I also agree with obonicus when bringing in "* with modifications" just invalidates any discussion. With enough modifications, you can make Wolf3D look as good as Uncharted 2 afterall. Grab the source code and a compiler and come back in 2 years.

The spirit of the thread, as I understand scalability of an engine, is "if I were to license this engine and had 0 hours on my budget for engine-programming could I make an Oblivion-type game with the same level of visual fidelity of Uncharted 2?". This isn't an academic exercise: many engine licensing decisions are strongly based on the (lack of) human/material resources to modify the chosen engine.

That is why you rarely get an engine license that deviates a lot from the engine's debut title either in genre or gameplay, unless that engine was specifically made to be as generic as possible thereby sacrificing some other aspect.

Everyone talks cpu, memory, threads, streaming but something as simple as your data structs used in the engine can have drastic implications on the applicability of that engine in some genres (off the top of my head, racing games' engines applied to RPGs).

But then what about UE3 being used in JRPGs?!?!

I'd have initially compared GeOW and Mass Effect, but ME is in fact pretty much a TPS of the same ilk but with RPG pajamas on.

But OTOH comparing jrpgs like lost odysee with GeOW? These are radically different games and I'm not all that convinced that UE3 was made to be as "generally applicable to all game types" as possible...

Maybe it's just me?:?:
 
And where did he say "cheat" there then?

he didn't, it's just how I read some of the comments - it seemed negative towards ND - in this way:

"yes, ND made the best looking game so far this gen, however that's because XYZ" (i.e. less about how well ND did (or how well the PS3 copes with what's thorwn at it) but more about 'anyone else could do this', it's nothing particulary special).

I'm probably reading wrong, but that's how SOME comments have come across (IMHO)...another example would be the simply wrong comparrison to an 'on rails' game (or did I read that wrong too!?)

Personally I've agreed with pretty much every point about how hard (read nigh-on impossible) it would be to translate to an open world game. However, that doesn't mean ND haven't (or aren't) using part of this engine to make a new 'open world' game.
 
But then what about UE3 being used in JRPGs?!?!

I'd have initially compared GeOW and Mass Effect, but ME is in fact pretty much a TPS of the same ilk but with RPG pajamas on.

But OTOH comparing jrpgs like lost odysee with GeOW? These are radically different games and I'm not all that convinced that UE3 was made to be as "generally applicable to all game types" as possible...

Maybe it's just me?:?:

My understanding is that UE3 is meant to be used for a wide variety of game types, because Epic's business is in selling their engine. I'm not sure how much modification is required. I guess that would be the question of budget for modification that Richard mentioned.

But what does that have to do with comparing rpgs to fps games? Not sure what the connection is in the context of this thread.

If someone has technical details on how the streaming works in Naughty Dog's engine, that might go a long way in understanding what it can and can't do. They may be streaming in a very different way than would be useful for an open world game. It could also be perfectly suited and designed to support open world in case they decided to go that route in the future.
 
well I did say 'almost' - but in your example...

"it is one of the best looking games in this generation...because the player is restricted in what he can do and where he can go"

Yeah - what's wrong with that statement? It is true, and it would be the same for any other game, on any other platform.
 
But

"yes, ND made the best looking game so far this gen, however that's because XYZ"

and
about how well ND did (or how well the PS3 copes with what's thorwn at it)

are pretty much the same, can't you see??

ND is making very clever and highly optimized use of the system's resources and strengths. Adding that it also means making compromises doesn't make it any less of an achievement, neither technically nor artistically.
 
But



and


are pretty much the same, can't you see??

ND is making very clever and highly optimized use of the system's resources and strengths. Adding that it also means making compromises doesn't make it any less of an achievement, neither technically nor artistically.

That's not what he's saying. He's saying that comments are coming across like this:

"Yes, Uncharted 2 looks good, but Naughty Dog does a lot of cheating, and their game design allows them to do that".

Which I think is unfair to the developer. It doesn't matter how they achieve their look, or what compromises they make to get there, at all. It's completely irrelevant. To somehow shrugg off their achievement as if to say "anyone could have done it under the same circumsatances" is silly.

Obviously the engine already handles some form of LOD, and given Naughty Dog's history, I think it's perfectly rational to assume they absolutely could use this very same engine to create a more "open world" experience.

The fact is, that is NOT the type of experience or game they are making.
 
I think some people are just getting really defensive about Naughty Dog. I don't think anyone is suggesting they're "cheating," just using visual tricks and smart optimizations like any other developer would. No one is downplaying their achievement.

Obviously the engine already handles some form of LOD, and given Naughty Dog's history, I think it's perfectly rational to assume they absolutely could use this very same engine to create a more "open world" experience.

The fact is, that is NOT the type of experience or game they are making.

I'm not convinced that's a rational assumption. Uncharted is not an open world game, by design. I'm not sure why we'd assume it could be. What does "a more open world" experience mean? If you mean bigger set pieces, then maybe or maybe not. They said Uncharted 2 maxed the PS3, which is probably not true exactly, but even providing bigger set pieces could push beyond their memory budget, depending on how their streaming works. If you're talking about a true open world experience, where you have many square miles mapped, that can be traversed in any direction, then you're talking about something very very different, and I'm not sure we could assume this engine could do that with any quality similar to Uncharted 2.
 
That's not what he's saying. He's saying that comments are coming across like this:

"Yes, Uncharted 2 looks good, but Naughty Dog does a lot of cheating, and their game design allows them to do that".

Which I think is unfair to the developer. It doesn't matter how they achieve their look, or what compromises they make to get there, at all. It's completely irrelevant. To somehow shrugg off their achievement as if to say "anyone could have done it under the same circumsatances" is silly.

Thanks. Like I said though, it may just be how I read it.
 
I think some people are just getting really defensive about Naughty Dog. I don't think anyone is suggesting they're "cheating," just using visual tricks and smart optimizations like any other developer would. No one is downplaying their achievement.



I'm not convinced that's a rational assumption. Uncharted is not an open world game, by design. I'm not sure why we'd assume it could be. What does "a more open world" experience mean? If you mean bigger set pieces, then maybe or maybe not. They said Uncharted 2 maxed the PS3, which is probably not true exactly, but even providing bigger set pieces could push beyond their memory budget, depending on how their streaming works. If you're talking about a true open world experience, where you have many square miles mapped, that can be traversed in any direction, then you're talking about something very very different, and I'm not sure we could assume this engine could do that with any quality similar to Uncharted 2.


I think it's fairly rational.

Look at GTA IV, yes it's a fantastic looking experience, however, the interactivity of everything is extremely limited. The majority of the buildings are not accessible, and it doesn't have the "intamate" geometry that Uncharted has. If you remove a lot of that "intamcy" of the indoor environments in Uncharted 2 (rather, the engine) and move that focus over to buildings and LOD, then you can make more room to load more for the unpredictable nature of the player.

I'm not suggesting their engine could make an open world game with the exact same quality as Uncharted 2, visually, I'm suggesting that the engine would almost certainly be capable of some "open worl" environments, i.e. area's you can traverse in any direction.

Granted, part of the reason it looks so good is because the LOD is obviously scripted in such a way to draw things cleverly when the player cannot see them, using camera angles and what not to hide this from view. However, I think that they could still tweak their engine to do something full scale ala inFamous or GTA IV.

Afterall, you can skip a cutscene and the next area is already completely loaded, which is pretty impressive considering the complexity of the environments.

So again, I don't think their engine can produce an Uncharted 2-like visual experience, but I completely think it would be capable of a GTA IV or better visual experience in an open world game.
 
But then what about UE3 being used in JRPGs?!?!
UE3 was designed to be fairly flexible. They had to compromise the system to accomodate this flexibility, hence no UE3 game is going to look as good as U2. If the engine were as finely tuned to produce U2 level visuals, it wouldn't work so well for racers or platformers who who-knows-what.
 
Back
Top