Yeah Dev’s are lazy they should be making new lighting system’s,new engine’s, new design, Etc there so used to old gen I don’t see dev’s making the effort.

What they said is getting into specifics, and is a clue that supports the idea that the speed/strength of the consoles depends on development approach. Keep in mind when they share their experience, its based on the current situation with the API. We all know things can change as it improves.

So add what they said and look at what was achieved with FH2, which was drawn up using a form of forward rendering and you can see that my assessment is sound. Close to 100% of games are done using deferred which as of now, do not cater to the strengths of the XB1. BUt the XB1 can keep up despite that. Things will change though, because its normal that it do as Devs get more accustomed to these machines. And I’m not talking about better API, just better software hacks like the ones employed in FH2, and from what it looks like, Quantum Break.

Can the witcher run on PS4 if it used forward rendering to draw, with all that expensive lighting? Can it run on XB1 the same? Deferred is the preferred method, and for now like CD Projekt said, not so friendly with XB1’s ESRAM. In this case PS4 > XB1 in development environment friendliness. Same old story. I get what they are saying, while most take it out of context and exaggerate the difference in power.

“I’m led to believe that the XOne would actually be at an advantage here since deferred requires lots of read/writes to the G-Buffer”

Dude, lets not pretend that we’re smarter than the dev in regards to how they build their games. CD Projekt said, ” targeting Full HD with a complex rendering pipeline is a challenge in terms of ESRAM usage. Especially when this pipeline is built on deferred rendering.”

So how is the XB1 at an advantage? He just explained in one lick why some games are @ lower res on the XB1.

About not trusting DF: I gave you 2 instances, you concentrate on one. There are plenty more, but thats a waste of time. Trust your eyes, not a pixel count.

I agree with everything else you said.

Smooth……..time will tell. Eh man thx for a nice lil debate, u and GHz. Good to see some constructive argument, not a bunch of unfounded, biased hatred lol. I can totally accept someone’s preference, but it’s tough to see the constant immaturity and hate online.

Sorry GHz but CD Projekt didn’t go into specifics of the ESRAM, they only said that it was complicated to work with its limits specially with an engine using deferred rendering, if you have a source that says otherwise, please send me the link.

Speaking of deferred rendering, I’m led to believe that the XOne would actually be at an advantage here since deferred requires lots of read/writes to the G-Buffer, but even then it’s hard for me to see how just the ESRAM bandwidth can compensate for the lack of cores on the GPU specially since its Bandwidth is not much faster than GDDR5. Moreover, the memory can only move information if the processor can provide it.

In both cases it seems to me that the XOne requires lots of optimizations to work efficiently and to get the same results it would have to simplify the calculations (lighting maps for example). This still can result in great visuals like RYSE but the end result is more about artistic direction than raw calculations.

QB was demoed in a controlled environment. I’m pretty sure everything shown was handpicked to display the best case scenario, you should know that, not doubting their capabilities but I don’t remember anything special coming from the studio behind QB.
Might be stunning? Yes, but we can’t judged products which are not final.

Digital Foundry learned that guerilla implemented a new upscale method in KZ which was probably why they missed it. Dismissing their analysis for one case isn’t handpicking the examples that suit you better?

I totally agree with the gaming media part. I’d say 99% of the fanboys on sites like IGN would never notice the difference if these same sites didn’t bring that up (of course they are loving it, so many clicks per article that is not even funny). Side by side a 900p vs 1080p game look the same, but when fanboys know about it, they just know everything 🙂

We’re talking about 1st party maximising the potential of their hardware. Each will approach development to better suite their respected platform. The results say a lot.

In regards to 3rd party tackling multiplats, and the conditions they have to work in, you’re right. In regards to the XB1, nothing can be done until it is powered by its true API. Then we can judge.

There is a presumption that since there are 50% more CUs and 100% more ROPs, that these parts can be used at 100% efficiency. The devs are building real-time rendered games, not running benchmarks.

Two MS engineers gave an interview with digital foundry. They stated that they kept XB1 as balanced as possible, and that overclocking was more effective than enabling the last two CUs (12 + 2). They stated that at the throughput number they threw out for eSRAM plus DDR3, those CUz could not saturated or even be cost effective vs. overclocking. Maybe they were full of it. Who knows…

You are dealing with two different games. No, not all devs are created equal, but it is silly to put unnecessary limitations on their work. If two teams are contracted to build a house, and you give one team proper tools and equipment, while the other gets makeshift tools, then someone has to make concessions or find time consuming workarounds. Think about how much wasted time there is when you try to do a job with a crappy wire stripper or battery operated tools vs. plug-in.

It is a fast lane, but their is way too much traffic waiting to use it. Being limited to “one lane” and the increased but finite speed limitation doesn’t adequately alleviate the bottleneck in the slow lanes.

In short, it is preferable to DDR3 alone, but doesn’t solve the problem. I suppose Microsoft committed to DDR3 to guarantee 8 GB per console.That solution lneeded to be propped up with embedded RAM. Now the die size is drastically increased, increasing complexity, thermal envelope & costs, while lowering yields. Thus, the chip is limited to 32 MB eSRAM and 12 functioning compute units. Viola, we have the XB1 as we know it today.

Sony took a gamble on DDR5 availability and lucked out. Who’s to say if both MS and SONY had insisted on DDR5, a shortage would not have resulted, yielding 4 GB DDR5 consoles and/or limited availability?

Nope, dont compare me to you. You’re the one that brought up CD projetk as an example. But you chose to stick with only a part of what they said, because it fits pretty in what you want to hear. You don’t even bother try to understand the other things they said to get the whole picture. And when i said the games are the proof, not only did CD projetk explained in plain english how ESRAM relates to deferred rendering, there is a game, FH2 that did well w/o it. CD projetk project understands this. But the witcher 3 is multi platform and its safe to say PS4 wouldn’t be able to handle forward rendering with all that global illumination. Not with the size of the worlds in that game. Deferred would be the better choice because even though the XB1 is at a disadvantage, at least it’ll be able to keep up. If it was the other way around, then that would’ve been headaches to get the PS4 running using forward. Understand? no biggie. it is what it is. In this instance, PS4 is the more powerful machine. They are talking about approach to rendering.

The problem is pple forever take what these devs say out of context, to better fit what they want to hear, and for the more technical minded, their limited understanding. Or in the case of a gaming site like this one or eurogamer, headline to get clicks, but like that CD projetk interview reveals, there more to it than just that. You just have to understand what they are saying.

About FH2, now your just making stuff up. Same advance physics engine from Forza 5, just tweaked for off road driving too. Its no exaggeration. You have websites dedicated to driving enthusiast, and they swear that F5 & FH2 is more or less accurate in displaying how these vehicles perform on road.

QB, not only have we’ve seen 16 minutes of real footage running on XB1, but it has been verified by attendees at Gamecon. QB stole the show, the most next gen looking game there at the time. No gimmicks. Why would you compare that experience to that of Uncharted 4? Sony haven’t shown no gameplay let alone showed it off to the press. All we got was an “in engine” cinematic intro.

As far as DX12, don’t confuse graphic res with graphic features. It’s accurate to answer a question like,” will DX12 improve graphics on XB1?’ with the the answer, “not much.” Because graphic fidelity can improve independent of DX12. Middleware software can handle that job. DX12 is about opening up more features and capabilities, and MS said that they are NOT ready to talk about what the features are yet. NDA. Throw in the fact that both platforms are only a year old, best believe that graphics will improve over time. It’ll be stupid to believe otherwise.

No one mentions anything about miracle in regards the stregnths of the XB1. You however ant to believe that 50% w/e translates to 50% better looking game. Thats false. Devs already cleared that up. You van believe anything you want though.

“multis will have very similar performances on both platforms”

YUP! its happening already, Destiny, CODAW, ACU, Alien Isolation etc.

“exclusives pushing the limits”

FH2>DC in so many ways its not even funny. Trumps it in overall presentation and scope. We can compare these 1st party efforts. No diss to the ps4, it is what it is.

Digital Foundry couldn’t even tell us that KZSF MP was 930 X 1080p. They told is it was full 1080p and a solid 60 frames per second. They were wrong on both accounts. They told us ESRAM was a bottleneck until they were told that its better code render targets to fit ESRAM 1st because it benefits all GPU’s with caches. So what else they got wrong? How trustworthy are they?

My advice, forget a pixel count and trust your own eyes and make your own assessment in regards to the games you want to play. Too much funny business with the gaming media.

If I’m repeating what other people are saying so are you since the argument ongoing for the XOne is exactly the same “Wait for the new engines”. There will be improvements but no revolution.

FH2 is really nice but hey, lets not exaggerate, the graphics are nice but the scope of the game is nothing extraordinary and it’s an arcade game not simulation, so its more eye candy than physical effects which exactly for not being simulation are simplified, so forza is not a good example of what a system is really capable.

Bringing Quantum Break to the discussion, sorry but I have to say this, is borderline fanboysm since the game is months away from release and we all know how hyped up games are these days. I could argue that Uncharted 4 will beat it in terms of graphic quality hands down, but then I wold be playing on the same field, which is speculation.

The DX12 is another miracle fans are waiting to happen that will not. Again, it will improve the system but will not make miracles, these are optimizations and lower level APIs that will allow developers to get more power from the system. And optimizations are always part of the console lifecycle, PS4 also receives optimizations to its libraries and since the architectures are basically the same, there is nothing the XOne can do that the PS4 can’t and vice-versa.

It’s Esram and parallel memory access vs GDDR5 and 50% more GPU cores. Engines that “understand” Esram will benefit from it, but there is no miracle. In the end, as I mentioned before, multis will have very similar performances on both platforms with exclusives pushing the limits and making no difference in the end since most of the time people will be doing comparisons without seeing both sides. We will all have to rely on Digital Foundry for the conclusions.