Digital Foundry Article Technical Discussion [2019]

[Recop]

The fact the game can run the whole campaign at 3 player split-screen on base XBONE and still have the campaign be up there with the best of the best in presentation, technical complexity and polish is the biggest highlight for me.

Performances are atrocious in this mode on XB1. It's more a gimmick than a real gameplay bonus...

 

[Globalisateur]

It's not only that... Gears 5 is a very linear game with limited physics for instance. I mean, there's no magic. There's no way the developers suddenly figured out what all other third party developers failed to do during the whole gen.

At this point, there's no debate anymore. The gen is finished and we have all the data we want.

I agree that exlusive games tend to be more polished, but there's a difference between a more polished game and something like closing a gap with a 40% stronger GPU.



It's an indisputable fact if we consider the resolution and the framerate. The Pro could not run Gears 5 in the same conditions.

We can argue that some games might have more advanced engines on Sony consoles, but they do it at the expense of the resolution/framerate.

I thought so based on DF article but I had the occasion of watching some different Gears 5 footage recently (VG tech video) and I am not sure about that anymore.

First Gears 5 has a very aggressive dynamic res from 1080p to 1800p (never reaching native 4K). Yes in small corridors highest res (1800p reconstructed to 4K) is achieved but otherwise it looks more like this at 1080p without proper taa coverage (and a blocky low res explosion). Secondly the game does not sustain 60fps at all, there are plenty of moments running in the mid 50fps range once shooting occurs.

Compared to God of war assets are much lower quality with low res assets everywhere like here:
Apparently the game is using base Xbox textures (and effects) which explain the low res assets everywhere and low quality of effects.

I say God of War is much more impressive, yes it's running only at 1080p on Pro using the performance mode, but at least it's consistent. I am sure if they had used a very aggressive dynamic resolution the game would have reached much higher res in linear corridors where the game is already easily locked at 60fps.

Yes in the main hub the game runs at 40-50 but this level is incredibly huge, much bigger than the biggest level in Gears 5. And the game in most moments run at solid 60fps on Pro. Notably in the snowy mountain, a beautiful level with snow deformation, weather effects, volumetric effects and such, all that with higher quality assets and high level of AF everywhere. I don't think there is much AF in Gears 5 from what I saw. I thought Xbox hardware was proficient at displaying high level of AF ?

They had others priorities in Gears 5, clearly. the aim was first to target 4K and target 60fps, whatever the costs (low resolution of assets, low resolution of effects).

 

[chris1515]

Yes. Just an example in North America. A title created in the bay area in California will cost roughly 3x the same title created in Montreal, CA.

So a title in the bay area at 30 million would be roughly the same as a title at 10 million in Montreal in terms of work done for money spent.

IE - comparing monetary budget across countries or even regions within a country (Houston is cheaper WRT development costs than San Francisco, for example) is not terribly accurate when trying to figure out how much of a budget X dev team had versus Y dev team in a different location.

Things get even more muddled when you start outsourcing parts of the development to countries like say China, the Ukraine, etc. where development costs are really low.

Regards,
SB

https://www.pcgamesinsider.biz/news...dget-for-remedys-control-between-20m-and-30m/

The CEO of Remedy talk about peak development at 100 and that it was very difficult to do a AAA with this budget...

Remedy's latest title Control is being made with a budget of between €20m ($22.2m) and €30m ($33.3m).

That's according to Remedy CEO Tero Virtala, who said in a talk at Devcom 2019 that the title was made with triple-A experience but a more moderate sum than its previous titles. A side effect of this is that there's less pressure on the Finnish studio to hit huge sales figure before it has recouped its budget.

"We would say it's about a sensible budget and scope. Being able to create a full-price triple-A game is really hard," Virtala explained.

"Being able to do that on a budget of €20m to €30m is very safe. It's been about three years now and at the peak of production we had about 100 developers on average."

He continued: "I can't give you the exact number, but within one-to-two million units, we will have broken even. After that we are profitable."

Naughty dog including the ICE Team they are 520 people, Guerrilla Games want to reach 400 people, Coalition they are 400, 343 Industry more than 500..

And I don't even talk about Rockstar with more than 3000 peoples in the credit of the game...

 

[Recop]

I thought so based on DF article but I had the occasion of watching some different Gears 5 footage recently (VG tech video) and I am not sure about that anymore.

Running a game at twice the framerate is extremely expensive and i have no doubt that the Pro would have performed notably worse, unless there's a major CPU bottleneck. The Pro would still perform worse whatever is the bottleneck though : it has virtually no advantage over the X.

Yes in the main hub the game runs at 40-50 but this level is incredibly huge, much bigger than the biggest level in Gears 5.

Actually, it can be worse than that :

 

[Scott_Arm]

Running a game at twice the framerate is extremely expensive and i have no doubt that the Pro would have performed notably worse, unless there's a major CPU bottleneck. The Pro would still perform worse whatever is the bottleneck though : it has virtually no advantage over the X.



Actually, it can be worse than that :

Looking at the frame time stats, god of war does run noticeably worse on ps4 pro than gears5 does on one x. Gears 5 stays at 60Hz 95% of the time, where god of war only stays at 60hz 85% of the time. At the same time, they both have different designs, different visual goals etc etc, and I'm not sure how anyone can make a fair comparison. PS4 Pro is also a less powerful console, so you wouldn't expect the same performance profile for the same game anyway.

 

[chris1515]

https://www.eurogamer.net/articles/digitalfoundry-2019-teraflop-face-off-current-gen-vs-next-gen

As for where this leaves the next generation consoles, it's all good news. On the face of it, the results may suggest that a six teraflop GPU like Xbox One X's would translate to an 'equivalent' 8.1 to 9.5TF GPU with a Navi-based successor. It's an overly simplistic calculation to make when there is so much more to the make-up of a GPU than its compute power alone - and the uptick we've measured here may actually be significantly higher depending on the workload. However, it would certainly go a long way towards explaining why Microsoft is considering the Project Scarlett box codenamed Lockhart rated at 'just' 4TF of GPU compute - and we may yet see this machine hit the market: Microsoft's PR focus is all on the higher-spec Anaconda box but I've yet to get any firm confirmation that Lockhart has been cancelled.

As things stand, a 40 to 60 per cent architectural improvement in performance is impressive, especially when further amplified by the inevitable increase in GPU frequency and the big increase in memory bandwidth. And this is just talking about hardware specs, when much of the magic comes from game developers. It's difficult to imagine that the likes of Uncharted 4, God of War and Horizon are essentially running on a customised Radeon HD 7850, while Forza Horizon 4 and Gears 5 are delivering a phenomenal return from what is basically an underclocked R7 360. And with that in mind, no matter what the configuration is that the next-gen consoles end up with, Navi-based silicon should be phenomenal.

 

[Recop]

PS4 Pro is also a less powerful console, so you wouldn't expect the same performance profile for the same game anyway.

As simple as that.

And since Sony and MS seem to have very similar hardware now, i think it will be easy to find out which console is the most powerful the next gen pretty quickly.

The secret sauce argument ended with the 360/PS3 era.

 

[Karamazov]

secret sauce are exclusives now
(and samuraï sauce with french fries)

 

[Johnglen]

This is actually debatable... airing more towards opinion rather than actual fact.

Yes, Gears 5 looks like a technological masterpiece across any platform... and The Coalition deserves lots of credit on making one of the best Unreal Engine 4 based games around. However, many would argue that God of War, RDR2, Forza Horizon 4, GTA V, FarCry 5, Minecraft, and the thousands of other game choices are more technology advanced than Gears 5 Unreal Engine tech. It doesn't mean they're right, or that they're necessarily wrong. Without knowing every aspect of a particular game's graphic pipeline, we can't honestly say one is more advanced than the other. Their maybe one or two aspects of one engine that outweighs all other features compared to another, the latter being considered to have the most advanced features.

Example: In my opinion, Remedy's Northlight Engine behind Control (multiplatform title), seems lightyears ahead of any other engine including the Unreal Engine, when dealing with physics and environmental destruction. Including aspects dealing with real-time scene deformation, transformation, and warping (such as the maze run) that are ungodly levels of complexity (completely unbelievable).

Point being, all engines have their strongpoints and weaknesses. So, there are points for debate on what constitutes a more advanced engine (or game) over another.

This is one of the funniest posts ever in this forum. Here’s a tip Gears 5 runs at double the frame rate of all these «more technically advanced» games while looking substantially better and , at least compared to PS4 pro games, at much higher resolutions too. Comparing Gears 5 with something like....far cry 5 and control is ridiculous anyway you see it. Whatever, bringing down Xbox games has been a thing since the xbox360 and will continue in the foreseeable future.

 

[BRiT]

Less game comparison on subjective qualities and more discussion of DF Technical articles, please.

 

[Shortbread]

This is one of the funniest posts ever in this forum. Here’s a tip Gears 5 runs at double the frame rate of all these «more technically advanced» games while looking substantially better and , at least compared to PS4 pro games, at much higher resolutions too. Comparing Gears 5 with something like....far cry 5 and control is ridiculous anyway you see it. Whatever, bringing down Xbox games has been a thing since the xbox360 and will continue in the foreseeable future.

Calm down, It's an opinion piece. So no need for the console warrior defending antics.

 

[Svensk Viking]

I miss back when Digitalfoundry tried to match the base console settings on a PC with similar specs. Now that it's late-gen games, I'd like to see if a HD 7950 or GTX 680/750Ti still unquestionably outperform the consoles.

 

[Shortbread]

Sometimes it works with great results, sometimes it doesn't - but like any technology, DLSS is evolving and improving over time and the Nvidia anti-aliasing technology gets more control, more flexibility and some impress new image quality enhancements in Control. Alex checks it out.

Love this game...

 

[techuse]

I miss back when Digitalfoundry tried to match the base console settings on a PC with similar specs. Now that it's late-gen games, I'd like to see if a HD 7950 or GTX 680/750Ti still unquestionably outperform the consoles.

A 750ti does not compete with Ps4 in the vast majority of cases but it can probably deliver results comparable to Xbox One. A 680/770 is more of a match for Ps4 these days. The 7950 performs a bit better.

 

[PSman1700]

A 680/770 is more of a match for Ps4 these days. The 7950 performs a bit better.

More like 7850 or 7870, a GTX670 or higher is asking too much in most cases.

 

[techuse]

More like 7850 or 7870, a GTX670 or higher is asking too much in most cases.

A 670 falls behind a Ps4 in lots of titles these days. It has not aged well.

 

[PSman1700]

A 670 falls behind a Ps4 in lots of titles these days. It has not aged well.

When that 670 pc is running at higher preset settings yes.

 

[techuse]

When that 670 pc is running at higher preset settings yes.

You should check some recent benchmarks. A 670 more often than not gets beaten handily by a 7870.

 

[PSman1700]

A 7870 also more often then not outperforms the base PS4. Looking at benchmarks, a 760, or anything in the 670 range seems to perform better then a base PS4, without having to go around with dynamic resolutions, at somewhat higher settings too. A 670 was a 7970 class GPU from early 2012. GTX 780 can't in life be compared to the base PS4 at all, its probably closer to the pro then.

 

[Deleted member 2197]