Just Cause 3 Engine Incompatible with SLI/CFX ?

For the most part you answered your own question.

Rewriting an engine for DX12 isn't something you can do in between game releases. UE4/Epic has been working on it since October 2014 (the first DX12 build was released, they've probably been working on it before that) and still doesn't have it fully rolled into their latest release.

For the second part, what's the point? Like you say manufacturers are pushing 4K, but it still has below 5% market saturation. And Multi-GPU setups are also rare. Does Avalanche really going to spend 50-60M rushing DX12 adoption for the handful of users it would actually effect? Like most of these people are buying the game already. There is maybe like less than a hundred people with SLI/Xfire/4K that would actually buy the game specifically for those features.

4K adoption has to come from monitor manufacturers making it worth it. Multi-GPU has to come from Nvidia/AMD pushing it to devs and making it easier to develop for (something AMD has been heavily focused on).

 

You're absolutely right, but what they could do instead is offer an upgrade by adding an external GPU.
I know this would pretty much violate the whole plug-and-play point of consoles, but this would be the only way I'd see consoles supporting multi-gpu configurations.

Plus peasants would think it's the second coming of jesus christ and claim that PCs do not support such a thing. Would be funny to see top kek.

Jokes aside, I'm sure this idea has huge sales potential and developers could not afford to ignore a decent chunk of console users.

 

If there's anything that has potential in DX12, it's discrete + IGP. The studios know that that is the most common multi-GPU setup right now.

 

Yea i was gonna say that to Denial, if you factor in iGPU's then that must be a huge percentage of people.

So yea most people already have 2 gpu's in their system but obviously disable one of them. Someone must have thought about all those disabled gpu's and thought, how can we use that disabled part of the system and make it useful.

Imagine how nice it would be if you could actually use that disabled part of the system for say a 20% boost For Free. :0

 

Here's the thing with SLI and Crossfire, its been around for over ten years, how are people still calling it a "gamble". It's a proven technology. "It's only a minimal boost" I don't care about frame rates per say, my eyes tell me a 60fps game looks really smooth and I'd rather play at that frame rate. When games fall under 60 I do notice it especially under the 40fps threshold.

SLI/Xfire boost means at 2k-4k res my game goes from 40fps to 50-65 where I want it. This isn't a guess it's noticeable. 4k gaming is also being touted as amazing (it does look really sharp), to get there at a respectable framerate without forking over $2,000 for next years ultra super single GPU solution you need SLI.

Nvidia knows this and they promote SLI. AMD knows this and they promote Xfire. 4k makers know this and they hold their breath. Game publishers however don't make a dime off of SLI or Xfire necessarily. They push their developers to get the game out of the door as quickly as possible.

Batman Arkham Knight - Rocksteady had their full attention on consoles, they hired Iron Galaxy to do the PC port... why? Money.. Consoles make the money (although that's really not even as fair as game publishers imagine).

Just Cause 3 has deadlines too, Avalanche just had to let go a bunch of people (Mad Max probably didn't do to well in sales).

The point? The real source of major triple AAA games not getting extra development time is Publishers. They hold the strings for development time.

Now I think I'm pretty realistic, I know SLI users represent a lower fraction of the gamer populace. So you release your game out and single GPU users are happier .. that's fantastic.

Just tell me its coming.

Don't like to me and say its coming (Arkham Knight) then don't ever deliver, and change your story to say its never coming.

Just tell me its coming, tell me you'll post release patch it in. Any game engine can do SLI, all you need is a developer who will put the time in, how much time? Considering how JC3 is the same engine just beefed up I would like to think its not a lot of time. Obviously their new effects broke SLI/Xfire sync. So just tell me its coming. That's all I want to know. I have a backlog of over 200 games I can play around with, I can leave JC3 on the backburner for now. To say its never coming from Avalanche, to me that's just lazy.

PS - For me JC3 hovers at 30fps-40fps no matter what I do at 2560x1600 res. It is noticeable. At least Arkham Knight runs 40-50fps for me with only one gameworks option turned off.

 

I'm not saying that multi-adapter will never be a thing.. but as a company you set schedules for game releases and have expected revenue targets for the year. Is Avalanche really going to delay JC3 for potentially a year (the time it took UE to implement DX12), just to get a 20% fps increase? Or should they be incrementing the development over time and multiple releases? I think the latter is smarter from a business perspective.

Like with JC3 they implemented a entirely new lighting system but that system can still be compatible with DX12 multi-GPU adapter. Just not SLI/Xfire. They are probably already working on their next game with a newer iteration of their engine, perhaps it will feature DX12 but not the multi-GPU adapter part, that may come even further down the line with a different game.

These studios aren't going to pause their game development process and have artists/designers just sitting around as their engineers attempt to get DX12 working. They are going to be working on finishing up games while the technology/engine team slowly iterates the engine in the background.

 

If they adapt JC3 for DX12 then there is still hope for multiGPU usage AND better performance.

 

What are you talking about? The 780 ti was faster back then, and it's still faster today. http://www.hardwarecanucks.com/foru...ws/70125-gtx-780-ti-vs-r9-290x-rematch-8.html