I think it is a good suggestion and I can explain in detail why:
Nobody (not even the critics in this thread) thinks DLSS and let’s add FSR would be negative additions.
Granted, the plugin might not work with CE but the normal SDK does. I don’t think there is a single finished game on the market using the plugin - it just makes it easier for future UE4/5 projects.
Funcom may think it is not worth their time, but how would you know that. It is not that much of the time needed after all.
In some videos, somebody said, that there won’t be changes on the engine. Okay, but there are new features getting added for example “character transfer”, the “anti-under-mesh-solution”, and even “performance” is listed as its own headline in many patch notes.
I don’t see the one unbeatable argument.
Also, I don’t “expect” the dev to add those features. On contrary, I expect them not to add anything we could ever suggest. But it would be neat, wouldn’t it?
Mmm, Well, all either of those systems do is add framerate. Both of them degrade the visuals in order to accomplish that. So, I’m not sure “neat” is the right adjective. If FC thinks that’s needed - like for example too many console users are getting really bad framerates and etc. then these are systems to consider. But if that’s not the case, then maybe Funcom’s response to such a “request” would be something along the lines of: “But why?”
In my own case I wouldn’t use it other than just to test out for fun. I get 90 to 130 fps on my system playing with the settings maxed, on a 4k monitor (though half the time in a 1440 UW window). Occasionally it drops to 60 when there’s A LOT happening in a very heavy area - but my personal solution there if I thought it was bad enough, would be to turn down some settings maybe. It wouldn’t be till the game was more consistently down around 30fps that I would turn to a solution like FSR or DLSS.
Others will have different takes on all this I’m sure.
To explain further, when I responded to drachenfeles with /thread. It was because I agreed with his statement of Facts. This was with the knowledge that Conan Exiles is in UE4.15CE, meaning Conan Exiles is in a modified version of UE4.15. I know this for a fact as I have it downloaded and have now made three mods with it. In addition I have seen dozens of developer responses and several videos saying they wouldn’t be going to 4.26 or 5.
drachenfeles did not post hearsay. They did not post rumors. They did not post opinions. They posted Facts.
That means DLSS as the OP, Korahan posted is not going to happen. Discussion isn’t disallowed no (and I never said it was). But it is ultimately futile. I’m sorry that is a statement of fact as well.
I would love for Conan Exiles to be moved seamlessly to UE5. That is a desire of mine. But I also realize this is a fantasy. So the question I would have for anyone who is still offended by the /thread comment, what purpose does discussing my fantasies have in this thread?
At what cost? That’s what determines if it’s a net positive or a net negative.
There’s a reason these features exist as plugins – precisely because it’s easier. It’s not impossible to implement support for DLSS or FSR without plugins, but then again, it’s not impossible to write a web browser from scratch using assembly language.
How would I know whether it’s worth Funcom’s time? I don’t know, but I can make an educated guess by looking at all that they’ve said so far, during their dev streams, the Reddit AMA, etc. So far, everything they’ve said points at their engine being complex and hairy. Like others already explained, it’s a heavily modified UE4.15, the key word here being “heavily”. There’s a good reason they haven’t upgraded the underlying UE any further. Ask yourself what it is.
While we’re on the topic of asking yourself some questions: how would you know that “it is not that much of the time needed after all”? Did they let you take a peek at their codebase?
Nobody said there wouldn’t be changes to the engine and nobody here claimed that, either. There won’t be further upgrades of the underlying UE, at least not according to what Funcom devs have said so far. Those new features have nothing to do with that.
And you won’t see one. It’s impossible to convince someone whose answer to everything is “yes, but”.
Not really, for reasons @TeleTesselator outlined, and a host of other reasons, too.
The topic name talks about DLSS. I didn’t know much about FSR (or cared, I’ve never bought an AMD GPU, didn’t see the point in researching it ESPECIALLY when the topic headline is DLSS)., and it wasn’t part of the original discussion. I put my 2 cents in about how I felt about DLSS. MY BAD, I guess?
I’ll be putting this discussion on mute, somehow I very much doubt that it’s as simple as you keep claiming it is.
This all seems like a good idea, but I’ve always experienced issues when pushing something through upscaling technologies, it just leads to further issues and displaced artefacts…but each to their own I guess!
It also names “Better Framerates”… But yea the two systems are VERY similar. The AMD one has a much better chance of long term survival as it’s open source and doesn’t require any specific hardware - it literally works on all GPS sold in at least the last 8 to 10 years. If you have a $2,000 video card DLSS can look better and even in exceptional cases better than the source - something untrue of FSR.
I don’t think it’s “bad”… It’s just, well, maybe like going into great detail about the taste of yogurt but only describing plain - err, no that analogy sucks… umm… well, it not “bad” is all. You probably know what I mean - just take it in the best way possible. It’s a compliment with caveats.
It is - you can trust me. I’ve worked with both. Of course so much always depends on the individual, but all things being equal…
Also from what I am reading you might be confusing DLSS with NIS. NIS is the Nvidia-Pendant to FSR. DLSS and especially 2.3 is really something else. NIS and FSR are just spatial upscaling and sharpening so in the best case the image gets unnoticeably worse. DLSS 2.3 can actually improve your image quality because it acts like an AA method as well - and NIS and FSR make Aliasing worse. NVIDIA and Epic are saying the image is better or equal 90% of the time to native, from my experience it is more like 50% but still. And we didn’t get at the point yet, what the Deep Learning in DLSS can accomplish - and what is the reason it only runs on RTX with tensor cores.
The good news is, today’s NIS is just as good as FSR and you can add it to your game (yes also CE) via the NVIDIA control panel or GeForce Experience. GeForce Experience automatically sets the resolution of all optimized games to the one set in the image scaler.
I myself wouldn’t use it though. You don’t buy an RTX 2080 ti to upscale images the old-fashioned way. You buy it to support modern tech like ray tracing and yes, DLSS.
I think a use-case for example could be where you have a 1060 and want to feed a UHD display. In this case, the NIS image should look way better than just setting the game to HD and blowing it up to fullscreen.
You may be surprised but I don’t do my “illustrious research” to annoy the people in the Funcom forums. If you don’t trust me, trust TeleTesselator, seriously.
And both of you, don’t take my sentence out of context, please.
Do you seriously think, it would not be “neat” if Funcom sometimes would consider/add something suggested by the community? And if it is just a bugfix? Or QoL improvement?
What @TeleTesselator said has nothing to do with what I said. Sure, DLSS and FSR SDKs might be easy to use. Adding logging to your program is also easy nowadays. There are tons of things in software development that are easy on their own, without any additional context. And for each one of those things, you could find a codebase where doing it is difficult, for a variety of reasons, but the most common one is that the codebase in quesiton is a hot mess.
No offense to @TeleTesselator or Funcom devs or anyone else, but I’ve long since lost my idealism when it comes to coding. Maybe that’s because 20+ years of doing it will give you a nice tour of all the haunting horrors you can find in the industry.
I’m not sure what I’m supposed to have taken out of context. If it’s the “wouldn’t it be neat” question, I didn’t take it out of its context, it lacked awareness of existing context:
Again, at what cost? No, I wouldn’t consider it neat if they spent time on this instead of things that I think matter more.
And yeah, that’s my own, personal opinion, so treat it as such instead of an argument. If you want an argument, here’s one: if they spend time on this, they’ll bring some limited framerate improvement to a certain portion of their PC users, whereas if they spend time on new game mechanics or on optimizing their netcode, they’ll bring improvements to all of their PC users and all of their console users.
This is a good argument - there are even plenty of things I myself would rather see changed/fixed.
Maybe I am too idealistic because I easily come out with any idea I find beneficial.
But I also think it is the role of the community to enjoy the game and give feedback and the role of the developer to sort their priorities and enhance the game (until it is time to move on).
Nah, we all post our ideas here. They all meet a lot of criticism, too. I generally try to be civil and fair with mine, but I also have a bit of a knee-jerk reaction whenever it looks like people aren’t taking seriously the complexity of game development. If my criticism was rude or too harsh, I apologize.
Funcom listens to feedback a lot more than people assume, so it’s definitely good to post it. As a consequence, a lot of people will chime in on other people’s feedback threads, because we know there’s a chance Funcom might listen and we all have our own ideas about where the game should go
It’s not just framerate. The ignorance in this thread can stack up to the moon, quite honestly.
I had an opportunity to put DLSS through its paces yesterday on a number of titles. Even if the engine supported it, this sort of integration, unfortunately, isn’t Funcom’s bag. Try playing a AAA title, you’ll notice all the creature comforts Funcom can’t, won’t or simply doesn’t think about implementing. It’s sad.
YOUR situation, though, should not require DLSS. Your card should be cranking the frames, and Funcom should address it.
Yup, all 100% true! Although just to edge it back my way a tad, both system affect such late stages in the pipeline it would have to be a horrifically hot mess to make either implementation too difficult or impossible - whatever “too difficult” may imply. That said, when I first came in to NewTek for an example, much of the Lightwave 3D code base was indeed a horrifically hot mess to be sure - so it might not be as uncommon as some including me, are thinking.
We always tend to think of programmers as a logical and competent bunch even giving them guru status in our minds but unfortunately the Pareto Distribution still applies and there are individuals we end up having to work around or even sacrificing what would otherwise be beautiful, for in almost every group development.
Well, all either of those systems do is add framerate. Both of them degrade the visuals in order to accomplish that. So, I’m not sure “neat” is the right adjective. If FC thinks that’s needed - like for example too many console users are getting really bad framerates and etc. then these are systems to consider. But if that’s not the case, then maybe Funcom’s response to such a “request” would be something along the lines of: “But why?”
It’s a compliment with caveats. 
