PC Pimps should consider this option

H

Hollowprime

New member
Hello everyone, long time no see. I've been waiting for A21 to fall before I posted but as a small unity dev I came across this video. I know 7 days to die is made in Unity and that the devs have a hard time balancing the graphics of the game. It's only fair as unity does not offer great solutions to multiple moving entities. URP despite being more polished still lacks in terms of performance. The 2022 Unity version also has significant loading times when trying to compile or import code (over a minute importing packages!)

Have the devs considered migrating to another more powerful engine like unreal which can handle thousands of moving entities?

I personally wouldn't mind having a couple orders of magnitudes more zombies on screen without the fps dropping bellow 120 😛 :



 
Last edited by a moderator:
Kosmic Kerman said:
They have considered it for their next project. It won't be happening for 7dtd for obvious reasons. Changing the game engine is not a small task.
Click to expand...


I agree it's not a small task. But unity engine doesn't seem to progress as fast as unreal not to mention the IDE lags more with each version. I personally consider checking unreal 5 when I'm done with my demo.

 
JCrook1028 said:
Are you seriously suggesting at this late stage they switch engines? Are you daft?
Click to expand...
ya something like this seems to indicate they have never tried to port a game from one engine to another. This seems more like a troll post than an honest question, because if they really considered the work involved, they'd realize what a bad choice that.is so late into dev

 
JCrook1028 said:
Are you seriously suggesting at this late stage they switch engines? Are you daft?
Click to expand...


warmer said:
ya something like this seems to indicate they have never tried to port a game from one engine to another. This seems more like a troll post than an honest question, because if they really considered the work involved, they'd realize what a bad choice that.is so late into dev
Click to expand...


This is a very serious post. I'm merely asking a question if the devs considered it. I even provided an example of a massive performance improvement. You don't have to be negative about it, especially since the devs keep reworking the same components (namely graphics) .

 
Hollowprime said:
This is a very serious post. I'm merely asking a question if the devs considered it. I even provided an example of a massive performance improvement. You don't have to be negative about it, especially since the devs keep reworking the same components (namely graphics) .
Click to expand...
they have updated through several major unity  updates to keep current. This took them a TON of time. To switch engines isn't as simple as pasting your code into a new engine. This would likely break a ton of code due to the syntax differences and make it unplayable for quite a while with no ability to see if there was a genuine pay off for the switch. Not trying to be mean, but when people ask "why aren't you doing x?" it is a question that comes from a place of ignorance to the facts at hand. This is why I said what I said. If you knew that, asking this question would seem like a pretty impossible ask. This would completely halt further progress for 6 months or more. NO WAY they would do that.

 
Last edited by a moderator:
1 major example;

C#(unity) can automatically manage your memory for you

C++(Unreal) you have to manually manage that, and this would like require re-writing memory calls to include dumping it when it isn't needed. If you didn't you'd have memory issues the longer you played a session.

Most people finish the road trip in the car they are using at the beginning unless it completely breaks down.
What they learn during that road trip will inform them on how to prepare their vehicle best for the next one.

The framework they've made is incredible. That's worth moving over to a superior engine performance wise at the beginning of the process. 10 years of code in C# is not something I'd move over to C++. I think at this point, they want to finish it and take lessons learned to their next game. If Voxel support in Unreal came out 5 years ago, they probably would have switched, but not when you are almost at the finish line.

 
An apt analogy would be like switching surgeons in the middle of a complex, multi-hour operation without prior planning.  If the two surgeons aren't extremely familiar with each other's work then significant complications will arise.  Unity and Unreal are skilled surgeons in their own right and appear to do many of the same things, but the details of how they do nearly everything vary significantly.  The surgery would not be finished in the original time frame and there may be a need to swap in a new patient, too.  :p

 
warmer said:
1 major example;

C#(unity) can automatically manage your memory for you

C++(Unreal) you have to manually manage that, and this would like require re-writing memory calls to include dumping it when it isn't needed. If you didn't you'd have memory issues the longer you played a session.

Most people finish the road trip in the car they are using at the beginning unless it completely breaks down.
What they learn during that road trip will inform them on how to prepare their vehicle best for the next one.

The framework they've made is incredible. That's worth moving over to a superior engine performance wise at the beginning of the process. 10 years of code in C# is not something I'd move over to C++. I think at this point, they want to finish it and take lessons learned to their next game. If Voxel support in Unreal came out 5 years ago, they probably would have switched, but not when you are almost at the finish line.
Click to expand...


I agree with everything you said but I read Unreal supports GC too.

 
theFlu said:
It says "official" .. some could translate that to "likely the opposite in the next 6-9 months" by now ... :)
Click to expand...


No, that is the technical definition of "official" in the TFP playbook. ;)

However....in this case there is no way that they will switch to Unreal for 7 Days to Die. Not now, or in 6-9 months, or ever.

Now, 7DtD2? We'll see how they manage Unreal with their next game and that will probably determine if they stick to Unity for 7DtD2 or make the switch. But THIS game will remain Unity until the sun goes nova. (yeah, yeah...that'll happen moments after this game finally releases. I'm saying it before anyone else does)

 
Last edited by a moderator:
unholyjoe said:
to voxel or not to voxel.... hmmmm :)
Click to expand...


UE5 can do voxels, there is a plugin for that . Now i don´t know if you can create a fully destructible world with it though. So we might see part 2 using UE and having voxels.

@Roland Is it offical that 7D Part 2 is coming? 

 
Last edited by a moderator:
pApA^LeGBa said:
UE5 can do voxels, there is a plugin for that . Now i don´t know if you can create a fully destructible world with it though. So we might see part 2 using UE and having voxels.

@Roland Is it offical that 7D Part 2 is coming? 
Click to expand...


It is interesting that Madmole has said publicly that if he had it to do over he wouldn't have chosen fully destructible so it is very well within the realm of possibility that part 2 will use UE, utilize voxels, but not 100% through and through. But by time they cycle back to do part 2 tech advances might make things possible that currently aren't. They have definitely said they plan to do a direct sequel. They have a couple other projects they want to do before they return to Navezgane so it will be awhile.

theFlu said:
So, you're saying there's a chance? Wohoo! :)
Click to expand...


Simmer down Lloyd Christmas.

 
Last edited by a moderator:
You must log in or register to reply here.
Back
Top