

Moving This Game To A New Engine
#21
Posted 09 September 2016 - 12:11 AM
#22
Posted 09 September 2016 - 12:14 AM
MrVei, on 08 September 2016 - 07:04 PM, said:
Can you please share the source for these claims ?
I've never heard PGI claim to want to move to a different engine. All I've heard is their response to questions about a new engine, which is the usual "yes, one day"™ answer they give to anything the players want but which they have no plan to actually work on.
PGI want to build and sell Mech Packs. They do not want to spend time on anything which won't directly earn them new sales. They have demonstrated this time and again for years.
Edited by Appogee, 09 September 2016 - 12:14 AM.
#23
Posted 09 September 2016 - 12:15 AM
Not having full Dx11 hardware means that the computer that you are playing this on is very old.
#24
Posted 09 September 2016 - 12:22 AM
#25
Posted 09 September 2016 - 12:57 AM
#26
Posted 09 September 2016 - 02:47 AM
LordNothing, on 08 September 2016 - 09:28 PM, said:
that said, what exactly is wrong with the current engine (other than buggy dx11 support)? is pgi going to leverage the new engine to bring new features, or are they doing it just to say they did it, then offer to sell you a mech pack? bringing in a new engine just to maintain the status quo is silly.
Honestly, I do not think your crashing is anyhow related to DX11 itself, or you running MW:O utilizing it ( unless your HW specs are not sufficient . )
My Specs:
Corei5 2320 @3Ghz
Pallit GTX 750 TI 2GB
8MB DDR3 @665 Ghz
Internal SATA HDD
Lenovo OEM MoBo ( yes, I am running this on a 5 year old Lenovo Minitower...)
Win7 Home P.
I have been running MW:O DX11 for roughly 3 months now . Never experienced ANY "Crash" .
#27
Posted 09 September 2016 - 02:57 AM
El Bandito, on 08 September 2016 - 09:00 PM, said:
The question here is: Will moving to new engine solve anything? Would it suddenly make PGI become better at coding, for example.
it could, since most of the stuff which is tricky right now and they hacked into the current CryEngine would be done by crytek.
#29
Posted 09 September 2016 - 06:26 AM
Kotzi, on 09 September 2016 - 04:02 AM, said:
Basically everything server-related (complete hitreg code and other net code) was hacked together by PGI onto a non-released beta version of the cryengine 3. They never had any crytek guys looking over their code, they had no support whatsoever.
MWO practically looks like this under the hood

The bottom car being thier CryEngine 3 and the top being their hitreg/server-auth-code.
If PGI would upgrade to CryEngine 5, all that stuff would not be necessary anymore, most heavy lifting would be done by Crytek engineers, most bandaids would not be needed, it wouldnt be a hassle to get Inverse Kinematics and PSR to work with each other. Features which are currently a no-go because implementing it would be a hassle, would be much easier.
Get it now?
Edited by TexAce, 09 September 2016 - 06:26 AM.
#30
Posted 09 September 2016 - 06:28 AM
El Bandito, on 08 September 2016 - 09:00 PM, said:
The question here is: Will moving to new engine solve anything? Would it suddenly make PGI become better at coding, for example.
I love these kinds of comments - "spaghetti code", "bad devs" and so on. I work in the industry and most of the people I have met are very competent (even brilliant) developers. Doing large projects, like this one, takes a lot more than coding skill.
Of course, the forum has many, many more expert coders, such as yourself, that PGI could rely on to help them improve things.
TexAce, on 09 September 2016 - 06:26 AM, said:
Basically everything server-related (complete hitreg code and other net code) was hacked together by PGI onto a non-released beta version of the cryengine 3. They never had any crytek guys looking over their code, they had no support whatsoever.
MWO practically looks like this under the hood

The bottom car being thier CryEngine 3 and the top being their hitreg/server-auth-code.
If PGI would upgrade to CryEngine 5, all that stuff would not be necessary anymore, most heavy lifting would be done by Crytek engineers, most bandaids would not be needed, it wouldnt be a hassle to get Inverse Kinematics and PSR to work with each other. Features which are currently a no-go because implementing it would be a hassle, would be much easier.
Get it now?
So, they should stop making any progress on the current game, take their entire development staff and re-implement the entire game in an engine of your choice. And the dollars will just keep rolling in. Amirite?
[edit] I take it you've done a comprehensive code review of their source?
Edited by Zibmo, 09 September 2016 - 06:28 AM.
#31
Posted 09 September 2016 - 06:28 AM
#32
Posted 09 September 2016 - 06:31 AM
TexAce, on 09 September 2016 - 06:26 AM, said:
Basically everything server-related (complete hitreg code and other net code) was hacked together by PGI onto a non-released beta version of the cryengine 3. They never had any crytek guys looking over their code, they had no support whatsoever.
If PGI would upgrade to CryEngine 5, all that stuff would not be necessary anymore, most heavy lifting would be done by Crytek engineers, most bandaids would not be needed, it wouldnt be a hassle to get Inverse Kinematics and PSR to work with each other. Features which are currently a no-go because implementing it would be a hassle, would be much easier.
Get it now?
You know that letting Crytek do the work does cost money?
#33
Posted 09 September 2016 - 06:35 AM
El Bandito, on 08 September 2016 - 09:00 PM, said:
The question here is: Will moving to new engine solve anything? Would it suddenly make PGI become better at coding, for example.
As a developer myself I can tell you that engine choice does have a significant impact. Unreal is known for being very artist friendly which is why between it and Unity they are the first choice for Indie devs.
By making the development process more streamlined and easy to do they can spend less time writing around complicated systems and more time actually doing stuff. Unreal also has much more community support, meaning that when devs run into trouble the chances of finding someone who's already solved that problem are much higher. StackOverflow is a developers best friend.
At some point the time lost writing around bad code and bad design slows down development so much that it would actually be more efficient to re-write the entire code base and start from scratch. I think MWO is close to if not already at that point, as several easy to fix bugs have been consistently overlooked because "the engine isn't compatible" aka "we have no idea how our code works because the person who wrote it left and didn't document anything"
#34
Posted 09 September 2016 - 06:36 AM
Zibmo, on 09 September 2016 - 06:28 AM, said:
I love these kinds of comments - "spaghetti code", "bad devs" and so on. I work in the industry and most of the people I have met are very competent (even brilliant) developers. Doing large projects, like this one, takes a lot more than coding skill.
Of course, the forum has many, many more expert coders, such as yourself, that PGI could rely on to help them improve things.
So, they should stop making any progress on the current game, take their entire development staff and re-implement the entire game in an engine of your choice. And the dollars will just keep rolling in. Amirite?
[edit] I take it you've done a comprehensive code review of their source?
Never I've said they should upgrade to a specific engine, I just showed you how it looks in the insides.
There are also a LOT of information on the forum from the staff as well as in townhalls about that topic.
Fact is, it will be VERY hard for them if they decide to stick with the current engine they are using, since updating it with new features will get harder and harder.
Most of the devs who wrote the current build do not even work anymore @PGI.
Thats why they completely revamp certain parts of the game than to try to decipher the old code.
if they want to be alive the next 5 years, they will have to update their engine. They are free to call it MWO2 then, I don't care. I'm jut sayin.
Kotzi, on 09 September 2016 - 06:31 AM, said:
Pro-tip for life: Everything costs money. Being it a new engine license (actually CryEngine 5 is FOR FREE and Unreal is pretty cheap for MWO's population) or the developers trying to shoehorn new features into a 5 year old engine.
Edited by TexAce, 09 September 2016 - 06:37 AM.
#35
Posted 09 September 2016 - 06:36 AM
Quote
Unreal has worse map making tools than Cryengine. But MWO's maps suck anyway so its not like that would make a difference.
Unreal also has less realistic graphics and worse lighting than Cryengine, but again PGI doesnt push the limit of what cryengine can do graphically anyway, so it wouldnt make any difference
Unreal is better in most other ways.
Edited by Khobai, 09 September 2016 - 06:38 AM.
#36
Posted 09 September 2016 - 06:37 AM
Kotzi, on 09 September 2016 - 12:22 AM, said:
By all accounts CryEngine is an absolute ***** to work with. Sure CryTek can make it do all these amazing feats, but for mere mortals it's much more difficult. CryEngine is an absolute resource hog compared to Unreal
#37
Posted 09 September 2016 - 06:38 AM
El Bandito, on 08 September 2016 - 09:00 PM, said:
The question here is: Will moving to new engine solve anything? Would it suddenly make PGI become better at coding, for example.
well you can talk to their coders and see what they actually know. We'll respect them more if they got that Bandito seal of approval.
Khobai, on 09 September 2016 - 06:36 AM, said:
Unreal has worse map making tools than Cryengine. But MWO's maps suck anyway so its not like that would make a difference.
Unreal also has less realistic graphics and worse lighting than Cryengine, but again PGI doesnt push the limit of what cryengine can do graphically anyway, so it wouldnt make any difference
Unreal is better in most other ways.
to be fair, Polar highlands to me looks freaking amazing
#38
Posted 09 September 2016 - 06:40 AM
Quote
polar highlands is far from amazing compared to what cryengine is actually capable of.
#39
Posted 09 September 2016 - 06:40 AM
Zibmo, on 09 September 2016 - 06:28 AM, said:
[edit] I take it you've done a comprehensive code review of their source?
They should invest in a secondary team to start re-writing the game . With 3-4 competent devs it should only take a couple of months as the slowest parts of development (design and art) are already done.
It's not like the game is rapidly adding features constantly
Edited by Troutmonkey, 09 September 2016 - 06:42 AM.
#40
Posted 09 September 2016 - 06:42 AM
TexAce, on 09 September 2016 - 06:36 AM, said:
Never I've said they should upgrade to a specific engine, I just showed you how it looks in the insides.
There are also a LOT of information on the forum from the staff as well as in townhalls about that topic.
Fact is, it will be VERY hard for them if they decide to stick with the current engine they are using, since updating it with new features will get harder and harder.
Most of the devs who wrote the current build do not even work anymore @PGI.
Thats why they completely revamp certain parts of the game than to try to decipher the old code.
if they want to be alive the next 5 years, they will have to update their engine. They are free to call it MWO2 then, I don't care. I'm jut sayin.
Pro-tip for life: Everything costs money. Being it a new engine license (actually CryEngine 5 is FOR FREE and Unreal is pretty cheap for MWO's population) or the developers trying to shoehorn new features into a 5 year old engine.
Ye, of course Cryengine is for free even for commercial stuff and Crytek will not charge anything if their employees work for you. And PGI will not mind to get the most expensive Engine and pay a lot for support just to give you your best gaming experience.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users