Band Camp statements: To state "But this one time I saw X doing X... so that justifies X" Refers to the Band camp statement.
Coined by Maxwell
The problem is the development time to do that probably isn't viable. I mean depending how rooted in it is you could be talking about remaking most of the game to change the engine.It would be really nice if the could rework there engine because that is the massive thing which limiting them to create something bigger.
Are you talking about Cyrodil? If you are one of the big issues there is Draw Calls.
If lyou ooked at your CPU usage for ESO you'd noticed that it burns through one of your CPU cores to almost maximum and touches maybe 2-3 others at a much lower level. This is because the OpenGL / DirectX 11 API the game uses is pretty much a single threaded operation with most of the draw calls being processed on this one core and limiting the amount of graphics surfaces / lights / people doing stuff you can see to a pretty fixed hard number something like 10/12 thousand surfaces. This doesn't go very far when you have huge battles with lots of action with hundreds and hundres of players around one objective.
The solution to this problem is to use Either DX12 on Windows and Vulkan on Mac or just Vulkan across the board. This will allow a massive expansion of the ability of ESO clients to process draw calls and reduce the lag / DCs and increase performance accross the board.
A good article on this is located here at Anandtech:
https://www.anandtech.com/show/11223/quick-look-vulkan-3dmark-api-overhead
On a standard benchmark test for GTX1060 and you have the following:
Millions of draw calls on GTX1060:
Vulkan: 26.4
DX12: 20
DX11 MT: 2.4
DX11 ST: 2.3
That's at least a 10 times improvement on drawing throughput on the same hardware simply by changing the API that's being used on a 4C/8T system. Would be interesting to see what happens on one of AMDs 8C/16T systems.
The long and the short of that is it would take a long time to reprogram the ESO clients to update them for this new API as well as testing it but it would result in vastly higher performance in environments where you have a LOT of draw calls happening (Cyrodil).
It would be really nice if the could rework there engine because that is the massive thing which limiting them to create something bigger.
The problem is the development time to do that probably isn't viable. I mean depending how rooted in it is you could be talking about remaking most of the game to change the engine.It would be really nice if the could rework there engine because that is the massive thing which limiting them to create something bigger.
Personally as the armchair dev that I am I think they should be looking at solutions like Brain in a Box which CCP/EVE considered. The idea of boxing up stats so the servers have less checks. Google a video if you're really interested I am pretty sure the presentations about bit are still around.
Last I heard that Dev had moved to Riot though which is why it never really made it into EVE.
Are you talking about Cyrodil? If you are one of the big issues there is Draw Calls.
If lyou ooked at your CPU usage for ESO you'd noticed that it burns through one of your CPU cores to almost maximum and touches maybe 2-3 others at a much lower level. This is because the OpenGL / DirectX 11 API the game uses is pretty much a single threaded operation with most of the draw calls being processed on this one core and limiting the amount of graphics surfaces / lights / people doing stuff you can see to a pretty fixed hard number something like 10/12 thousand surfaces. This doesn't go very far when you have huge battles with lots of action with hundreds and hundres of players around one objective.
The solution to this problem is to use Either DX12 on Windows and Vulkan on Mac or just Vulkan across the board. This will allow a massive expansion of the ability of ESO clients to process draw calls and reduce the lag / DCs and increase performance accross the board.
A good article on this is located here at Anandtech:
https://www.anandtech.com/show/11223/quick-look-vulkan-3dmark-api-overhead
On a standard benchmark test for GTX1060 and you have the following:
Millions of draw calls on GTX1060:
Vulkan: 26.4
DX12: 20
DX11 MT: 2.4
DX11 ST: 2.3
That's at least a 10 times improvement on drawing throughput on the same hardware simply by changing the API that's being used on a 4C/8T system. Would be interesting to see what happens on one of AMDs 8C/16T systems.
The long and the short of that is it would take a long time to reprogram the ESO clients to update them for this new API as well as testing it but it would result in vastly higher performance in environments where you have a LOT of draw calls happening (Cyrodil).
It would be really nice if the could rework there engine because that is the massive thing which limiting them to create something bigger.
Removing 32 bit support is the first step in the right direction, to focus on the capabilities of 64 bit systems.The problem is the development time to do that probably isn't viable. I mean depending how rooted in it is you could be talking about remaking most of the game to change the engine.It would be really nice if the could rework there engine because that is the massive thing which limiting them to create something bigger.
Personally as the armchair dev that I am I think they should be looking at solutions like Brain in a Box which CCP/EVE considered. The idea of boxing up stats so the servers have less checks. Google a video if you're really interested I am pretty sure the presentations about bit are still around.
Last I heard that Dev had moved to Riot though which is why it never really made it into EVE.
Well, the game engine is modular enough to support different backends; it currently supports at least OpenGL and D3D11 (though not simultaneously in any platform anymore).
It wouldn't be too hard to implement a new backend in Vulkan.
In fact, a "Vulkan" string as a backend option was datamined from the PTS .exe some time ago, so we may actually see a Vulkan renderer for ESO in the future.
Are you talking about Cyrodil? If you are one of the big issues there is Draw Calls.
If lyou ooked at your CPU usage for ESO you'd noticed that it burns through one of your CPU cores to almost maximum and touches maybe 2-3 others at a much lower level. This is because the OpenGL / DirectX 11 API the game uses is pretty much a single threaded operation with most of the draw calls being processed on this one core and limiting the amount of graphics surfaces / lights / people doing stuff you can see to a pretty fixed hard number something like 10/12 thousand surfaces. This doesn't go very far when you have huge battles with lots of action with hundreds and hundres of players around one objective.
The solution to this problem is to use Either DX12 on Windows and Vulkan on Mac or just Vulkan across the board. This will allow a massive expansion of the ability of ESO clients to process draw calls and reduce the lag / DCs and increase performance accross the board.
A good article on this is located here at Anandtech:
https://www.anandtech.com/show/11223/quick-look-vulkan-3dmark-api-overhead
On a standard benchmark test for GTX1060 and you have the following:
Millions of draw calls on GTX1060:
Vulkan: 26.4
DX12: 20
DX11 MT: 2.4
DX11 ST: 2.3
That's at least a 10 times improvement on drawing throughput on the same hardware simply by changing the API that's being used on a 4C/8T system. Would be interesting to see what happens on one of AMDs 8C/16T systems.
The long and the short of that is it would take a long time to reprogram the ESO clients to update them for this new API as well as testing it but it would result in vastly higher performance in environments where you have a LOT of draw calls happening (Cyrodil).
Band Camp statements: To state "But this one time I saw X doing X... so that justifies X" Refers to the Band camp statement.
Coined by Maxwell
Band Camp statements: To state "But this one time I saw X doing X... so that justifies X" Refers to the Band camp statement.
Coined by Maxwell
Are you talking about Cyrodil? If you are one of the big issues there is Draw Calls.
If lyou ooked at your CPU usage for ESO you'd noticed that it burns through one of your CPU cores to almost maximum and touches maybe 2-3 others at a much lower level. This is because the OpenGL / DirectX 11 API the game uses is pretty much a single threaded operation with most of the draw calls being processed on this one core and limiting the amount of graphics surfaces / lights / people doing stuff you can see to a pretty fixed hard number something like 10/12 thousand surfaces. This doesn't go very far when you have huge battles with lots of action with hundreds and hundres of players around one objective.
The solution to this problem is to use Either DX12 on Windows and Vulkan on Mac or just Vulkan across the board. This will allow a massive expansion of the ability of ESO clients to process draw calls and reduce the lag / DCs and increase performance accross the board.
A good article on this is located here at Anandtech:
https://www.anandtech.com/show/11223/quick-look-vulkan-3dmark-api-overhead
On a standard benchmark test for GTX1060 and you have the following:
Millions of draw calls on GTX1060:
Vulkan: 26.4
DX12: 20
DX11 MT: 2.4
DX11 ST: 2.3
That's at least a 10 times improvement on drawing throughput on the same hardware simply by changing the API that's being used on a 4C/8T system. Would be interesting to see what happens on one of AMDs 8C/16T systems.
The long and the short of that is it would take a long time to reprogram the ESO clients to update them for this new API as well as testing it but it would result in vastly higher performance in environments where you have a LOT of draw calls happening (Cyrodil).
The solution to this problem is to use Either DX12 on Windows and Vulkan on Mac or just Vulkan across the board.
Correct me if I'm wrong but no ESO live this year has mentioned any steps in how they intend to improve ESO, there has been no further testing mentioned with the intention of fixing or stabilizing PvP, and there has been no update to what changes will arrive to stabilize PvP.
It's getting out of hand and the lack of outrage is amazing especially with how long we've been placed on the back burner with lack of any form of updates such as even addressing the topic and giving detailed steps that have been tried thus far.
http://xboxdvr.com/gamer/Maxwell/video/44638373
Band Camp statements: To state "But this one time I saw X doing X... so that justifies X" Refers to the Band camp statement.
Coined by Maxwell