The settings have NOTHING to do with this.Elijah_Crow wrote: »Turn down your graphics settings and the issue will go away. Your trying to run on max with a card that can't handle it without those issues.
As of patch 2.3 (Thieves Guild DLC), it is no longer possible to play the game in D3D9, because this particular rendering path has been removed. D3D11 ist the only way now to play the game on Windows (barring a future addition of DirectX 12, ofc).I also noticed that you apparently can force D3D9 and D3D10 that way too, so you dont have to use D3D11. But you must set the file to read-only for this to work as well, otherwise the game just forces it back.
Interesting. Im wondering what the point of this line in the usersettings.txt is for then?KhajitFurTrader wrote: »As of patch 2.3 (Thieves Guild DLC), it is no longer possible to play the game in D3D9, because this particular rendering path has been removed. D3D11 ist the only way now to play the game on Windows (barring a future addition of DirectX 12, ofc).
The settings have NOTHING to do with this.Elijah_Crow wrote: »Turn down your graphics settings and the issue will go away. Your trying to run on max with a card that can't handle it without those issues.
Do you folks not understand the difference between a Mobile GPU and a Slot-Card GPU?
Total Polygons per second. Thats it.
Bus speed is irrelevant, hell even RAM Is typically irrelevant.
Changing resolutions and settings only increases frame rate which is irrelevant to the problem here.
Ive already tested multiple resolutions, and lowest settings.
That only results in me getting a whopping 90fps of mostly ugly, but does NOT resolve the delayed texture/mesh load.
Bro...the 470 was released in 2010.Elijah_Crow wrote: »I was playing with a GTX 470 and had the same issue you are having. With a GTX 970, it doesn't happen. I'm sorry if you don't like the answer, but there it is.
Ive noticed these "loading" screens too, which dont always occur and seem to vary from player/mob density.I would add that when I ignore these texture loads and don't stop and wait for them to finish I get thrown into a loading screen while the game loads them (therefore, it is not a random load screen). So, I either stop after every two dozens steps or I get a loading screen. This makes riding a full speed horse in Wayrest impossible (or sprinting longer distances). It takes a lot less time to load them second time (i.e. I travel to dungeon and then back to Wayrest).
The point was that back when there was a choice, it could be used to choose between different possibilities, e.g., D3D9, D3D11, or OPENGL. Potentially, when D3D12 will be added, it might become meaningful again. For now, it's deprecated. There are a few settings that have either become obsolete (e.g., through the use of newer technology), or will only be evaluated on the Windows client, but not on the Mac client, and vice versa.Interesting. Im wondering what the point of this line in the usersettings.txt is for then?
SET GraphicsDriver.7 "D3D11"
You can alter this line to "D3D9" or "D3D10", and as long as you set the file to read-only, ESO's exectuable cant alter it back (which it will do if its not read-only).
Whats the point of having the entry, or forcibly altering it back if changed, if its not relevant?
Bro...the 470 was released in 2010.Elijah_Crow wrote: »I was playing with a GTX 470 and had the same issue you are having. With a GTX 970, it doesn't happen. I'm sorry if you don't like the answer, but there it is.
Thats a 6 year old card.
I'd be surprised it can run the game at all, and is nearly a dinosaur by comparison to the 870M.
Thats not even a relevant comparison.
Why are you comparing apples to oranges?Elijah_Crow wrote: »[snip]The GTX 470 has very close performance to the card you are using and in some pass mark tests is actually better.
http://gpuboss.com/gpus/Nvidia-GeForce-GTX-470-vs-GeForce-GTX-870M
Your running basically the same as the card you just said was a dinosaur.
So, the assumption here is that Zenimax simply hasnt removed the relevance of that line in the usersettings.txt, nor the executable's specific instruction to attempt to ensure that it remains set as "D3D11".KhajitFurTrader wrote: »The point was that back when there was a choice, it could be used to choose between different possibilities, e.g., D3D9, D3D11, or OPENGL.
Why are you comparing apples to oranges?Elijah_Crow wrote: »[snip] The GTX 470 has very close performance to the card you are using and in some pass mark tests is actually better.
http://gpuboss.com/gpus/Nvidia-GeForce-GTX-470-vs-GeForce-GTX-870M
Your running basically the same as the card you just said was a dinosaur.
Throughput is the only difference between a mobile GPU and a Desktop Slot-card GPU.
[snip]
https://www.facebook.com/l.php?u=https://www.quora.com/What-is-the-difference-between-laptop-GPUs-and-desktop-GPUs&h=cAQEtbrPS
This is NOT an issue of framerate. Ive repeatedly said that multiple times.
If it was a performance issue, it would impact the entire render process equally, unless specifically coded to do otherwise via the game engine priority.
As far as what "equivalent" laptop GPUs have in them as opposed to desktop GPUs, they generally acheive approximately half the performance level of the equivalent desktop part.
Seriously? This is what the PS4 does? Well then, now we know for sure that it has nothing to do with the hardware of either system, and it is a specific issue pertaining to the Hero engine itself.Clarkieson wrote: »Ha! You have the ps4 version installed somehow. Welcome to console!
Babylon, what is your point?Regardless the guy basically just told you your laptop is half as good as a PC of same type (ie gtx970 vs your 970m), and you think that somehow supports your position that your potato can run game at max settings?
Doncellius wrote: »"The topic of this thread CAME out differently than I expected it would.... based on the title." -Guy who just had to say it
lordrichter wrote: »One comment... If you run a speed test to find your internet ping so that you can compare it to the latency number, pick a server in Frankfurt, Germany (EU) or Dallas, Texas (NA) to measure. If you post a screen shot, make sure it shows the server in Germany or Texas that you are using. The speed tests want to use the closest server. By selecting a server in the same area as the megaserver, you get a better picture of the route your ESO data has to take.
Be aware that the latency number displayed in the game and the internet ping response time are not the same thing. The ping time is the absolute fastest that you can get a response back from the ping test server. It is a measure of how slow your internet connection is to whatever data center that ping test server is in. Even if the ping test server is in the same city, it is not going to be exactly the same, but we hope it will be <10ms off.
While we do not know exactly what is included in the latency number provided by the client, I think that a ball park estimate of how long it takes the server to turn around a client request is going to be around the difference between the ping time and the latency time. My client reported latency is around 50-60ms larger than my ping to Dallas.
lordrichter wrote: »One comment... If you run a speed test to find your internet ping so that you can compare it to the latency number, pick a server in Frankfurt, Germany (EU) or Dallas, Texas (NA) to measure. If you post a screen shot, make sure it shows the server in Germany or Texas that you are using. The speed tests want to use the closest server. By selecting a server in the same area as the megaserver, you get a better picture of the route your ESO data has to take.
Be aware that the latency number displayed in the game and the internet ping response time are not the same thing. The ping time is the absolute fastest that you can get a response back from the ping test server. It is a measure of how slow your internet connection is to whatever data center that ping test server is in. Even if the ping test server is in the same city, it is not going to be exactly the same, but we hope it will be <10ms off.
While we do not know exactly what is included in the latency number provided by the client, I think that a ball park estimate of how long it takes the server to turn around a client request is going to be around the difference between the ping time and the latency time. My client reported latency is around 50-60ms larger than my ping to Dallas.
What does connection have to do with textures? I am 99% sure they are stored on your PC.
Most maxed settings players get 99 fps (max is 100 fps). So I think you're way overestimating your potato.
It is overall system issue. You know, PC is not only CPU, RAM, HDD and GPU - it;s also the motherboard. The MB connects them all and governs any process/communication that goes between those parts. So, I'm pretty much sure there's a bottleneck in Omnido system. If it's not the CPU, not the GPU, not SSD - it must be motherboard(maybe faulty SATA driver or slow CPU-RAM-GPU communication interface?)
I continue to notice this Loading effect...
Before:
5-10 seconds later...
Would be nice to know if this is even on their agenda to "fix".