What are the changes in performance as per GPUs used in Total War: Shogun 2 - Fall of The Samurai

[S.I.K]

Hello friends, does anyone have notice what are the changes are been seen in performance as per GPUs used in game, I am trying to find out, but not getting exact thing which I am looking at it. So anyone among you have done and got some result then do tell me about it. Thank you.

[Shikamaru Nara]

Well I have done it and doing lot of Direct X 11 benchmarking, got the following performance impact in game:

Particularly with shadows its value to lower the quality, the big performance gain comes with 'medium' shadows, instead of un-checking 'soft shadows' In My Humble Opinion since it will give better performance and a better look.

[Badmannah]

Let me tell you that performance impact may differ with card model/generation and driver version. The CPU can score 29.7 fps in Shogun 2 CPU benchmark. Whereas ATI cards appear to have a great deal to get minor drop in performance in consider of shadow feature whereas medium to high only about 10% instead of 25% for Nvidia cards. The latest Nvidia 275.27 drivers , it’s beta was released with new GTX560 card and it will give a 5-10% additional performance and depending on settings and CPU speed as balance to 270.61 on a 560ti@1020Mhz on 1680x1050 resolution, 2500k@4.4Ghz, Windows 7 64-bit.

[DevenderD]

If you are going through above user method then I would suggest that you better do preferred settings which I have done it and it will be best. Let me tell you that I manually set the VRAM to 1.5 GB so that I can keep away from any downgrading. I have not gone through any troubles till now other than that it can guide to sever slowdowns, such that during texture swapping and unsteadiness in some situations/configurations.

[S.I.K]

When I have 30fps with no AA settings and I will prefer 4xMSAA, I will have 15fps and it doesn't matter if this is correct or not, that not my focus. In the event that I hold no AA with 30fps and pick MLAA I will have 24fps. In any case In the event that I have MLAA and I will prefer 4xMSAA I will have 18fps or 22.5fps-depends which number did you take as base number. So essentially, consistent with your estimations, depending on if I need more fps with 4xMSAA I ought to decide on MLAA first and afterward overhaul to 4xMSAA. It is possible that or your legitimacy is too interesting for me.

[Nabhoj]

What did you individuals run to get those qualities? Just campaign map, simply a combat map? A drop of 33% for 4xAA is not so unfavorable and 50% for 8 x. I feared it could be worse but those qualities are around the ordinary for amusements. MLAA however looks "blurry" contrasted with no AA or MSAA-IMO not worth utilizing. Side note on the VRAM regulation of yours: Just have a 1GB card and 1012 MB VRAM utilized for the event is for all intents and purpose 100%-the rest is the regular "overhead" required for portion tables and whatnot. So on ultra settings you probable at present have a VRAM "top" in place, without a doubt at present initiating FPS drops-or the recreation may computerized change down.

[Halle]

The built-in benchmark which is there is unfortunately does not incorporate with any battles. While my recreation settings score 37fps in the benchmark, around the same time as a quit for the day a great combat with dead figures, the fps heads off down to about 20fps. In the battle guide, I have in the middle of 30-45fps. I began the entire benchmarking practice being as how I would be unable to empower 4xMSAA with great framerate. Until I have found that the shadow portion prices so far appearance. Now I am in addition utilizing 4xSSAA (constrained with Nvidia control board) which basically dispenses with all shining at the edges of items.

[Daksha90]

You are right; the actual VRAM usage is probably above the 1024MB on my graphics card. Although I never found any difference in the benchmark between high and ultra textures. I guess it was just slightly above and since the 275.27 drivers, it actually shows 991MB memory usage during the benchmark. Maybe there was some optimization. I edited the VRAM configuration, setting 1.5GB of VRAM to avoid any downgrading (I checked for the log file during the benchmarks as well). Since then I have maybe played 15 hours with ultra textures and never had any problem. Maybe because I have 8GB of RAM and a fast SSD so that any texture swapping is 'relatively' fast.