The TWC Total War Benchmark thread - compare your real TW performance

[Biggus Splenus]

The Tech Myths, Benchmarking and Preference.Script.txt pages have been linked in my thread now they should prove invaluable for system builders, new and old alike.

[alQamar]

thanks for your new results with optimsed settings for CPU and RAM crzyrndm added them.

Possibly because the GPU bench is a pre-battle cutscene with no blood involved. Having the setting on or off should have minimal/no impact until there is actually some combat (and therefore blood).

Also a quick note for anyone with borderline performance and a multiple monitor setup, just having a secondary screen rendering the windows background was costing me 2 fps in the CPU bench.

thanks. Well I don't know about the blood patch. But it is definitely something software based and not on the graphics card. I edited the reason for your improvements.

[Martin N]

My system is no where near some of those required to run the benchmark so i wont even try, but i just tested and can run two armies with 40 units of Yari ashigaru on S2 (160 men) on medium ish at 1280x1024 any chance of having similar results on R2? The map i tried was one of the very basic ones with just a few trees and hills, im thinking a big performance drop in sieges but i hope otherwise.

(I believe the blood patch was turned on too i dont know if this takes a little more resources)

[Dominian]

My system is no where near some of those required to run the benchmark so i wont even try, but i just tested and can run two armies with 40 units of Yari ashigaru on S2 (160 men) on medium ish at 1280x1024 any chance of having similar results on R2? The map i tried was one of the very basic ones with just a few trees and hills, im thinking a big performance drop in sieges but i hope otherwise.

(I believe the blood patch was turned on too i dont know if this takes a little more resources)

Any system that can run Shogun 2 is elligible for the benchmark. It it about gathering information about technology and it's capabilities. Of course people like me can boast here about their new systems, but in the end the real question is, where the cutoff point lies and for that mid and low-end systems are much more helpful. Cause in the end we all know that Rome II will perform swell on a Titan. Just run it and send your report in a PM to alQamar. Help him out!

[babydoc]

AMD partnered with SEGA and Creative Assembly on Total War: Shogun 2 through the AMD Gaming Evolved program to implement DirectX® 11 features into the game. Richard Gardner, the Lead Graphics Developer for the game discusses these features and how AMD effectively supported the Shogun 2 team.

And yet it plays better on Intel

[baldrick13]

And yet it plays better on Intel

it even advertised amd graphics cards in the recommended specs.

i still wonder how much the "improved" multi core support will change things

[alQamar]

And yet it plays better on Intel

it even advertised amd graphics cards in the recommended specs.

i still wonder how much the "improved" multi core support will change things

Apples and cherries. ..
They assisted them on GRAPHICS
NOT PROCESSORS.

The AMD graphics performance is very good!

In addition to my previous post:

It is really not my fault we have so less results for AMD based rigs.

The dx9 benchmark can be run on ANY rig with my preset but people aren't doing it. What can I say?

[k_161]

"Intel Core i7-3770K, 4,5 GHz, Ivy Bridge, 22nm lithography, 77 Watts TDP, DDR3 1333 MHz CL7-7-7-21, Dual Channel: 40 FPS"

Why am I only getting 29 FPS with my 3770k @ 4.5 : (

Edit: Ok I used your benchmark file for the CPU test and I got a score of 39 fps with a lazy overclock by just adjusting my turbo multiplier. What is the difference between the file you provided and the normal one ?

[alQamar]

Hi k161 Higher resolution and blood pack disabled. Can you please submit your logfiles incl cpuz log. Thank you very much.

[k_161]

Hi k161 Higher resolution and blood pack disabled. Can you please submit your logfiles incl cpuz log. Thank you very much.

Sure thing I've sent you an email. Thank you for the thread it was very informative

[alQamar]

here is the bios
http://www.gigabyte.com/products/pro...?pid=4481#bios
You have F3 newest is F7. copy the file on a usb stick and then update in the bios. If you are not sure how to setup the RAM speed with XMP or update bios please refer your manual.
It is not very complicated. Before this a resubmission of your results dont make much sense as we have not an optimal setup for your rig especially on the RAM.

When I got your results I will update the thread as I got also other submissions.

Thanks for everyone participating.

[djxput]

Just wanted to say alQamar I really like the thread here and cant wait for the new one based on Rome2. Also once the game comes out Im sure many people would be very interested in what settings one could change to produce better fps or better graphics without much of a fps hit.

for example; I plan to run rome2 at 2560x1440 and know Im going to have to limit the settings I use.

[alQamar]

Hello friends of Total War,
after getting some feedback my instructions may be to complicate for non tech persons, I've decided to record a step by step video and linked it in the OP aswell.
http://www.youtube.com/watch?v=9zzvWigd9rg

Sorry for any language issues in advance but I hope I helps you enough to make the benchmarks now without bother you to read my wall of text instructions. Please leave comments and reps if you like.

[A Barbarian]

@Nippy
Do the H100i fans properly spin up? If not then there might be a problem with the sensor (sometimes my board reports the wrong CPU temperature and doesn't speed up the fans properly). If they do, then it's most likely something wrong with the cooler mounting. With default voltage/speed and the H100i you should see temperatures around 70c (max 80c if you have a high voltage chip).

[Biggus Splenus]

Not Prime95 for Haswell its a bit buggy as of now, and has been reported to request more voltage than the max limit sometimes. This should be fixed soon apparently though. AIDA64 includes the new instruction sets introduced with Haswell, so that's probably the best stress tester you can get right now, for Haswell

Sent from my HTC One X using Tapatalk 4

[A Barbarian]

Not Prime95 for Haswell its a bit buggy as of now, and has been reported to request more voltage than the max limit sometimes.

That's not really correct. Prime95 has no bug and it can't 'request more voltage' from the CPU. What the newer versions do (and newer AIDA64 does the same) is to use special Haswell instruction sets like AVX (which is also used to a small extent in newer MP4 encoders). In combination with the adaptive voltage mechanism of Haswell CPUs (they raise voltage automatically above the baseline from the BIOS in certain situations) you get a higher voltage than what you configured.

Here is a quote from ASUS overclocking manual:

Additionally, Haswell has a more specialized point of consideration for synthetic stress tests. When using an adaptive VID, voltage levels will be automatically increased by the iVR when AVX instructions are detected from an application. An example is noted below:
A defined voltage of 1.245/1.250 would produce a CPU voltage of 1.245 or 1.250 under a normal realworld multi-thread application under load. Using the same defined voltage of 1.245/1.250 but running a synthetic stress test like Prime 95 with AVX, produces a CPU load voltage of 1.370 This increase must be considered as it could induce thermal throttle and overwhelm the cooling capabilities of the thermal solution being used.

[♔KillZoneGB♔]

Using your preference scripts CPU 35 FPS (rounded up) and Gpu score of 70 Fps (rounded down)

Spoiler Alert, click show to read: 

Attachment 291513

[alQamar]

Hi kill zone please attach the log files. Thanks a lot.

I just notice that for some reason people with low end or mid end desktop or laptop don't take part here. Once again THIS IS NOT thread for Overclockers

[babydoc]

More results make it easier for people to compare their scores and find whats hampering them.
Maybe you are not utilizing your hardware enough and could gain performance by some easy tweaks?

Send in more results.

[Epic]

AlQamar,

Below are the log files for my benchmarks using your settings.

Motherboard - Asus P8P67 LE
CPU - i5-2500k OC 4.6 GHz
RAM - Kingston HyperX 8 GB 1600
SSD - Samsung 830 series 256 GB
GPU - MSI Twin Frozr 7950 OC 1100/1450

CPU test - 40.7391
GPU test - 78.675

Overall I'm really impressed with my results, esp my CPU. It preformed better than the other i5-2500ks listed in the results.