jgonz
Can some one give RAGINGCANE a Blue Ribbon! Good work mate the community appreciates it.
 
Still rocking my original Kepler 680s here. Honestly no need to upgrade until we see what DX12 has to offer in the future.

I second the blue ribbon nomination. Those charts took a LOT of time and effort. Bene!
 
And true that jgonz. Currently the only thing that my SLI 780 Ti reference cards struggle with at max settings is the piece of un-optimized garbage known as Arkham Knight. (disclaimer - @ 1080P, but from 7 feet away with a 65" LED TV it looks grand)

arestavo
Wouldn't Icestorm be a bad test to compare high end graphics cards? Besides, it looks like you have two different systems (at least for the CPUs) - which cannot be compared by their overall score because of the physX test.
 
EDIT: Looking at the graphics scores alone tells the tale. The two 980's in SLI devastate the others.

Yes your right but that 980s in SLI at is 3dmarks official mark. against one 780 and two 760s that I did. I don't have 980s.  I was showing that to show how I cant even get the same scores anymore with the same setup backing the claims of the post.

UnReal-4-Life

arestavo
Wouldn't Icestorm be a bad test to compare high end graphics cards? Besides, it looks like you have two different systems (at least for the CPUs) - which cannot be compared by their overall score because of the physX test.
 
EDIT: Looking at the graphics scores alone tells the tale. The two 980's in SLI devastate the others.

Yes your right but that 980s in SLI at is 3dmarks official mark. against one 780 and two 760s that I did. I don't have 980s.  I was showing that to show how I cant even get the same scores anymore with the same setup backing the claims of the post.

I appreciate an attempt to help, but honestly that data is not repeated, nor are the testing conditions the same, therefore it has no intrinsic value nor can conclusions be drawn.

Despite all the work I have done thus far, my test conclusions can only essentially apply to 3DMark, despite it being a reliable gauge of performance.  That's why I am doing more testing, GTA V is next for example.

The fact that so many people have already made up their minds without numbers to back it up is a sad state of affairs. Facts are the bane of ignorance.

I've been running 340.52 flawlessly in BF4 since it's release a year ago.
 
Nice work btw, you deserve a ribbon!

I am currently using 350.12 drivers and playing GTA5.  I have crashed a few time with this error "Failed Intializaion. Please reboot and restart the game.  Try reinstalling the game if this persist.  
 
Is this a driver issue or a possible overclock issue???

bcavnaugh

ragingcain

UnReal-4-Life

arestavo
Wouldn't Icestorm be a bad test to compare high end graphics cards? Besides, it looks like you have two different systems (at least for the CPUs) - which cannot be compared by their overall score because of the physX test.
 
EDIT: Looking at the graphics scores alone tells the tale. The two 980's in SLI devastate the others.

Yes your right but that 980s in SLI at is 3dmarks official mark. against one 780 and two 760s that I did. I don't have 980s.  I was showing that to show how I cant even get the same scores anymore with the same setup backing the claims of the post.

I appreciate an attempt to help, but honestly that data is not repeated, nor are the testing conditions the same, therefore it has no intrinsic value nor can conclusions be drawn.

Despite all the work I have done thus far, my test conclusions can only essentially apply to 3DMark, despite it being a reliable gauge of performance.  That's why I am doing more testing, GTA V is next for example.

The fact that so many people have already made up their minds without numbers to back it up is a sad state of affairs. Facts are the bane of ignorance.

"The fact that so many people have already made up their minds without numbers to back it up is a sad state of affairs. Facts are the bane of ignorance."
 
You are also Not Adding in the Different Versions of 3DMark, Windows Patches/Updates, you are looking only at the Driver Versions.
 
"Facts are the bane of ignorance" You are Correct!

Why would he care about different versions of the bench? It's been 1.1 for over two years. Irrelevant for the claims being made regarding the 700 series performance.  Sure I guess he could generate a huge mess including the literally hundreds of Windows updates and patches which would vary per system.... but again, for the purposes here, what is the relevance? Since the claim IS the drivers gimping performance, it would seem most relevant to focus on the DRIVERS while leaving the other variables relatively the same.

bcavnaugh
3DMark v1.5.884
3DMark v1.5.893
3DMark v1.5.915 Current Version
 
Last know good version of NVIDIA Drivers - Unknown.

Minor updates, benchmark scores unaffected.  The actual Firestrike benches were version 1.0 until May of 2013 when they were updated to version 1.1. What do you mean by, 'last known good version of Nvidia drivers'. I don't get what your issues are doing a direct comparison of drivers since the release of Maxwell.