Nano better than 1070?

[ishoeyi]

just installed a new 1070 a few hours ago

I noticed I was a few fps short from the same game with a nano

pc 1
i5 6400
16gb ddr4 2400
1070 FE

pc2
xeno x5650
24gb ddr3 1600
nano

I load up metro 2033 redux
test both at exact location save point
same settings
nano giving a solid 60fps
1070 giving about 55fps
at 4k resolution

I haven't tested all the games yet
but I was abit shocked I expected the 1070 to be a faster card
or is the i5 6400 to blame but 1070 is running at full load
or is this the expected result?

 

[erazor51]

Run both graphics cards on the same rig to get right results. Two different machines won't give you right results.

 

[ishoeyi]

so there is a possibility that the rest of the system is the weak point?
I can't be bother changing about
just wondering if this is expected

my i5 6400 running at stock
the x5650 oc from 2.6 to 3.2

 

[erazor51]

For gaming the i5 should be better. Test other games to see what you get. If it still underperforms then it might be a configuration issue.

 

[ishoeyi]

ok I just figured it out

I forced advanced PhysX off thru config file

now I got solid 60fps

and gpu load dropped from 100% to 70%

unbelievable
can't believe PhysX can tank the card so much
I don't have much experience with NVidia card
mostly been with the red team

 

[kaku]

PhysX can be pretty awful. I'm not sure if it's just poor implementation rather than the fault of the library though.

For a good example, try Borderlands 2 and set the physics detail up. FPS will tank like crazy when you kill an enemy that leaves PhysX effects behind despite it being an old game.

 

[Blackjack Davy]

PhysX can be pretty awful. I'm not sure if it's just poor implementation rather than the fault of the library though.

For a good example, try Borderlands 2 and set the physics detail up. FPS will tank like crazy when you kill an enemy that leaves PhysX effects behind despite it being an old game.

Like their other Gimpworks offerings then? Using God Rays in FO4 at anything other than the lowest settings will tank your fps unless you have a lot of grunt to spare with very little to show for it.

 

[Rroff]

Metro uses quite a lot of physics effects so you'd be far from comparing like for like in terms of visual effects.

 

[Nasher]

Physx was good when they actually used the physx hardware, but now everyone just seems to use the software implementation which isn't half as good :/

 

[davewhite04]

just installed a new 1070 a few hours ago

I noticed I was a few fps short from the same game with a nano

pc 1
i5 6400
16gb ddr4 2400
1070 FE

pc2
xeno x5650
24gb ddr3 1600
nano

I load up metro 2033 redux
test both at exact location save point
same settings
nano giving a solid 60fps
1070 giving about 55fps
at 4k resolution

I haven't tested all the games yet
but I was abit shocked I expected the 1070 to be a faster card
or is the i5 6400 to blame but 1070 is running at full load
or is this the expected result?

Are you running that game 4k on ultra settings? I have to turn off MSAA(I think) to get 60FPS on 3440x1440 using a 1070.