[Coco10]

i did some interesting comparison between GTX560 Ti clocked to 1Ghz VS GTX 470@875

here are the excat versions from newegg GTX560 Ti and GTX470 with ZALMAN VF3000F

my System

Q9550 @ 4.0 Ghz
GigaByte Ep45-DS4P
FSP Fx 600W
KINGSTON HyperX T1 4GB 2x2GB

Benchmark in

3Dmak Vantage
3Dmak 11
FarCry 2
Resedent Evil 5
Lost Planet 2
Tom Clancy's HAWX 2
S.T.A.L.K.E.R

i used two resolutions just check the pics

keep in mind GTX560 Ti @ 1Ghz equal to a stock GTX570 in most of the Games Check this Review here

Benchmark Results

3Dmark 11 GTX560 Ti

3Dmark 11 GTX470

3Dmark Vantage GTX560 Ti

3Dmark Vantage GTX470

,
,

,
,

,
,

,
,

S.T.A.L.K.E.R GTX560 Ti

S.T.A.L.K.E.R GTX470

when my ViewSonic VA2702w arrived i will re tested in higher res

hope this can help you guys

bbye

 

[TheRealHeavyG]

Benches at those resolutions don't tell me much.

 

[BallaTheFeared]

Quote:

Originally Posted by TheRealHeavyG;14704144
Benches at those resolutions don't tell me much.

I have to agree, outside the ones that aren't CPU limited there isn't much here.

Not to down play your effort OP, but a c2q at those resolutions with those cards is the bottleneck.

The 3dMark11 run shows some power differences with the 470 getting about 1k higher on the gpu score.

The only other bench I can think of for you to run where the c2q won't matter much is Heaven 2.5 with x8 aa 16x af and extreme tess.

 

[Booty Warrior]

Hmm, that is kind of interesting.

I'm surprised the 560 Ti won as many as it did, since the 470 is so heavily OC'd. The vantage results are surprising anyway.

+Rep for the effort at least.

 

[Coco10]

Quote:

Originally Posted by BallaTheFeared;14704280
I have to agree, outside the ones that aren't CPU limited there isn't much here.

Not to down play your effort OP, but a c2q at those resolutions with those cards is the bottleneck.

The 3dMark11 run shows some power differences with the 470 getting about 1k higher on the gpu score.

The only other bench I can think of for you to run where the c2q won't matter much is Heaven 2.5 with x8 aa 16x af and extreme tess.

Q9550@4.0 can hundle more than these resolutions check here
so i doubt it was a bottle neck

more over

a stock clock GTX560 Ti 822Mhz perform better like 10% more than Stock GTX470 in some games for exp Farcry 2 which i test it
gtx560 has 10 frames more than gtx470 look here
822 to 1 Ghz still good OC

other Exp which i tested
HAWX - 1680 x 1050 - Maximum Quality + Low AO + 4xAAFrames per Second - Higher is Better here
gtx560 ti@822mhz got 14 fps more than stock gtx470

the huge gap between gtx560 ti and gtx470 in 3dmark11 that because even at stock clock gtx470 beat gtx560ti here

so these results sound resonable is it ? lol

 

[neoroy]

Nice review but i still love my Zotac GTX470Amp! daily running at 860/1025 with 1.112v

produces GPU score 21.400 in Vantage ^_^ ofcourse with SB procie 2500k.

 

[BallaTheFeared]

Quote:

Originally Posted by Coco10;14704937
Q9550@4.0 can hundle more than these resolutions check here
so i doubt it was a bottle neck

more over

a stock clock GTX560 Ti 822Mhz perform better like 10% more than Stock GTX470 in some games for exp Farcry 2 which i test it
gtx560 has 10 frames more than gtx470 look here
822 to 1 Ghz still good OC

other Exp which i tested
HAWX - 1680 x 1050 - Maximum Quality + Low AO + 4xAAFrames per Second - Higher is Better here
gtx560 ti@822mhz got 14 fps more than stock gtx470

the huge gap between gtx560 ti and gtx470 in 3dmark11 that because even at stock clock gtx470 beat gtx560ti here

so these results sound resonable is it ? lol

Well let me get your settings and I'll run them on my system. I can already tell you in Vantage I beat your 470 score at the same clocks by 1000 points.

It's possible they're close, the 560 has better texture fill rate.

What were your advanced settings in CoP?

 

[mdatmo]

Quote:

Originally Posted by Coco10;14704937
Q9550@4.0 can hundle more than these resolutions check here
so i doubt it was a bottle neck

Those tests were run at 1920x1080. The lower the resolution the more important the CPU is because the GPU is stressed less due to the lower amount of pixels being rendered. Of course the amount varies with the test but it is a general run of thumb. Nice work though.

 

[Coco10]

Quote:

Originally Posted by BallaTheFeared;14705742
Well let me get your settings and I'll run them on my system. I can already tell you in Vantage I beat your 470 score at the same clocks by 1000 points.

It's possible they're close, the 560 has better texture fill rate.

What were your advanced settings in CoP?

advanced settings in CoP MSAA 4X the others are Off

 

[Razi3l]

Quote:

Originally Posted by BallaTheFeared;14705742
Well let me get your settings and I'll run them on my system. I can already tell you in Vantage I beat your 470 score at the same clocks by 1000 points.

It's possible they're close, the 560 has better texture fill rate.

What were your advanced settings in CoP?

Well the GTX 500 series seems to do much better in vantage compared to before, which is some sort of driver optimization or something.. I think Heaven 2.5 and Metro would be better tests, both on highest settings.

 

[BallaTheFeared]

I would say you're cpu limited.

CoP is one of the more demanding benches you ran graphically and my setup with the same gpu clocks scores 6-12 fps higher on average.

Switching over to something like LP2 at that res I get 79 fps in one test and 103 in another, thats with high dx11 features set.

In Far Cry 2 I averaged 106 fps in each test.

I won't go so far as to say the 560 ti isn't either, I'm just saying you are for both cards most likely.

You should enable tessellation whenever possible, it is a key feature of DX11 and is one aspect the 470 does noticeably better in than the 560 ti.

RE5:

 

[Coco10]

guys can some one of you apply these setting with the same Res and the rest of the setting in HIGH

and overclocked your GTX470 to 845Mhz and your CPU for 4.0Ghz and post it here For Test A and i will do the same to see howmuch different it will be

 

[Booty Warrior]

Quote:

Originally Posted by BallaTheFeared;14706690
You should enable tessellation whenever possible, it is a key feature of DX11 and is one aspect the 470 does noticeably better in than the 560 ti.

Well, yes and no.

According to the results in my Heaven benching thread, the GF100/110 cards were averaging ~32% perf loss vs 40% on the GF114 @1080p (interestingly, Drag and I found that the performance hit decreases substantially as resolution increases, but that's another topic).

That would mean, if the 470 and 560 Ti were clocked so that each managed 60fps in Heaven with no tessellation, extreme would knock the 470 down to 40.8fps and the 560 Ti down to 36, giving the 470 a ~13% performance lead.

However, it is important to note that that's a worst case scenario. No games use the Unigine engine or the extreme levels of tessellation in that bench atm, and at lower levels (normal tess) there wasn't nearly as much variation.

In benching actual DX11 games, the 560 Ti doesn't seem to lose ground compared to the chips with beefier polymorph engines.

Crysis 2 is probably the heaviest tessellated game released atm and Guru3d measured the performance hit from DX9-11 to be only 10-15% on Radeon cards, which take a beating in Heaven.
http://www.guru3d.com/article/crysis-2-dx11-vga-and-cpu-performance-benchmarks/4

 

[HeadlessKnight]

There is a CPU bottleneck on these results. at 1080p results will be more accurate.
Both GTX 470 & 560 Ti should be tested on similar setups, to get more accurate results. Testing for example a 470 on 2500K & 560 Ti on Q9550 will give unfair results. The GTX 470 should be 6% faster (at 875 MHz) vs. than the GTX 560 Ti at 1 GHz.

Thanks for the effort OP +Rep.

 

[HeadlessKnight]

Quote:

Originally Posted by Coco10;14706831
guys can some one of you apply these setting with the same Res and the rest of the setting in HIGH

and overclocked your GTX470 to 845Mhz and your CPU for 4.0Ghz and post it here For Test A and i will do the same to see howmuch different it will be

I would like to help you on this but I can't. since my 470 is a POS.
Waiting for some people who have good oc'ers to see the results.

 

[Coco10]

Guys if someone has bench these games with i7 2600k@4.0 and the other one got higher scores because he overclock@5.0ghz

Can we saw the first guy was bottle necked his gpu because he got lower scores ?

Can i say both gpu gtx560 and gtx470 will gain equaily more if the cpu was sandybridge@4.5ghz ?

 

[BallaTheFeared]

Quote:

Originally Posted by Booty Warrior Well, yes and no. According to the results in my Heaven benching thread, the GF100/110 cards were averaging ~32% perf loss vs 40% on the GF114 @1080p (interestingly, Drag and I found that the performance hit decreases substantially as resolution increases, but that's another topic). That would mean, if the 470 and 560 Ti were clocked so that each managed 60fps in Heaven with no tessellation, extreme would knock the 470 down to 40.8fps and the 560 Ti down to 36, giving the 470 a ~13% performance lead. However, it is important to note that that's a worst case scenario. No games use the Unigine engine or the extreme levels of tessellation in that bench atm, and at lower levels (normal tess) there wasn't nearly as much variation. In benching actual DX11 games, the 560 Ti doesn't seem to lose ground compared to the chips with beefier polymorph engines. Crysis 2 is probably the heaviest tessellated game released atm and Guru3d measured the performance hit from DX9-11 to be only 10-15% on Radeon cards, which take a beating in Heaven. http://www.guru3d.com/article/crysis...e-benchmarks/4

Yeah there isn't a huge difference between the 470 or the 560ti, the only real difference comes from the price which favors the 470 and the power draw with favors the 560ti.

560ti's lose the same amount of performance as 6950/6970s do from tessellation.