gtx 560 vs 560ti real world difference?

C

c1001

Gawd
Joined
Oct 24, 2004
Messages
943
Anyone have any experience with a gtx 560 vs. gtx 560ti in gaming? i.e. will I see any noticeable difference in frame rates?

fyi, i am looking to get a second card for SLI and keep expenses down.
 
The GTX560Ti is 14% faster than the standard version on average.

Shogun 2: Total War: estimated frame rate 29 min. 35 ave. (GTX560 Ti) (DX10 mode, MLAA), 19 min. 25 ave. (DX10.1 mode, 8x MSAA), 17 min. 22 ave (DX11 mode, 8x MSAA) - assumes max detail

Metro 2033: estimated frame rate 65 min. 75 ave. (GTX560Ti) (Low detail DX10, AAA), 67 min. 63 ave. (Medium detail DX10, AAA), 42 min. 50 ave. (High detail DX10, AAA), 19 min. 35 ave. (Very High DX10, AAA), 13 min. 23 ave. (Very High DX11, AAA), 9 min. 17 ave. (Very High DX11, MSAA)

Deduct 12.5% from the Ti results for the standard version.
 
Or 10% faster depending on your source:

http://www.techpowerup.com/reviews/Palit/GeForce_GTX_560_2_GB/27.html

I tried a EVGA GTX 560 Ti 1GB, returned it and then got an EVGA GTX 560 SC (both 850 Mhz). I only game at 1680x1050 with a PII X4 955 @ 3.7 Ghz, but I couldn't tell any difference between the two (and was glad I saved the money). Just been playing RPGs such as Fallout 3 and New Vegas, Fable III, Witcher 2 and MMOs such as LotRO. I suppose with more demanding games and a higher resolution you could very well see a difference, but in my "real world" scenario I couldn't.
 
Those tests are crap, I ignore them. Subjectively assessing benchmark suites is much more scientific.
 
samuelmorris said:
Those tests are crap, I ignore them. Subjectively assessing benchmark suites is much more scientific.
Click to expand...

Sorry, the gentleman who runs that website is very thorough and well respected in his reviews. I tend to give more credence to his results than someone posting unsubstantiated results in a forum. At least he's not pulling assumptions out of the sky as to the percentage deficit that the GTX 560 would have vs. the Ti.
 
Last edited:
My 'assumption' is derived from real-world benchmark data, including some from TPU.
The problem with TPU is not their data and test methodology, it's those 'overall performance' charts at the end - TPU include a lot of tests which have very little relevance to modern hardware, and many are included in that chart at the end, which skews the results. People who rely on them are showing themselves as people who can't be bothered to read a benchmark properly, and have to be spoon-fed a summary, but still want some numbers in it.
 
You must log in or register to reply here.
Back
Top