Author : Daeguen Lee

(Any action violating either CCL policy or copyright laws is strictly prohibited)





It seems clear that NVIDIA's coming two-GPU product, the TITAN-Z, will run at 705MHz at base and 875MHz at its boost state, according to several sources. Compared to my first assumption (http://iyd.kr/637), base clock slows down a little bit (711MHz -> 705MHz) while boost clock goes higher (811MHz -> 875MHz) by 7.8%. Since gaming performances in general stick together with GPU's boost clock rather than base clock, there's no doubt to believe that the performance of TITAN-Z is also increased by around 7%. Here's (maybe) the last speculation for its performance. Let's see:



Based on new boost profile, the TITAN-Z is anticipated to be 69% faster than original TITAN (not TITAN Black) and win current 2-GPU product, the GTX 690, by 32% or less. FYI, there's also attached a "speculative" performance for already-launched counterpart : Radeon R9 295X2.



The 295X2 gain the score 84% higher than that of 290X. By assuming that performances of GTX TITAN and R9 290X are roughly same, we can compare those two score -speculative performances of TITAN-Z and 295X2- directly : 169%p vs 184%p. In other words, coming TITAN-Z would be slower than existing 295X2 by more or less 8%.


Well, my part is finally over (for this generation). Let's just wait until the TITAN-Z is formally unveiled, and check whether my speculation is right or not. Stay tuned!


//

 

(아래 위젯은 티스토리의 크라우드펀딩 시스템인 '밀어주기' 위젯입니다. 100원부터 3000원까지의 범위 내에서 글쓴이에게 소액 기부가 가능합니다. 사견으로는 이러한 형태의 펀딩이야말로, 성공적으로 정착될 경우 이해관계자로부터 독립된 벤치마크가 지속가능해지는 원동력이 될 것이라 생각합니다. 제가 작성한 글이 후원할만한 가치가 있다고 여기신다면 밀어주기를 통한 후원을 부탁드립니다. 물론 글을 '가치있게' 쓰는 것은 오롯이 저의 몫이며, 설령 제 글이 '후원할 만큼 가치있게' 여겨지지는 못해 결과적으로 후원을 받지 못하더라도 그것이 독자 여러분의 잘못이 아니란 건 너무 당연해 굳이 언급할 필요도 없겠습니다. 저는 후원 여부와 관계없이 제 글을 읽어주시는 모든 독자분께 감사합니다.)