By using this site, you agree to our Privacy Policy and our Terms of Use. Close
dahuman on 08 November 2013
Edit Thread
eyeofcore said:
dahuman said:
Espresso is actually not as weak as some people think it is and I never once thought it was. The 15GFlop estimate is based on estimation that's on outdated documentations as well. They already modified the 750 once to get better FP performance on the Gekko and nobody knows what they did with the Broadway or the Espresso afterwards. The only thing we still know is that the Wii U is incredibly efficient while not matching up in power to the X1 or PS4 is about it. We need more data than all this guess work bullshit.


15GFLOPS is based around outdated document that stated that Wii's CPU was 2.9GFLOPS or something...

I always asked if it is possible to improve performance of PowerPC750 cores while maintaining compatibility and people said not possible yet I don't agree since it is still basically PowerPC750 design. I wonder how much they could have improved the IPC from Wii's Broadway to Wii U's Espresso... 5-10%,15-20% or higher?

We will never know except if someone gets hands on development kits and then we would probably be under NDA.

Marcan42 on Twitter is sure it is regular PowerPC750, yet I beg differ.

It's possible to make an avaliable technology to be more efficient until they hit a theoretical cap in the design depdending on the process and material used. Marcan said it's 750 based, not that it's a regular one since we already know the Gekko wasn't a straight up 750 but a variant to start with so there is noway the Espresso is a straight up 750CL OC'ed in a tri core configuration, shit isn't that simple since they actually did R&D into the final hardware. What we need are benchmark tools that maybe a dev can run on Wii U systems to get a good average without leaking actual specs. Which I doubt they'll do in their spare time since they are not getting paid to do so.