Though I wasn't talking about feature sets like MMX or 3DNow. Even if Intel and AMD both make x86/x64 processors, the way they meet the instruction set requirements can be achieved using vastly different methods. Some use direct hardware, and the level of complexity determines how many clock cycles a result can be returned. Intel themselves have been using Microcode from a RISC based internal module since the days of the P6. It's those differences where compilers optimised for certain manufacturer hardware comes into account; It's still instruction code compliant with all compatible hardware, but performs better with specific hardware.
I just thought I would elaborate on those particular points... As whilst things like 3DNow! did gain traction for awhile, AMD's marketshare was never significant enough for it to gain industry wide support, hence why it was shuttled in the end.
As an example, Intel optimised x86 code can utilise answers from branch instructions quicker, thanks to their stupidly complex, yet surprisingly accurate branch predictor logic. Likewise, AMD has logic where code executed in a certain order can achieve a result quicker than Intel hardware. Methods like prefetching and ILP have also made a huge difference in optimisation considerations.
The same goes for IBM PowerPCs and Cell; both use the PowerPC instruction set, but achieved those outputs from different hardware methods.
Compilers optimised for certain hardware are hardly new. I started x86 programming in 1995, and I remember them being around then; Usually Intel, AMD and Cyrix.
I agree, but for simplicity sake, I try to keep the explanations simple so that other forum goers aren't lost.
AMD's branch predictors are probably rivaling Intels right now in Zen 2, Branch Prediction was always a massive strength of Intels going back to even Sandy Bridge, maybe even Core/Nahelem. - They invested a ton of effort to that end.
I have to admit, I never strictly "programmed" for the PC, I started on the Commodore 64 using Beginners All Purpose Symbolic Instruction Code... And tended to gravitate towards Objective C on the PC back in the 90's/early 2000's for it's ease of use. (Time is money and all that.)
Even assisted in a few projects like reverse engineering the shaders for Oblivion/Fallout 3/Bioshock to run on GPU's that those games were never designed to run on... Since haven't really done much, life changed course and priorities are different.
I would have no issue with anyone saying MS is aiming to have the most powerful or if that was their objective or roadmap.
But there is no legal way MS could be 100% sure as those people claimed. They can't make a contract with AMD with "assure XB4 HW is at least 10% more powerful than PS5" or discover the power of PS5 without spying. So there is no way MS could be certain. They could of course make a 800USD HW selling for 400USD that they think Sony wouldn't ever go above it, but nothing would stop Sony from doing it.
This. Until we actually have Microsoft and Sony outright state their hardware specifications, we honestly have no idea which platform is more powerful, we can always speak in hypothetical's and what-ifs, but it's not going to be productive either way, just the same old endless circle.
Microsoft and Sony wouldn't have any clue on what the other company is cooking up either.
Cheers. Hopefully it turns out to be true. Having feature set parity would mean big things.
Still not a source as legitimate as directly from AMD or Sony though.
At the end of the day when Sony approached AMD for building their next-gen chip, AMD would have given Sony all the same options it would have given Microsoft on the table, so we can assume they made the right choices.