Originally Posted by SpeedyVT
Oddly enough there are still a few things Bulldozer is still doing better than the last few generation i7s, not defending AMD. I know their performance in general is subpar, but this is fact. Not all processors are made equal. I think Zen will still be weaker per core performance in an HT disabled environment to the i7 counter part, however with the scalability research done on their FX series and APUs it's quite possible that some of the CMT technology has made it's way into the SMT especially since they did experiment on dual schedulers on the Carrizo chips.
Well, interesting. What you’re seeing with Zen is its versatility. What is Cinebench trying to represent, or the benchmark that we showed today? They show off—we show off a number of multithreaded applications. And you saw that we’ve done a true simultaneous multithreaded implementation. That really helps double the effective cores for those applications.
But we did it—and I mentioned this in the presentation—by increasing the resources in that execution pipeline. So if you are running single-thread, you get the benefit of these additional resources. It’s a versatile core; it’s going to play well to single-threaded and multithreaded applications.
Interestingly enough, this quote implies they would do better relatively in single threaded applications. They have a lot of extra resources dedicated for multithreading, but those same resources can still be used in single threaded scenarios, just not with the same efficiency as full on SMT.
Originally Posted by black96ws6
I remember when Bulldozer came out, there were some things (very few) it did better than a 2500k, based on multi-threading and perhaps special insruction sets, I can't remember it's been too long now.
Anyway, Zen may very well be faster or equal to Broadwell in this one specific instance, just like Bulldozer vs Sandy was in those 1 or 2 instances.
But we all know how that turned out...
Not the same scenario. Bulldozer had "8" cores, running against Intels 4 core 8 threaded processors. Their per core resources were quite different, and the core counts were different.
In this scenario we're talking same configurations and same clock speeds. Blender is also a fairly well rounded application in terms of demand from the CPU, as it tests many different parts of the CPU.
It's as apples to apples as it gets.Edited by lolerk52 - 8/19/16 at 12:18pm