07-27-2013, 06:43 AM
nsahawks7 Wrote:Honestly, I don't see how that quote translates into that. Maybe it's just me, but it's not equivalent to saying "we are done building CPU's, we give up." Distinctly, they're different, marketing or not, it came from their mouth.
The rivalry will still be there, but AMD isn't focusing on Intel and how they can beat them anymore, they're doing their own thing.
Then who or what are they focusing on?
Intel is their only competitor in the cpu industry right now (where they make the vast majority of their revenue). If they don't compete with them they go bankrupt, it's as simple as that. As I said, it's fluff.
nsahawks7 Wrote:Also, scroll down. ALL the specs are listed there.
The shader throughput is not listed there. And it's the single most important spec. of a GPU.
nsahawks7 Wrote:It's to show you that synthetic benchmarks and specs vs specs don't always work out, just like synthetically, an FX-83XX should be better than an i7, but in reality, we both no that is not true...I think you misinterpreted what I was saying.
But I was never comparing the two synthetically nor did I even bring up their specs.
Also synthetic benchmarking has nothing to do with comparing specs. Synthetic benchmarking simply means your benchmarking application is meant solely for benchmarking, not an application that people use for another purpose.
nsahawks7 Wrote:I used synthetic benchmarks because you used synthetic theories to show that to compete with Haswell, AMD needs to do what you stated in the quote below, which is not true. Synthetically yes, if you use math and all that it will probably be pretty close. But it's not the case.
I didn't use synthetic benchmarks. I mentioned how much faster Intel cpus currently are in dolphin, which is not a synthetic application.
This is why I couldn't figure out why you were suddenly talking about synthetic benchmarks.
nsahawks7 Wrote:3. Explain to me how those benchmarks are terrible? They're real life situations, 7-Zip, Photoshop, Video editing, and then some synthetic benchmarks to add it in to show how your theory is not very accurate.
Again I never used synthetic benchmarks.
What I meant when I said "those benchmarks are terrible" was "the FX-9590 is doing terribly in those benchmarks". Hopefully that clearly things up.
nsahawks7 Wrote:Also, TrueCript was the only one on there that took real control of multicore prcoessing, the other didn't or it was minimal.
Not true at all. Nearly all of those applications have perfect or near perfect multithreaded scaling to 8 cores. Which is why they picked them.
x264, mediacoder, POV ray, 7-Zip, truecrypt, and WPrime all have near perfect scaling.
Notice that the FX-8350 performs twice as high as its quad core counterpart the FX-4300 in all of those tests. That could not happen without near perfect scaling.
CS6 and 3Dmark do not but CS6 has very good multithreaded scaling.
nsahawks7 Wrote:Yeah, I said that the 9XXX is an OC 8350 pretty much and I know the TDP is 200+W, but this is not about that. Overall, AMD is not far behind from intel.
They are, particularly in dolphin, which is what we're talking about since this is the dolphin-emu forums.
nsahawks7 Wrote:Single threading, intel kills no doubt, but when you look at the facts, this statement quoted below just doesn't match up.
If you don't quote the context of the sentence then sure. If you did quote the context you would realize that I was talking about dolphin as was Thorhian.
nsahawks Wrote:I'm not trying to stir an Intel v AMD flame thread, all I'm saying, people need to cut AMD some slack.
We're just making projections and wondering what the future of using AMD cpus for dolphin will be like. If things will stay the way they are and for how long. There's nothing wrong with that. This isn't really a flamewar so much as it is a discussion (mostly).
