Its commonly accepted that above a certain refresh rate threshold that the difference is un-perceivable to the naked human eye, as we cannot distinguish one frame as being separate from the one that succeeds it due to the lack of time between the transition. Basically, at such a high refresh rate the motion on screen is perceived by your mind as motion is in real life, instead of as a series of pictures that are being rapidly progressed through. If you really focus you can see one frame transition to the next at 30hz or 60hz, because the time between refreshes is long enough that your mind can take note of that frame and differentiate it from the next. At say 300hz, this would be impossible. Most people would say much lower than 300hz though, as many people cannot distinguish between even 120hz and 144hz. It is in fact possible though to the trained eye at these ranges, as I myself have seen when using my 1080p 144hz monitor when set to lower refresh rates.
For me, if its not all the way to 144hz and is at say, 100hz, I can distinguish it when playing a fast pace shooter like cs:go. Others are better than me at this, and many professional esports players have reported seeing differences going from even as high as 240hz to 144hz or 165hz. I remember even one of the first 165Hz 1440p monitor reviews came out and the reviewer noted that at 165Hz it was close enough to real life motion that to them they experience an uncanny valley affect of sorts, due to the fluidity of the motion.
I should also at this point note I was talking about monitor refresh rates when your original question was regarding fps. There is a difference between 350fps and 550fps, but its dumb. Basically at 550fps your getting a closer to "most up to speed" frame, because at 550fps the displayed frame that goes out on screen was rendered closer to the time the monitor refreshed. That is though to talk about milliseconds if not fractions of milliseconds of difference between when the frame would be starting to render at 550fps and 350fps. It in practice, makes literally no difference.
I did the math of the second per frame at 550fps and 350fps
at 550fps it is 0.00182 seconds per frame at 350fps it is 0.00286 seconds per frame
So yeah, if assuming both are going to output a frame at exactly the same time, the 550fps system would start rendering the frame 0.00104 seconds later, so it would have 0.00104 seconds more information than the system running at 350fps would have again assuming the frame output both finish at the same moment in time, but don't start at the same time. This in practice means absolutely nothing, as both systems probably aren't being fed new information by the server they're connected too often enough for there even to be a difference. Even assuming the server is ridiculous and non-realistic, the difference is probably millimeters at most if we're talking something like a CS:GO or other common shooter.
There is a difference between 350fps and 550fps, but its dumb.
Why do you go into the "looking at the 550fps with an naked eye" while whole purpose of such benchmark is to show computational power of CPU when DX11/12 fremowork is used?
By going into logic of "naked eye" to compare computational power of chips is like comparing Formula1 car with a semi truck in regards how many real alive horses can fit in them.
I was more addressing his claim that there was a perceivable (aka can be seen with naked eye) difference, which there isn't. Yes I understand that its to demonstrate a measurable performance difference (which does exist), but that wasn't addressed within my writeup because it wasn't the subject which I was discussing. I was discussing his specific wording that there is a "perceivable difference."
Let's be real here. People that are spending $300+ on a cpu aren't going to be gaming at 1080p for the most part. People that are spending $200-$250 on a gpu are probably going to be gaming at 1080p. So marketing the 480 as the "everyman" card makes sense. And marketing the R7s as "high end" cpus also makes sense.
It was an interesting perspective. My 8350 has gotten better over the years. There is an issue with things being optimized for Intel and Nvidia. Not too surprising since they hold the largest market share. We have an all new platform and architecture and stuff has been put in place to improve performance of Ryzen. It will take time. Ryzen has shown some pretty solid chops.
I like how its like: what? You have a setup where you benefit most from having a strong CPU? You shouldn't get a Ryzen, because Ryzen is for enthusiasts so they don't need a strong CPU.
Regardless of the gaming performance in contrast with Intel jay does have it right that Ryzen is still the best all-around value. But, I think that the complaints of some pure gamers are valid that it isn't the gaming monster that many wanted. Compared to the 6900k or 7700k, it (ryzen) isn't as good purely in gaming because ipc is still king, and has been king, and will probably still be king for the next couple of years. Hell I mean even still with DX12, thread 0 is by default taking more load than thread 1 and that continues down the chain. Its not perfect multithreading with evenly spread load, its just improved over DX11 which favored single threads to a greater degree to others than what is implemented in DX12. If your doing workstation tasks with your system as well as gaming, a Ryzen 7 1700 is easily recommendable. If your a pure gamer with only that want, than a 7700k is still your best bet. Although I'm sure the AMD fanboys will try to lynch me for saying it, and the Intel Fanboys for recommending a 1700.
I want the cores. I like the versatility and while most games don't do a good job advantage of them, doesn't mean games 6 years down the line they won't do a much better job.
Again: It is. I haven't seen a game running poorly. And that is not getting worse with better optimized everything coming. And I am saying that as someone who actually is gaming on a 144Hz 1080p screen.
Plus: Pure gamers care about games, not about numbers in the corner of the screen. It simply does not matter if that number shows dips down to 90 or 110 as long as the gameplay is smooth.
Is it valid to do these benchmarks? Yes, you can generate numbers with it. Do these numbers matter? No, they don't.
People are using very extreme ways to define how Ryzen 7 is supposed to be. I think the discussion in the community went a bit overboard and lost the point.
The Ryzen 7 was never meant to be a gamer's CPU. It is intended for someone that has a need for a very good mutli-threaded CPU with some very sufficient gaming performance but not needing it purely for gaming.
A purely gamer system makes more sense to use a Ryzen 5 (or an i5).
The question is not about R7 to being a gaming monster, but if it is better or worse to the i7 8 cores solution that also targets the same market, always in conjunction to the difference on the more professional applications.
To me Steve Burke from Gamers Nexus till now has made the best review. He did extended tests, confronted AMD directly about it etc. And i totally agree with his statements about Ryzen. Idk who AdoreTV is, but their statement about low res benchmarks being worthless, of course hits nor land nor sea. Still in my opinion Ryzen performs basiclly okay for gaming. Its not the best, but overall its reasonable for a new platform. Its great for productivity at its price point.
Not not really, i think that Ryzen is a great cpu for productivity workloads at its price point. And as far as gaming is concerned, it basiclly performs okay for a new platform. But its not the best choice for gamers mainlly, depending on each individual use case scenario ofc. But the same can be said on intel side of things, X99 vs Z270 debate for gamers. Never the less Ryzen was very hyped towards gamers and overclock enthusiasts by AMD, and i can see why its dissapointing for certain people.
The main things that dissapoints me personally about Ryzen is its overclock potential and X370 motherboard pricings. It also seems that iommu support is lacking atm. Never the less compaired to X99 you cannot deny that Ryzen offers great bang for buck.
First over clocker in a while that is not that great. Not one of you pointed out the fact that Adored tv appears to have relied upon on one set of benchmarks/ source for their assumptions :) We all know how bad that is ? From my perspective the ipc had to at least catch up to the I5's as something like the 8350 was limping a bit in modern games. From an Amd user perspective ? It is welcome.
Personally I hate that everybody is looking at just the gaming when talking about Ryzen that's not the only thing Ryzen was designed for imo. I love the idea of streaming on a 1700 and being able to have an affordable chip to allow me to produce content at the quality I desire. Also the productivity results of this thing are insane. Take that same 1700 with some ECC ram and you have as great entry level video editing rig
I dont know who adoredTV is, nor do i exally care. Allot of their video´s seem to have clickbaity titles, which tells me enough. I mean making statements that low res benchmarks to test and compair actual cpu performance for diffrent generations in gaming are worthless, is of course just a laugh.
He makes good stuff imo though he does release a dud video now and then. Before the ryzen release he was doing some amazing things with wafers and polaris really technical at the same time easy to understand for non techy