All that is true , but it's not a hardware problem. It is a poorly written software problem. Patch or re-write sections of the software and the bottleneck disappears. One can , and many do , choose hardware based on what current software they plan on running. Personally , I can't justify spending $1,000 - $4,000 on hardware just for a couple of $60 games I might not be playing a month or three down the road when the next "Must Have" title comes out.Dribble said:... but the performance wouldn't change because that maxed out thread is still maxed out and each individual core doesn't run any faster.
That's the problem with the "more cores is always better" approach. You will always end up with a bottle neck on one thread. As long as you have enough cores so that thread gets to run continuously (i.e it gets a dedicated core, although you can't tell in task manager because windows keeps switching which one it's using) then the only way to improve performance is higher single thread performance.
You don't know you're missing something until you try something better. The 6700k and 7700k probably feel smoother than an older i5. When I switched from a 5400rpm HDD to a 7200rpm drive, I could tell a difference. Does that mean that when people changed out their 7200rpm drive for an SSD they shouldn't have been able to tell a difference because they didn't feel their HDD was slow to begin with?
You make some bold claims against respected reviewers and forum members. Maybe you should find a buddy with a Ryzen rig and try it for yourself.
You say that, but you can't just throw lots of threads at everything efficiently. You can easily get a massive increase in coding complexity, and a huge decrease in efficiency. This is due to you having to do lots of copying of data for the separate threads, locking, merging data back afterwards. Hence it might well be say for example the renderer using two threads goes slower then the single threaded one, and it only gets faster when you are using 3+ threads and twice the memory.All that is true , but it's not a hardware problem. It is a poorly written software problem. Patch or re-write sections of the software and the bottleneck disappears. One can , and many do , choose hardware based on what current software they plan on running. Personally , I can't justify spending $1,000 - $4,000 on hardware just for a couple of $60 games I might not be playing a month or three down the road when the next "Must Have" title comes out.
This is exactly why no one mentioned issues with the Intel chips before. You need a new experience in order to form a new view. Relativity is how you can distinguish where something sits. Whoever mentioned confirmation bias, that cuts both ways. If I just bought a new rig and it's having minor stutters or hiccups in some scenes of a game I'm not likely going to assume it's my CPU. I will accept the issue because it's highly unlikely I'd be willing or able to purchase new gear just to find out if there is a better experience. There isn't a choice to do anything else.
We have benchmarks showing i7-7700K ahead in both games you mentioned in frame times (smoother in multiple recent/older games according to Tech Report and others).
That's why some testing, or blind testing needs to be done, but it shouldn't just be between the i7 7700K and Ryzen, there needs to be an i7 6900K too. Then we can get to the bottom of this.It very hard to disprove a persons opinion. and that's what this is, a handfull of people's opinions. There are a thousand variables. My guess it will never be proven, just repeated by AMD marketing.
10 years from now we will have 8k TV's and somehow people will see the difference between that and a 5k TV. impossible to see the difference.
We will have 4k @ 200fps and people will say its smoother than 140fps. Impossible to tell the difference.
If I told you I can tell the difference between 13ms frametimes and 18 ms frametimes , I would be a liar.
It is true that most people use balanced mode. I do as well. You get 99% of the performance, without your CPU running at full clocks at the desktop, when doing almost nothing. Once in a game, it should be nearly full clocks all the time, if not, something needs to be fixed with Ryzen.I don't think you bothered to read how the Tech Report tested their Ryzen system at launch. If you're going to show graphs from them at least let everyone know that they did not use High Performance power mode when testing the Ryzen system http://techreport.com/review/31366/amd-ryzen-7-1800x-ryzen-7-1700x-and-ryzen-7-1700-cpus-reviewed/3
"After consulting with our readers, we've decided to enable Windows' "Balanced" power profile for the bulk of our desktop processor tests, which means power-saving features like SpeedStep and Cool'n'Quiet are operating. (In the past, we only enabled these features for power consumption testing.) Our spot checks demonstrated to us that, typically, there's no performance penalty for enabling these features on today's CPUs. If there is a real-world penalty to enabling these features, well, we think that's worthy of inclusion in our measurements, since the vast majority of desktop processors these days will spend their lives with these features enabled."
Heck, even their 6700K and 7700K CPUs are showing that they are running at 3866 DDR4 speed vs. Ryzen at 2933 with the same memory. This was at launch, I'm sure with all the BIOS updates and tweaks now the Ryzen system will only perform better now in gaming as we have seen with higher memory speeds achieved nowadays.
Imo it doesnt come more trustworthy than computerbase and personal experience from your fellow anandtech forum guys.
see, here is where you and like minded are wrong. when you use higher resolution and crank up the details, the GPU needs more data to carry out its magic. and it is the CPU who has to preprocess it and move the data around. so your assumption that GPU usage will scale linearly while the with resolution while CPU usage stayes the same is just unfounded. you can test it out your self. turn on V
The only way to look at CPU bottleneck is to look at the minimum fps, while analysing the CPU usage at the same time (it should be 100%)... remember a lack of optimization does not equate to CPU bottleneck
I dont know because i never participated in those threads because bd was crap for me and i can only think of one or two who could claim it back then. Certainly not the +30 people who have already bought a ryzen in this cpu forum. I havnt come across a single person here who upgraded from a slow singlethread cpu like 8350. Though by chance surely there is an outlayer.Fellow AT forum guys are not exactly trustworthy, some are, some aren't. Where do you think "FX8 feels smoother" came from?
I dont know because i never participated in those threads because bd was crap for me and i can only think of one or two who could claim it back then. Certainly not the +30 people who have already bought a ryzen in this cpu forum. I havnt come across a single person here who upgraded from a slow singlethread cpu like 8350. Though by chance surely there is an outlayer.
Peoples opinion is interesting because there is data that backs it up. From eg extremely strong throughput in eg cb15 or frametimes in some situations. Non of that was the case when bd was hot for a handfull of guys.
And then its worth remembering a 8350 is faster than a 2500k today in the newest games. Who would have thought that? Lol. Now look at a Ryzen. Its like from another planet.
dx 9 & 720p benchmarks are important, believe me! I cannot tell you why, just believe me! Fallout4 and arma3 are incredibly relevant and need to be benchmarked 24/7, because reasons.And i have to say its utterly pathetic that nearly all reviewers have conveniently taken the 2500k out of their suite and replaced it with the 2600 as the low end. Way to go. And then claim 7700k is a gamers cpu with a suite with eg arma 3 and other drawcall limited stuff from the dx9 age.
I say bs.
It is true that most people use balanced mode. I do as well. You get 99% of the performance, without your CPU running at full clocks at the desktop, when doing almost nothing. Once in a game, it should be nearly full clocks all the time, if not, something needs to be fixed with Ryzen.
My 1700 idles at 59w at desktop even with High Performance mode enabled. Same as my 4790K in balanced mode.
So the 6 core chips may not run this as well as people thought, relative to the 4 and 8 core chips?AotS automatically reduces details in-game (not in benchmark) on systems with fewer than 6 physical cores, no matter what settings you choose.
The 7700K is running at gimped details, so of course it will run faster than it "should" relative to Intel i7-6900K and R7 1800X.
Source:
http://www.pcgameshardware.de/Ryzen...ecials/AMD-AotS-Patch-Test-Benchmark-1224503/
Anything higher than 4 cores 8 Threads is affected. Or to be more precise, anything lower than 6C is given an advantage by the game via reduced image quality. That is why 7700K is performing better than any 6C+ intel or AMD SKU.So the 6 core chips may not run this as well as people thought, relative to the 4 and 8 core chips?
Hate to dis-illusion you, but these forums are the *last* place I come for objectivity, especially in the past few years.Imo it doesnt come more trustworthy than computerbase and personal experience from your fellow anandtech forum guys.
Benchmarking mp consistent in a game is next to impossible. You need to be into the game to know where the problems is. Like this German guy who reviewed bf1. But then you get a conquest bm because thats what he game lol and operations take to much time...so its not easy.
I think the gamers need to learn and have the tools to benchmark themselves. The most needed competence is actually knowledge of the game.
dx 9 & 720p benchmarks are important, believe me! I cannot tell you why, just believe me! Fallout4 and arma3 are incredibly relevant and need to be benchmarked 24/7, because reasons.
lol yah rightBecause Fallout 4 and Arma 3 are popular games with respectable player counts? Not hard...
http://steamcharts.com/toplol yah right