I haven't claimed that benchmarks are 100% accurate. Please give me reference for your 3% claim.
Multiple publications/sites, many test runs and Ryzens are consistently slower than Cores in gaming. If it was due to "real" variance we'd be seeing those CPUs leapfrogging each other from site to site or test to test. The consistent trend however is that Ryzens are slower. Even if I grant you that anything below 3% difference doesn't count (which I do not until you provide some reference for that 3% claim), we have instances where Cores' advantage goes beyond that. So we have a situation where a Core CPU is going to be as good or better than equivalent Ryzen. Sounds like and advantage to me.
I haven't said that doing multiple runs eliminates margin of error but hey, nice strawman.
If the 9900K (and 9700K for that matter) is above that margin at some games and never drops below then in fact it has an advantage over the 3900X. Moreover please explain to me why is it that somehow, that variance you speak of, allways works against Ryzens ? The Core CPUs have been consistently shown on multiple sites, test setups, test samples etc. to be faster in gimng than Ryzen CPUs. That's a fact. Which then leads to a natural conclusion that they have an advantage in gaming. Deal with it.
https://www.gamersnexus.net/guides/...ability-standard-deviation-of-game-benchmarks
You do not seem to understand what margin of error is. It doesn't matter if the i9 9900K is consistently 2.4% ahead in benchmarks, the tests being run simply do not have the resolution to declare one better then the other. CPU benchmarks are not accurate within 3% and the only thing that would change this is improvement to the testing methodology itself. More data does not equal better as it is all limited by how the testing was conducted.
Here's another review from GN that shows the margin of error bars
https://www.gamersnexus.net/hwreviews/3474-new-cpu-testing-methodology-2019-ryzen-3000-prep
"They’re all about the same. The top half of results – that’d be the i5-8600K through the i7-8700K – are all within margin of error"
Given that the 9900K is actually a bit worse then the 8700K, that's enough said on the subject.
A small sample size ? CSGO is one of the biggest eSport titles around. It's consistently in the top 5 (top3 more so than not) of concurent players on Steam. Sees milions of players each month (
https://blog.counter-strike.net/index.php/category/updates/) and yet that's a small sample, OK.
Oh please, could you possibly manufacture any more false outrage?
CS:GO is the only eSport where below 1080p is common and you tried to pass it off as if every pro player. This is just you creating a straw man (which you ironically accused me of) so that you don't have to answer why every other eSport has a standard resolution of 1080p as shown in the links I provided. FYI steam doesn't at all represent the whole PC scene, not at all.
Nice cherry pick btw. I said some play at 800x600 resolution to point how low some players go. I also said many play at 1024x768 but you have conveniently left that part out. In fact, only around 6% of CSGO pros play at resolutions 1080p or higher. This is as close to a strawman without being a strawman one can get. Nice job.
Now, let's check out your candidates. Let's see what % of players game on resolutions below 1080p. I have excluded players who have no resolution stated in the respective tables. We have: Fortnite - over 22%, PUBG - around 40%, Apex Legends - around 29%, R6 - around 8%, Overwatch - around 7%. Please tell me again about how small the sample size of pro players playing under 1080p is, please.
No you were not correct and you saying otherwise doesn't change anything. It's not semantics, it's reality and facts, some of which you have even provided. When you say: Pros play at 1080p and anything below is not realistic, the only thing one has to do is provide one pro player that does not and your point falls apart. Unfortunately for you, it's not just one pro player, it's a lot of them. Get your facts straight and maybe learn the meaning of semantics.
You cherry pick one game
I provide links to every other eSport with a standard res of 1080p
Yep I'm the one cherry picking
Now, let's check out your candidates. Let's see what % of players game on resolutions below 1080p. I have excluded players who have no resolution stated in the respective tables. We have: Fortnite - over 22%, PUBG - around 40%, Apex Legends - around 29%, R6 - around 8%, Overwatch - around 7%. Please tell me again about how small the sample size of pro players playing under 1080p is, please.
Um, I appreciate the evidence that supports my point. If you take the average, that's 21.4% of players who play below 1080p. Yes, my point that 1080p is the standard for eSports player is more valid now then ever.
No you were not correct and you saying otherwise doesn't change anything. It's not semantics, it's reality and facts, some of which you have even provided. When you say: Pros play at 1080p and anything below is not realistic, the only thing one has to do is provide one pro player that does not and your point falls apart. Unfortunately for you, it's not just one pro player, it's a lot of them. Get your facts straight and maybe learn the meaning of semantics.
Yep and my dog loves chocolate so I guess it's ok for all dogs to eat chocolate. You don't seem to realize how utterly illogical your statement is. By the way, in case you really don't know, don't feed your dog chocolate. My dog is a freak that can eat a whole pan of brownies and be A-OK but a single example or a few examples does not disprove the overwhelming amount of evidence that chocolate's effect on small animal's hearts. Your statement is exactly this, you are trying to disprove the vast majority with the minority.
You can disprove my statement with a single example? Where exactly did I say "all pros play at 1080p"? That's right, I didn't. At this point, when you read "pros play at 1080p" you assumed it meant "all pros". At this point you are arguing against what you assumed I said. The sentence "Pros play at 1080p" is pretty vague in and of itself. I didn't specify the quantity of "Pros" so the reader can assume any number. A normal person would assume "a majority" or "average" amount. Of course that would be the normal rationalization of a statement like that. This is the internet and you had to assume on the extreme side of the spectrum because you wanted to pick a fight. I'm not going to bother being explicit with my nouns on the internet. There will always be trolls who assume the irrational to try and prove their point. It's honestly more funny to watch them try and rationalize their extreme interpretation to a statement they made a bunch of assumptions on. Suffice it to say, if this were a conversation in real life you would have agreed with my statement long ago, most likely because people would think you need help if you are jumping to wild conclusions like that.