robax91 :
Man this guy just doesn't get it. I kinda feel bad for him at this point, I'll just quote this little tidbit while I laugh to myself, no point in going on further after this gem.
"FPS is not a strong enough argument on it's own to determine the best CPU for gaming right now as the differences in any currently meaningful context are minimal." - Bigpinkdragon286
Lmao. Yes, yes it does.
Vitriol and condescension are not arguments and they invalidate my argument no more than they validate yours.
robax91 :
FPS is basically the only metric you can use to determine how well each CPU works with the same GPU.
Sub-second frame-rate has been used by benchmarking and review sites for many years now, and while similar to FPS, strictly speaking, it's not FPS. Frametime can give a far better picture of whether a game is smooth and playable or not, whereas FPS often masks those problems. If this wasn't the case, most people would still consider multi-GPU configurations superior, despite the problems they have with jitter and frame pacing. This leads to another measure, which is frame pacing, but I don't see too many sites looking into that.
When talking about FPS, it's more useful to avoid using strictly max and min averages. Not sure exactly what you're referring to as you just blurt out FPS, but 99th percentile, 1%, and 0.1% are better metrics than just raw max and min averages.
A perfect example is the old G3258 vs X4 760k. The G3258 would spit out higher FPS, but return an overall worse user experience due to it's inability to maintain a consistently fluid experience.
Consistent delivery of frames can be just as important to the fluidity of motion in games, if not more, as how many frames you're getting.
robax91 :
Your point is invalid about which GPUs you should pair with what CPU, that's not the point of the article.
I'm not suggesting which GPUs you should pair with what CPU, never was. I had some rhetoric in one of my replies, maybe you mean that? I also pointed out the most common GPU according to a common hardware survey, taken by many gamers the world over.
CPUs for gaming don't exist in a vacuum. It just so happens that you can't game on a CPU without a graphics device, and it just so happens that, according to the numbers, most of the graphics devices in use around the world are going to limit the FPS you can achieve with either a 2600X, 8600k, 2700X, and even an 8086k, meaning that once you are FPS limited by the GPU, you will want another metric for deciding which CPU to purchase within the pricing tiers.
robax91 :
Also, I see you have trouble figuring out how this article works. They have price tiers, and are supposed to show the best CPU for gaming in that price tier.
If you're going to make general assertions rather than point out specific instances where I am wrong, I'm going to ignore what you're saying.
You're the one that began comparing across price tiers, declaring a CPU from the $200 - $300 tier was better and cheaper than one from the $300+ tier. I just fleshed out your argument with actual numbers, and added the 2600X, since it actually occupies the $200 - $300 tier with the 8600k. My comparison of CPUs within a tier goes along with the premise of the article. Your comparison across tiers doesn't.
robax91 :
"The 8600k actually competes in the price segment that the 2600X competes in, and the 2600X pretty much ties the 2700X in general gaming performance, but for much less."
Yea, again you are dodging the point to try and win the argument, something people do when they know they can't win. The $300 suggestion for the 2700x isn't justified when compared to an i7, and my point, it's not even justified when compared to an i5, which beats all of Ryzen chips (including the 2600x). Your arguments only point back to my suggestion of changing the title to "best CPUs for the money (that can game)".
I'm dodging your point?
Maybe your point isn't clear?
I'm not even sure the point you're trying to get across when you say "The $300 suggestion for the 2700x isn't justified when compared to an i7...".
robax91 :
The selections here aren't the best gaming CPUs, they are the best value CPUs. Period. Now move on, before you repeat yourself yet again about Ryzen being the better value, which has nothing to do with the title of the article "Best Gaming CPUs". If it's really that hard to understand, have your parents sit down and try to explain it to you, because I don't have the time for that.
You can't game in any fashion where frame times are a meaningful measurement without a graphics device, therefore you can't have a best gaming comparison without some modicum of graphics device. You have to imply some sort of GPU in a best gaming decision.
Graphics devices impose limits on the number of frames you can output from a CPU in a given time frame based on the performance of the graphics device.
Because a graphics device must be implied, and that implied graphics device must have a particular performance level, a decision must be made as to what performance level to use.
If only 4% or less of GPUs currently being used are capable of demonstrating your FPS difference in a meaningful way, logically speaking, those probably aren't the right GPUs to be using the performance of when recommending CPUs for most people.
If you advocate for using the high end GPUs with market penetration below 4% to measure with, what is the reasoning there, to remove the GPU as the bottleneck? Once you remove the GPU as a bottleneck, you have reverted back to the best CPU comparison, the comparison without the gaming part.
If over 50% of the GPUs in use for gaming on a service such as Steam are at a GTX 1060 level or below, then that seems a much more logical performance level for the graphics device.
Once you normalize the FPS generated across differing CPUs by means of a GPU bottleneck, you have to look at other reasons to recommend a CPU, such as value.
Have a nice day.
Facebook
Twitter
Instagram