I'm not sure what you're looking at, because
I've referenced that GamersNexus review many times, and just took another look at it to be sure, and the benchmarks for the 9700k at stock settings show it usually nudging out the 8700k with a 5Ghz OC, in just about everything. Considering that's with no OC on the 9700k, and we ARE talking about 1% and .1% low benchmarks which are what typically matter the most because that's where trouble tends to develop in terms of visible stuttering or lag, I'm not sure how you could even consider the 8700k except as a better alternative in terms of price, only, and we're talking a five dollar difference, so.....
View: https://www.youtube.com/watch?v=uXepIWi4SgM
There ARE a few instances where the overclocked 8700k has a better .1/1% score but they are few AND in almost every case, a similar OC on the 9700k wipes out even those few results.
Plus, AND MORE IMPORTANTLY, that review was done BEFORE the patches and mitigations for all of the various Spectre and Meltdown vulnerabilities, so you can be sure that there would be an even bigger amount of disparity in those comparisons now, than there was then, because the hyperthreading performance has been hampered to varying degrees (depends on who you ask, as to how much we feel this has actually had an effect on performance but where OTHER processes are involved that are not strictly gaming, we can assume the hit on performance is likely to be higher than on "just" gaming) and since the 8700k has two fewer actual physical cores then anything that hampers the hyperthreading performance is going to have a decent negative effect on those CPUs.
In some cases on the 8700k, as much as a 24% impact on performance.
https://www.extremetech.com/computing/291649-intel-performance-amd-spectre-meltdown-mds-patches
So unless you can get a PRE-patch era motherboard that has a BIOS from before these began going into effect, and can install an old enough build version of Windows that does not include any of the cumulative mitigations, it would be pretty hard to get the kind of performance from the 8700k that you see in the GamersNexus review, which occurred prior to any of these BIOS revisions or software patches happening.
Right now, I wouldn't even spend any extra money on an Intel CPU simply to gain hyperthreads. It's the main reason I haven't done a platform upgrade yet, at this point, because I'd LIKE to wait and see if Intel at some point offers a 10 or 12 core no-hyperthreading model with some optimizations. If that doesn't happen in the next 12 months or so, and it probably won't, then I'll likely either upgrade my 6700k to a 9700k or a 3700x/3900x, unless something else, more appealing, comes along before then. I'm ready for one now, because I can definitely tell that my performance is not as good as it was prior to those mitigations AND because being five years old with only four physical cores, I could be doing a lot better especially in the area of heavy multitasking, than I am now.
As far as YOUR performance expectations, yes, I believe you would likely see a pretty fair jump in performance, even WITH the mitigations and patches, from what you have now to either of these choices. I just believe your better option is the 9700k. The six cores on the 8700k plus it's hamstrung hyperthreading, doesn't seem to overcome the performance of the all physical 8 core 9700k, regardless of whether you overclock the 8700k or not.
If there is something out there that refutes any of this, and clearly demonstrates that this is not the case, I certainly haven't seen or heard about it. Anything is possible, but the MOST recent information I can find regarding the effect of the cumulative effects of these mitigations seems to clearly make the case that those early reviews are probably not representative of today's performance numbers and "feel". Having 5fps higher maximum or average FPS doesn't make much difference if you have an 8-10 FPS lower .1/1% that translates into visually noticeable problems.
Likely, none of these is really an issue, but since we know that THIS metric is the one that matters most, then it's what we have to point to when it's clear that there is a difference between two products and one doesn't perform as well as the other on this baseline.
Then again, I've been wrong before so I'm sure somebody will come along and tell me how full of crap I am.
Facebook
Twitter
Instagram