Gadgetory

welcome back to harbor unbox i've now tested all ten retail core i7 87 hurricane processors and it's now time to go over the results if you missed yesterday's video then check that first as it provides a bit more of an explanation of what we're doing here basically though when reviewers like myself checked out the core i7 87 okay for the first time around six months ago now we were all or rather most of us were impressed with the overclocking capabilities most reviewers hit 5.1 gigahertz using around 1.3 5 to 1 point 4 volts but reported scorching hot temperatures at this frequency temperatures that were hell into the 90s half-a-dozen reviewers including myself push things to their limits and managed 5.2 gigahertz these are obviously very good chips but again they did run super hot at that frequency and would require deleting for long term usage quite a few reviewers though were limited to 5 gigahertz and well you could speculate as to why that was stuff like they didn't push hard enough they were happy with 5 gigahertz and just stopped there or whatever other story you can come up with I'm no pro overclocking though I've certainly never claimed to be on the channel and for the most part overclocking doesn't really excite me much these days now that said hitting 5.2 gigahertz on my review sample was extremely easy and I didn't say with very little effort I'd say most of the reviewers that were limited to 5 gigahertz know at least as much as I do when it comes to overclocking in fact I'd say most of them probably enjoyed tinkering with this stuff more than I do and possibly have more experience I guess what I'm saying is if I can do it I'm confident they can as well and I'm certain they would have as a side note I was able to boot into Windows a five point three gigahertz with my sample from Intel but I wasn't able to complete any tests I've since spent quite a few hours with a few different high-end motherboards and despite temperatures remaining quite reasonable even with dangerous voltages I haven't been how to stabilize the chip at that frequency it seems to be able to do 5.2 you can Hertz with ease 5.3 gigahertz though seems to be just out of reach anyway despite a mixed range of frequencies from reviewers there's always claims of samples being cherry-pick so I decided to look into this further supporting this endeavor is pck skier who handed over 10 brand-new unused car I 787 here ok retail boxes for testing so I've spent the last three days swapping 8700 KS in and out of the test bed to find the maximum stable frequency each chip is capable of none of the CPUs in this test have been deleted including my samples from Intel and asrock for testing I'm using the gigabyte Z 370 or s gaming 7 motherboard and for cooling I have the thermal tape Pacific our 360 d5 liquid cooling kit for this one I'm just sticking with basic multiplier overclocking so no adjustments have been made to the base clock and in order to validate each overclock I'm running a blender workload which takes daily 700k a little over an hour to complete so no it's not a 24 hour extreme stress test I just don't have over a week to invest in this video right now certainly not with second gen rise and CPUs knocking anyway this is still a very heavy workload and chances are if it can survive this it's really good for any gaming session you can throw at it ok so starting with my Intel review sample which we know is a beast and I wouldn't for a second argue this isn't top grade silicon using just one point three seven volts in the bias it manages a rock-solid at five point two gigahertz overclock under load hardware info reports drawing up to one point three nine two volts and peaking at 91 degrees which is obviously very hot and for prolonged use you will need to deal with this thing it also pushed total system consumption at 275 watts based on an average of six passes we recorded a multi-threaded Cinebench score of sixteen hundred and sixty three points and a corona benchmark completion time of one hundred and twenty-two seconds in comparison the sample provided by asrock which has a slightly different batch number isn't quite as good although it managed the same 5.2 gigahertz overclock it did require more voltage and therefore did consume more power and ran a tad hotter now time for the first retail chip and this one managed 5.1 gigahertz at a very reasonable 1.3 eight volts which saw the system consumed 266 Watts under load and maxed out at 88 degrees in terms of performance it was very close to the eyes rock sample despite being 100 megahertz down though of course we are only talking a 2% reduction in frequency I was able to load into windows at 5.2 gigahertz but even with 1.4 to five volts it wasn't completely stable therefore while not as good as the samples that I got from asrock and Intel this retail ship is still very good moving on we have retail chip number two and we find a very similar result 5.1 gigahertz but this chip did require a bit more voltage and this increased the power consumption and familes a little further and this chip wasn't able to even load windows at 5.2 gigahertz no matter how much voltage we gave it then retail chips three four and five are all limited to just five gigahertz with around 1.3 five volts I could have done a bit more voltage tuning here to be honest but I was mostly focused on the frequency all three could load windows at 5.1 gigahertz using 1.4 to 5 volts but stability wasn't 100% they did appear stable at 1.4 5 volts but temperatures were getting well into the 90s like very high nineties almost knocking on a hundred degrees and that is without open a loop system so without a dealer these results are unacceptable in my opinion retail chips 6 7 and 8 were all much the same hitting 5.1 gigahertz a stable 5.1 hurts though they did require over 1.4 volts to do that and therefore power consumption a thermal output was quite high so again good overclock is but there clearly not as good as my Q s sample finally though I did have my first real win when I got to chip number 9 this one was very similar to the Q s chip provided by asrock I managed 5.2 gigahertz with the v core set to one point 4 volts and the bias and performance was identical to that of the chip provided by Intel that said the voltage was much higher and this increased power draw and thermal output quite substantially then last but not least we have retail chip number 10 and this was another 5.1 gigahertz chip that required just over one point four volts so not as good as the Q s samples but still quite a good retail chip then wrapping up the testing just out of interest sake I decided to install each CPU again load the BIOS defaults and then enable multi-core enhancement as well as XMP then I ran the tests again monitoring things like the frequency voltages power consumption and temperatures all 12 CPUs hit 4.7 gigahertz on all calls for both the tests and performance varied by no more than a single percent which I'd say is margin of error even for a 6 run average again we see that my sample from Intel was a little you're running quite a few degrees cooler than the retail CPUs and despite hitting a similar or even higher maximum vehicle and maybe the heat spreader on this chips pressed down a little harder or something overall we had just one of the 10 retail chips that could do 5.2 G Hertz six that did 5.1 gigahertz and obviously there were three that did five gigahertz but of course there was nothing worse than that so overall I was a bit disappointed not to see at least one other chip capable of 5.2 gigahertz but I suppose at least we've got no lemons so overall probably a pretty good result there honestly though the results I found with these 10 retail chips really aren't that different to what the tech media found in fact overall I'd say they're actually a little bit better and while I wouldn't dare say that you're guaranteed 5 gigahertz if you were to buy one of these ten chips then you would have achieved at least five gigahertz and while some reviewers that were given chips by Intel didn't even get that and we are talking about seasoned highly experienced tech guys you know guys like Steve from gamers Nexus I know if I were going to spend loads of time and energy sifting through hundreds if not thousands of CPUs to find the golden eggs I'd be sending them to guys it'll get the most out of them whatever the case what these results mean and what they are by no means are conclusive but what they should indicate is that you have an extremely good chance of getting an 87 here okay that will do at least 5 gigahertz of course even for a 5 gigahertz overclock you will require a high-end air cooler or a liquid cooler with at least a 240 millimeter radiator and for best results you will need to deal it and well I've given my strong and negative opinion on this subject several times before on the channel in a nutshell I find it unacceptable the Intel forces are those who purchase their expensive unlocked CPUs to spend even more money and time deal letting them just so they can achieve acceptable operating temperatures but that's an entirely different issue so let's not get into that here things to look out for in reviews that include overclocking is stuff like power consumption and thermals when overclocked as well as take note of the motherboard and cooler used for my review I used the extreme and very good MSI Zed 370 GOG like with a 2 and 40-millimeter all-in-one liquid cooler and I had this to say with a room temperature of 21 degrees the 8700 K titled 25 degrees stressing both the CPU and FPU course temperatures 84 degrees while only stressing the CPU saw temperatures max out at 60 degrees once I have a clock to 5.2 gigahertz we reach within six degrees of the t.j.maxx while running the CPU stress test peaking at 97 degrees briefly obviously a deal it would help tremendously for those seeking extreme overclock and or a more extreme cooling solution so basically what this means unless you're willing to deal with the 87 ROK and couple it with an expensive cooler you aren't running over 5 gigahertz even if you do win the silicon lottery whether or not Intel is cherry-picking review samples honestly I still don't know I got lucky for sure but there are those that didn't for me though the evidence so far probably suggests that they're not but if you believe they are based on the evidence at hand then you'd also have to strongly believe that AMD did exactly the same thing with Rison whatever the case it's hardly the biggest con job in the world if they did in fact cherry-pick them and reviewers mostly show 5 to 5 point 1 gigahertz with a few showing 5.2 gigahertz then based on the evidence in this video if you were to buy them off the shelf you'd likely see at least 5 gigahertz maybe five point one and possibly five point two so yeah big deal about nothing probably if the average retail CPU did let's say four point seven to four point eight you can Hertz maybe even four point nine gigahertz then sure that's a bit of a problem but with five gigahertz looking almost guaranteed assuming have a good motherboard and cooling then it's hard to see what all the fuss is about finally as I said earlier the results from our 10 retail CPUs are by no means conclusive I really looked at batch l7 33 and even then there's a chance you could get a cpu from this batch to do four point nine gigahertz or you could get one that'll do five point two gigahertz or maybe even better again silicon lottery and that is going to do it for this video really big thank you to PC case KeyUp for supplying all of ten retail 8700 K CPUs if you enjoyed this video and you appreciate the effort involved in credit then please head over to the PC case key website or if you're not in Australia and maybe stop by their Facebook pages give them some love there you can also support us directly by a patreon page and if you do so you'll gain access to our discord chat and monthly livestream and that's with Tim and myself and we often give our patrons a bit of a sneak peek at what's going on behind the scenes as well stuff like this very video anyway thank you for watching I'm your host Steve and I'll catch you again very soon for some second gen arise an action see you then