Gadgetory

welcome back to harbor unboxed now it's been roughly three weeks since we reviewed the core oh nine ninety nine okay and in that time basically no one's been how to buy it and locally here in australia it doesn't look like we'll be getting any more stock for at least another week and how much more will be getting in a week is anyone's guess but i'm guessing it will be bugger-all anyway despite the fact that you can't really buy a core i nine nine are okay we're seriously considering re-reviewing it because we think it's probably worth doing and we'll discuss why in this video and then I'll actually turn it over to you guys to vote on what you think the best course of action is today's video has been sponsored by as rock and then you phantom gaming range of Zed 390 motherboards the Zed 390 phantom gaming six and nine include a blazing fast 2.5 gigabits per second network interface offering gamers and content creators two and a half times the bandwidth compared to standard Gigabit Ethernet for more information please check the link in the video description now since the 9-yard k reviews first went live we saw a few strange anomalies mostly relating to power and thermal performance and the results did stir up quite a bit of chatter regarding whether or not the night are okay in its stock trim really ran it over 80 degrees with a high-end cooler or is it more like 60 to 65 degrees to better answer that i guess you have to first define what stock actually means for the ninety nine hundred k officially intel specifies the core i nine ninety nine hurricane has a TDP rating of 95 watts they Staton all core base frequency of at least three point six gigahertz and this is what the TDP is measured from however they also state that if just a single core is active it will operate at up to five gigahertz so the two frequencies officially given are a three point six gigahertz base and a maximum boost of five gigahertz this means there's a 1.4 gigahertz disparity between the base and turbo boost frequencies now it's important to note that prior to the release of the 8th gen core series intel did actually specify the all core turbo frequency for example that made it quite clear that the core i7 7700 K would target and operate at four point four gigahertz when all four cores were under heavy load however with the release of the 8th gen series which brought about the first mainstream six core processor from Intel they made the decision to no longer disclose the all called turbo frequency of course a predefined maximum all call frequency still did exist and for the 8700 K it was 4.3 gigahertz Intel was just no longer specifying this frequency and therefore and no longer had to guarantee that it would be achieved by all hardware configurations for example low-end motherboards the 8700 K was right on the edge of the 95 watt TDP and depending on the motherboard quality this figure could be exceeded under optimal conditions with optimized voltages the 87 ROK pushes package TDP to roughly 95 watts at the 4.3 gigahertz all core frequency opposed to the 75 watts the 7700 K maxes out with its 4.4 gigahertz or core frequency so where is the 7th gen core series part came in well under the 95 watt TDP rating 3700 K was right on the edge and again we saw in lower quality boards that pumped more voltage into the CPU in an effort to maintain stability that the TDP was often pushed well over the 95 watt rating now for the new core I 999 hundred K the official default clock multiplier table states that with all eight cores active it will operate at 4.7 gigahertz and achieves this same frequency with 7 and 6 cause active then with five and four cores active it will go up to four point eight gigahertz four point nine you Hertz with three cores and as I said earlier five gigahertz with one and 2 cores active however as I noted earlier Intel only states and all core base frequency of three point six key Hertz in a single core maximum at Turbo Boost frequency of 5 gigahertz so whatever happens between those variables depends on the workload and crucially how long the workload runs for but this Intel specifies a time duration for certain power limits commonly referred to as pal 1 and PL 2 now we've covered this on the channel before but in short pail 2 is our boost consumption it defines how much power the CPU can use in bursts and okay the limit is 119 watts Intel says that the pale to state could be sustained for up to 100 seconds but in reality with a 1911 that it's just shy of 30 seconds in a typical workload called workload at least with the said 390 motherboards that we've used with the 95 watt TDP enforced at which point the power limit kicks in and that limits the frequency down to whatever fits within that 95 watt envelope and we've typically seen during stuff like our blender workloads that all eight cores will clock down to four point three gigahertz rather than the maximum four point seven gigahertz that can be achieved without any power limits in place so the problem we have here is that motherboard manufacturers for the most part aren't abiding by either of these power limits instead they're just targeting the clock multiplier table as they would have done with the seventh sixth and so on and generations of core processors yeah you could say that the motherboard manufacturers are cheating but we don't think that's it we still feel it's Intel who are cheating their own spec there's just simply no way the board manufacturers did this on their own and it'd be a mighty big coincidence that they all decided to break the spec in the same way and surely a seuss MSI asrock an gigabyte have all worked alongside Intel engineers to create their Zed 390 motherboards anyway let's not get into that just yet for now let's look at how badly the TDP rating is getting abused for sustained periods here I've tested the 9900 K in our blender workload without any TDP limitations in place using a number of core count configurations leaving hyper-threading enabled I've tested the 99er okay in its stock trim with all eight cores enabled and then seven more times with a specified number of cores enabled at the BIOS level I've also set the frequency to meet that of Intel's multiplier table with just a single core enabled the ninety-nine hundred K targets five gigahertz and here we see a packaged TDP of just 35 watts well under the 95 watt rating now please note all this testing was done with the voltage held at a steady one point one eight eight volts moving on with two cores enabled the 9900 targets five gigahertz and here the package TDP is raised to 49 watts 49 watts with just two cores enabled with three course the frequency drops to 4.9 gigahertz and this resulted in a package TDP of 66 watts for course targets 4.8 gigahertz and here we're hitting 83 Watson already with just half the COS enabled we're not that far off exceeding the specified TDP rating interestingly whereas the core i7 77 Eric a maxed out at just 75 watts the 900k with the same amount of cores and threads enabled hits 83 watts that's an 11 % TDP increase though we also do have a 9% increase in clock speed and the rest of that margin can likely be attributed to any changes in voltage along with a much larger l3 cache of the 99er okay of course we're not stopping here the 1900 km a package TDP of 96 watts with five cores enabled at 4.8 gigahertz then with six cores we dropped down to 4.7 gigahertz and here at the TDP hits 114 watts then 130 watts with seven cores and technically this shouldn't be possible with Intel's PL 2 spec as the limit here is a hundred and nineteen watts and then finally 153 watts with eight cores enabled at this point we've exceeded the base TDP rating by a little over 60% and the pl2 short burst spec by almost 30% so what happens to the clock speeds on a long run test if we enforce the 95 watt TDP as expected these single and dual-core configurations still hit 5 gigahertz because as we saw previously this is the 99er okay come in under the 95 watt rating then with three cores active we still hit the target of frequency as this only saw a packaged TDP of 66 watts however with 4 cores active we see a 50 megahertz return in frequency with the 95 watt limit in place this is then extended to 250 megahertz with 5 cores active 350 megahertz with 6 cores 500 megahertz with seven cores and then finally 700 megahertz with all 8 cores active this means we're seeing up to a 15 percent decrease in clock speed for our blender workload and for less optimally can get motherboards running more voltage this margin will likely increase I should also note that I'm not using the integrated GPU for any of this testing now since first reviewing the kora 999 okay I've tested over a dozen well over a dozen said 390 motherboards ranging from the cheapest 120 dollar to $150 US models right up to the much more expensive flagship versions out of the box all boards from MSI gigabyte and asrock run with a dirty DP limit in place even with the default bass configuration however this isn't the case for any of these suitable as I've used the default out of the box are cleared BIOS configuration employs a 95 watt TDP limit to remove it without having to dig into the power settings and do so manually you'll have to load an extreme memory profile and then agree to using the Isuzu optimized settings rather than Intel settings basically a seuss optimized means ignoring the TDP limit and running at the default clock multipliers like boards from msi gigabyte and asrock do essentially then Intel has two separate specifications for their high and CPUs a TDP limited specification that they loosely define or a clock multiplier table specification and enabling one means it's impossible to achieve the other the TDP limit means you won't reach the intended all core clock speed while the clock multiplier table spec means you're running well above the TDP this seems to have caused a divide between our viewers on how we should test Intel processors shall we make them abide by the specified TDP rating just as AMD processors do or should we test them running at the maximum allowed clock multipliers as they seem to do out of the box on pretty much all motherboards right now we're hearing a lot of back and forth between my suppose AMD fans AMD loyalists and then Intel fans Intel those extremely loyal to Intel for whatever reason it's not really good to be in either of those camps but we seem to have a lot of people that are anyway the point that I'm trying to make here is that opinions differ obviously between these two but we've also seen last year that there's even the people at each group seem to be unable to agree on how we should be testing these processes so for example with the 8th gen release AMD's pitchfork crew were shrieking at the top of their lungs how we should test with a 95 watt TDP in place the 95 watt limit well the well let's call them the Blue Man Group members they strongly disagree with that they wanted to see the Intel processors pretty much tested how they well how they perform out of the box on the motherboards that we were testing with which I suppose is fair enough even up was technically out of spec depending on how you define the spec hey now with the 9th gen series it's a bit all over the place some AMD extremists want us to actually test with the TDP remove now I guess they want to see the 1900 can then on the other side we have some Intel extremists that now say motherboard manufacturers aren't abiding by the Intel spec and are accusing them of making the 99er ok and run much hotter than it should and well that's partly true I suppose but I'm again not convinced that motherboard manufacturers are really the ones at fault here testing with the 95 watt TD spec also introduces a few testing issues for reviewers you can no longer just show a single or even worse the best Cinebench r15 run with an Intel CPU the same goes for benchmarks such as VRA or really any short rendering test like Cinebench or v-ray or really any short test that uses all 8 cores so ideally you'll need to show short and long run results for all applications as performance could vary by as much as 20% anyway to try and avoid hearing from the most passionate brand loyal fans we've created a poll and we want to know what you think is showing 95 watt limited results misleading even if it is included alongside unlimited testing personally Timur myself think the best course of action is probably to revisit the 99er ok review with both Unlimited and 95 watt limited results with both short and long run testing but as I said we're really keen to see what you think and what you believe is the most appropriate way to handle this sort of thing in the future and basically whatever you guys come up with in this poll will really influence how we test Intel CPU is moving forward I guess that's kind of useful as well since Intel themselves don't seem to be able to create a concrete concise specification for their own CPUs and even if they could it appears they're powerless to enforce it and I'm sure that's the case so please click the link below it'll be at the top of the video description that'll head you over to a public post and our patreon page where you can vote for your preferred option and yeah I'm very interested to see where we end up on this one anyway that is gonna do it for this one thank you for watching I'm your host Eve see you next time