Gadgetory

welcome back to harbor I'm boxed now yesterday we finally got to show you what the new Intel Core I nine ninety nine hundred K and core i7 ninety seven hundred K processors have to offer in summary the ninety nine hundred K was fast but not really fussed enough to justify the price and that now appears to be the general consensus among reviewers along with the high price the other major issue was the high temperatures again most reviewers were reporting very high stock temperatures using high-end coolers and that basically killed limited hurt the overclocking performance of these new chips in my review I spent quite a bit of time talking about these shockingly high thermals I expected that the night a hurricane was going to be a seriously power-hungry CPU and therefore it would be a rather hot item I just didn't expect Intel's first soldered chip in a very long time to run quite as hot as it did using relatively low voltages at five gigahertz or the 9900 k peak at 100 degrees with the coarser hydro series H 100 I Pro RGB or Noctua NHD 15 and we're talking about some pretty premium coolers here this was actually worse than what we saw previously with the 8700 K at 5.2 gigahertz granted the 9900 k-pax 33% more cause but it's soldered whereas daily similar k uses Intel's notoriously rubbish thermal paste in the past I've deleted the 7700 k and 8700 chips with pretty amazing results using liquid metal drop temperatures at least 20 degrees there a big part of that improvement was achieved by removing the IHS glue which reduces the gap between the CPU die and the heat spreader still we know that Seoul during CPUs works a lot better than the pace method that Intel's been using to save on production costs fears now we know this because Intel's desktop CPUs ran much cooler back when they were sold in 2011 and that was the Sandy Bridge days for the mainstream desktop processors and then the high-end parts were soldered right up tool Broadwell II in 2016 as a quick example the chorus 3770k running at 4.7 gigahertz using 1.35 volts would peak at just over 90 degrees running an Ida 64 stress test and that was with a large sort of tower style air cooler under almost the exact same conditions but with one point 4 volts so a bit more voltage the 2600 K around at least 20 degrees cooler so that being the case I was expecting the 1900 be something special and not the kind of special that it turned out to be anyway what I wanted to know was how much better is the soldered method used by the 99er okay than the paste of the 8700 K or 8086 k as they're essentially the same CPU and I will be using the 8086 K for my testing as I just had that CPU laying around the 87 okay is busy in my GPU test rig anyway to find out I disabled to 1900 K cores which effectively turns it into an 87 or okay or ATO you 6k middle II it's far from an exact science as the 99er okay is a physically larger CPU packing more l3 cache but it's about as close as we can get with the harbor I have available the core i5 9600 K might make for a better comparison but I am still waiting for my chip to arrive so we can revisit that down the track if need be that said the 9600 K is basically just a 99 or okay with two of the dies disabled but you do get a smaller l3 cache anyway this comparison should give us a pretty good idea of just how much of an improvement Intel stim makes for the first test configuration I locked both the 19900 K and 8086 K at 4.5 gigahertz on the MSI Zed 390 godlike blender was used to place full load on all cores and again both CPUs were tested with just six cause active and for this test the voltage options were left on auto this saw the 9900 K running at one point one six volts for the most part and the 8086 K at one point two six volts the 9,900 case of a peat core temperature of 61 degrees and the XTU software also reported a 61-degree package temperature the 8086 k on the other hand went as high as 74 degrees in the package with a peak core temperature of 70 two degrees so for the 4.5 gigahertz comparison the solid method reduced temperatures by 11 degrees a 15% reduction though as I noted voltages are also 8% lower the package TDP was also 16% higher on the 8086 cake-eating 125 watts opposed to just 107 watts for the 9900 K that being the case for the next test I locked the voltage at 1.3 5 volts on both CPUs leaving the operating frequency at 4.5 gigahertz this evened things up and now the 8th gen processor was hitting a package TDP of 145 watts while the 9th gen ship hit 144 Watts so basically the same figures there that's about a 16% increase for the 8086 K from the previous test and 35% increase for the 19900 K this saw the 8086 hit a peak package temperature of 88 degrees and a core temperature of 87 degrees with a typical low voltage of 1 point 3 7 3 volts the 9900 K on the other hand peaked at a package temperature of 79 degrees and a 78 degrees for the cause while running at one point 3 6 6 volts so that's a 9 degree improvement for the 9900 K allowing it to run 10% cooler that's a decent result for the soldered CPU that said we are almost already at 80 degrees so how much hotter do things get a 5 point one gigahertz with the same 1 point 3 5 volts and what are the improvements over the 6 core 8th gen part if you recall I opted to run my 87 or okay test rig at 5 gigahertz as the temperatures were much more acceptable than what I was seeing at five point one gigahertz and the same is true for the sole at 900k running an average of one point three eight volts the 8086 K had a peak temperature of 94 degrees for both the core and package this resulted in a package TDP of 188 watts so pretty toasty stuff and again this is why we decided to run these processors at 5 gigahertz for our GPU test system in comparison the 19900 K peaked 91 degrees for the court and 90 degrees for the package the average voltage during the test was one point three six six volts and this resulted in a package TDP of 181 watts so this means when overclocking the soldered chip was just three degrees cooler than what we saw with the 8th gen model using paste this also explains why we saw a 98 degree operating temperature with all 8 cores active that's actually a smaller increase than what you'd probably guess based on the six core results so why does the 9900 KT go from offering a nine degree improvement over the 8086 at 4.5 gigahertz to just three degrees cooler at 5.1 gigahertz well my guess would be that we're reaching a thermal bottleneck with the 99er okay devourer recently discovered Wendy letting one of the ninth gen processors that the dye is significantly thicker than that of the 8700 K I won't give his exact measurements but if you want those please go check out his video it's certainly worth having a look at as Devourer notes the thermal conductivity of the silicon isn't great so the more of it you have are the more thermal resistance you'll face and this appears to be an issue for the 99 hundred K it's my assumption that this isn't as much of an issue at 4.5 gigahertz but as we increase the thermal output the silicon bottleneck becomes more of an issue as it starts to degrade thermal performance to the point where Sol during the IHS is of little benefit proving this tube our sanded down the core i5 9600 K so they can buy just 0.2 of a millimeter and this reduced temperatures by five and a half degrees and that's a rather significant improvement again if you want a full rundown of all the testing please visit to Bauer's channel on a slightly different subject a few viewers were a bit confused or concerned by my findings and yesterday's video claiming that the 9900 entrie aliy be that much hotter than the Rison 720 700 X as they both are eight core 16 thread CPUs and they are both soldered well the main reason why the night of decay was so much hotter was because of what we just saw in this video clock speed the note on her okay with just six cores active went from 78 degrees at 4.5 gigahertz just 91 degrees at 5.1 gigahertz and that's a 17% increase in operating temperature the 2700 X can be overclocked to around 4.2 gigahertz but by default runs at an all core clock frequency of 3.8 gigahertz with the Wraiths prism box cooler about 3.9 gigahertz with an aftermarket closed-loop cooler for example also helping manage thermals is the CCX design which doesn't pack the course as densely basically you get grouping of four cores rather than one big block of eight so that I imagine that helps quite a bit basically intel has just been too aggressive with their clock speeds though and this has resulted in eight core CPU that's almost too hot to handle as for the silicon well there's likely a number of reasons why Intel has had to increase the height and I doubt any of them have anything to do with them being incompetent as some of the viewers have suggested in the comments of yesterday's video pretty sure the Intel engineers know exactly what they're doing and they may be quite limited in what they have to work with at the moment but they know what they're doing all the same as I understand it sold during the 80s 700k would be much more risky as the silicon could crack during the heating of the soldering process therefore Intel's been forced to increase the thickness of the silicon to better handle the stresses of the heating process required to melt the solder a thinner silicon would complicate the process likely making it more time-consuming and therefore more costly then there's the issue of the thickness of the solder which is quite thick and some of you suggested in yesterday's video again that it is too thick and Intel messed up here and well that could be possible maybe it is too thick but I suspect Intel faced a few issues here as well one being damage a thinnest solder I would be more prone to cracking that seems quite obvious particularly after prolonged use these chips heat up quite fast and then depending on the cooler use they can cool down quite rapidly and this places a lot of stress on the solder layer now you can deal heed the 1900 K is Intel I didn't want to go with a super hard solder they're gonna try and avoid cracking there's certainly more work involved to clean it up opposed to what we saw with previous generation CPUs such as the 87 are ok it's much harder to remove the solder than it is just wipe away the paste but when doing so and reapplying with liquid metal that will reduce temperatures by a further 5 to 10 degrees so there is that however we're not really fans of D leading here at harbor unboxed it voids your warranty it can be a risky process and really we just rather you didn't have to particularly when spending at least $500 on a brand new CPU not to mention $500 on a CPU that is unlocked and is intended to be overclocked at mid elite is a fun and challenging experiment Maddon thues with deep pockets but those of you who just want to overclock without getting the lab code out the crazy high 99 okay temperatures will be quite frustrating they certainly were for us at the end of the day it seems like Intel is really just pushing everything to the limits and therefore they were forced to solder these chips in order to maximize the thermal conductivity overall though it's still a better method as we are seeing improved thermal performance especially out-of-the-box thermals unfortunately though we are only talking about three to four maybe five degrees when overclocking and that is going to do it for this one if you did enjoy the video be sure to the like button subscribe for more content and if your preciate that work we do Harrow box then consider supporting us on patreon thanks for watching I'm your host Steve see you next time