Gadgetory

ái Oh temperatures are lying to you it's true but allow me to explain let's pretend a guy named Andy has been sporting a cooler master hyper 212 Evo for the past year he's satisfied with its 4.7 gigahertz i7 7700 K temperatures but wants to push the overclock a bit higher right now his core temperatures idle at around 30 °c and thing to jump up to around 80 degrees Celsius immediately after starting i-264 still not bad for a run-of-the-mill air cooler but things could definitely be better and he realizes this hops online and decides to purchase an a io like this one here the deep cool captain 240 X just insert your a IO liquid cooler of choice these coolers use liquid let's assume water with a few additives it's usually what it is as a medium of exchange between CPUs than the outside world the water isn't making direct contact with the CPU there's usually nickel plated copper or an aluminum base of sorts sandwiched in between but nonetheless it's the water that ultimately transfers heat away from the source and in comes the radiator these vary in size and fin density but essentially reverse the process that occurs at the block instead of heat being transferred from metal to water the radiator transfers heat from water to metal and relies on installed fans to dissipate or radiate heat into the atmosphere but here's where the lye begin Andy removes his hyper 212 Evo installs the AIO let's assume a constant max pump speed and a normal fan curve and immediately checks idle temperatures much lower right let's say 23 24 degrees Celsius wouldn't make much sense for core temps to be lower than room temperatures Andy then runs the i-264 and notices that his immediate temperature jumps are much more controlled whereas his core temperatures jump to 80 degrees Celsius before they're only jumping to around 55 degrees Celsius now job well done right the AIO was definitely worth it let's see if he's still saying that in about an hour or so in comes the misleading nature of a iOS and comes specific heat it's defined as the amount of energy required to raise the temperature of a substance by a single degree Celsius or Kelvin it is often listed as an intensive property with respect to mass hence joules per gram Kelvin or joules per Graham Kelvin if we assume a copper cooling system and a perfect translation of energy from the CPU heat spreader to the cooler space that's what thermal glue seeks to do and the cooler in question would be able to absorb only point four joules of energy per gram of copper before increasing in temperature by a single degree Celsius so in a nutshell copper heats up very fast it's a thermal conductor and most metals behave in this way this is why Andy sees an immediate temperature jump with his air cooler when he first starts out of 64 copper is absorbing energy at the same rate that water would be in its place but copper has a much lower heat capacity which means it heats up quicker it's absorbing a certain amount of joules per second denoted as watts which is where TDP comes into play and is either passively or actively radiating this heat water on the contrary has a very high specific heat capacity which means that it can absorb a substantial amount of energy per unit mass before increasing in temperature by a single degree it's specific heat capacity it's 4.2 joules per gram Kelvin nearly 10 times that of coppers as a relational example picture is stove with a pot of water if the pot is made of metal usually a sort of stainless steel Teflon combo the metal will heat up at a much quicker rate than the water will not only because the metal is in direct contact with the stove but because it's specific heat capacity is much lower your cookware reaches water's boiling point long before the water itself does even though it's in direct contact with the metal if both the water and alloy had the same heat capacity they'd heat up at the same rate and reach water's boiling point at the same time water's high heat capacity and low viscosity make it an excellent coolant in several applications including PCs however ironic that may be but do not be fooled there are several misleading claims on forums and even other YouTube channels in regard to a iOS and they're ultra cool yields the first has to do with load testing when Andy first started i-264 the a iOS temperature spike was much more controlled if you recall 55 degrees Celsius verse 80 degrees Celsius but this isn't the entire story the entire story would reveal the thermal property differences between water and the metal in question usually copper or aluminum first off Andy's initial temperature spikes were lowered because water requires nearly ten times the energy to reach the same temperature as the copper in the air cooler not only that but the Loup requires a length of time to equalize I demonstrated this with Heisenberg's i7 6700 K and the NZXT kraken X 62 it took nearly 42 minutes for the core temperatures to stabilize much longer than a typical air cooler whose respective core temps would plateau much sooner you can verify this yourself if you have your own air cooler in your PC doesn't matter what size it is your temperatures will peak in a few minutes or so now this isn't to say that a iOS are unnecessary that's not the point of this video of course this a IO from deep cool will better cool your CPU than a hyper 212 Evo hands down it does a better job at dissipating Heat plain and simple but not only thanks to the water involved also the 240 millimeter radiator further down the loop what I am trying to say is that a iOS are often misunderstood because their testers are often too impatient to wait for temperatures to stabilize at least those who are posting on forms all the time that their temperatures drop by 40 degrees Celsius which is usually not the case if you let those things run for a couple hours with i-264 you'll find that that Delta is much smaller I was watching a video from Joker Productions a few days ago he built a sweet new PC and was discussing CPU temperatures after running i-264 for just over an hour he said on his 7700 K captain 240 x combo his core temperatures leveled off at around 75 degrees Celsius that's what I want to hear he let the thing run for more than just a few minutes 100% necessary for liquid cooled systems the t delta can be much lower for air cooled systems because metals heat up rapidly by contrast but in comes the other side effect of a iOS the fact that they stay hotter for longer after prolonged burnin even if all Andy does is play video games for a few hours at a time they don't even have to be CPU intensive his core temperatures will be higher for a much longer time after returning to an idle state because water tends to retain heat much longer than most metals do thermal diffusivity is a variable at play here of which specific heat is a function but without using numbers you can test this yourself as well if you heat up an equivalent mass of both iron and water to the same temperature let's say 80 degrees Celsius and then set each aside to passively cool the iron will reach room temperature again much faster than the water so while water requires more energy to reach a certain temperature it all wants to hold on to that temperature for a longer amount of time which means that your idle CPU temperatures could be in the 40s or 50s long after you finish to gaming running benchmarks or rendering this isn't necessarily a bad thing it's just a side effect of an AI oh you can tweak fan curves to speed up this process by the way it is something to keep in mind though as it's often overlooked by first-time users of closed and even custom loop assemblies blue sky will also have this problem but not to any significant degree thanks to its two large radiators you can't change the properties of the compound but you can change the environments in which the compounds react if you liked this video be sure to give it a thumbs up thumbs down for the opposite be sure to click the subscribe button if you haven't already I will catch you in the next video waiting to finish good old blue sky this is Salazar studio thanks for learning with us