Gadgetory


All Cool Mind-blowing Gadgets You Love in One Place

Do you REALLY NEED Fast RAM? Core i5 8400, Z370 vs. H370

2018-06-02
welcome back to harbor unboxed today we have a hybrid benchmark / build video safely that turns out to be a lot of fun this idea came about when Coolermaster sent over their latest line of tough addition hardware as you probably guessed they have partnered with a soos and they've jumped on the tough branding to create the MB 500 tough edition we have the master watt 750 tough Edition then we have the master air coolers and that includes the master air ma 620 P tough Edition and the master ma 4 10 M tough Edition basically what this means is we now have a tough themed PC case we have a power supply there's actually two but we have one of them here and then there are only two coolers and we do have both of them here so they will all go nicely with the asou stuff themed motherboards and it's not just cooler master who's been hitting the gym team groups also been working hard to try and get buff with the new T force Vulcan tough gaming alliance ddr4 memory damnit guys enough with the ridiculously long prior names you get two words two words that's it that's the new rule apologies for that outburst I have taken a moment to calm myself down so let's get back to it the final piece of the tough puzzle is the asou tough h 370 pro gaming and I have to say this is a component that I own least component in that's largely due to the fact that the motherboard sells for $135 us and that means that roughly half a dozen zedd 370 boards on the market that actually cost less not only that but as Souza's own tough said 370 board comes in at a mere $15 price premium this is a problem because while the said 370 board allows for overclocking of the unlocked K model CPUs as well as ddr4 memory the H 370 board's support neither of these features if you're going with a core i5 8500 processor like we are see if you're overclocking really isn't a big deal as this is a lock CPU and therefore can't be overclocked and an e board but we are limited to ddr4 2666 memory and honors ed 370 board you could take advantage of much faster memory but how much of an advantage is the faster memory under typical gaming conditions we know from previous benchmarks using the gtx 1080i that you stand again about 10% more performance at best but is this still the case for lower end graphics cards depending on the quality settings it might be someone effort to determine just how much difference memory frequency makes we're going to test three graphics cards the GTX 1063 gigabyte gtx 1070 TI and GT x 1080 using medium high and ultra quality settings for the h3 70 board we're using g.skill sniper x26 66 CL 19 16 gigabyte memory and we'll be comparing that to a kit of the sniper X 3600 CL 1960 here byte memory so this isn't low latency memory but the latency is the same at both tested frequencies this means we will be installing both the asou stuff h 370 pro gaming and tough zed 370 plus gaming inside the cooler master masterbox mb 500 tough edition i'll also be testing out both of the air master tough edition cpu cores and i'll compare them to intel's epic box cooler ok so here we see when testing with battlefield 1 at 1080p with the ultra quality preset the gtx 1060 delivers the same performance on the HR 70 board using ddr4 2666 as it does on the zed 370 board was much faster ddr4 3200 memory upping the GPU firepower with the GTX 1072 ICS the faster memory improved the frame time performance by 8% though the average frame rate is increased by just a few percent then with the gtx Eternity we see the frame time performance improved by a 9% margin while the average goes almost unchanged reducing the quality preset to high doesn't change much as we see very similar performance trends to the ultra quality preset so let's see how things look using the medium quality settings interestingly here the frame time margins close up as we are likely seeing less load on the CPU using these lower quality settings instead now we're seeing the average frame rate start to increase for the gtx 1070 TI and 1080 when using the faster memory meanwhile the gtx 1060 is still the primary bottleneck as you probably expect when jumping up to 1440p this see the results become much more GPU bound and although the gtx 1070 TI and 1080 are pushing over 60 FPS at all times the faster memory doesn't really help eke out any extra performance reducing the GPU load with the higher quality settings doesn't change much and again the faster ddr4 memory is if no real benefit here then with the medium quality settings we again see no performance gains with the faster ddr4 memory we need to increase the GPU horsepower to see any real benefit here next up we have fortnight and 1080p use in the epic quality preset the GTX 10 6 delivered these same moments on both motherboards using ddr4 2666 and 3200 memory we see a very small 2% increase in performance for the gtx 1070 TI and a 4 percent increase for the gtx 1080 so nothing to write home about now if the higher quality preset we see a small benefit to using the higher clocked member at the gtx 1060 is the average framerate is improved by a 3% margin we do though see a much more impressive 11% improvement for the gtx 1070 TI and much the same for the gtx 1080 though we do appear to have reached the limits of the core i5 8400 in this test suggesting that at most these dead 317 chipset and the fasted ER for memory well it boosts performance by a little over 10% but let's move on to see what the medium quality results have for us again small gains can be seen when testing with the gtx 1060 while the 1070 was 9% faster than the 1080 11% faster so reasonable formats gains though we do only see a 4% increase for the frame time performance once again increase in the resolution to 1440p basically eliminates any advantage of using the higher frequency memory at least with the ultra quality settings we see some improvement 1440p with the high quality preset but for the most part the gains are very small then finally the medium quality settings and we see up to an 8% improvement for the higher-end GPUs but again the frame time results are much the same the last game I test it is far cry 5 and again we'll start with the ultra quality settings at 1080p and here we see a 7 percent improvement for the 10 84 percent for the 10 72i and just a single percent for the gtx 1060 the margins are much the same using the higher quality preset so again the to memory offers a very little in the way of out of performance with these GPUs even with the medium quality ceilings the margins remain much the same and this really was our best chance to see any kind of noteworthy gains in far cry 5 of course as we've already seen twice now increasing the resolution results in a heavily GPU bound scenario and now the improved memory speeds offer nothing in the way of extra performance this is also true when you than the higher quality preset even the gtx tay delivers the same performance then finally with the medium quality settings we see once again the exact same performance using either the h3 70 or z3 70 motherboards just quickly before moving on from the benchmarks i tried out the cooler master master at ma 620 p and ma 410 m tough edition coolers but i only had time to compare them with the intel box cooler i'm a core i5 8400 although you can't overclock the 8400 the advantage of these big aftermarket coolers is that they reduce load temperatures as well as operating noise under 100 percent load in our blender stress test the temperature drops from 77 degrees to just 49 degrees with the ma 410 m and then a further three degrees with the ma 620 p impressive stuff and if you care at all about temperatures an operating volume these coolers are worth the upgrade even for a locked core i5 or i7 processor although we saw a little to no performance gain when using the Z 370 board along with the faster ddr4 memory I still find H 370 boards bit of a tough sell pointless even H 370 boards do start at $90 us while these ed 370 models start at $100 so for just $10 more you can have as in 370 board and in this case the tough model as I mentioned earlier is just 15 dollars more so I feel like for that small investment you are better off with this board opposed to the tough H 370 board and yeah $15 really isn't much to pay for the fully fledged model the biggest cost difference can be seen when looking at the memory 16 gigabytes of ddr4 2666 starts at a hundred and sixty dollars us while the 3600 kits start at a hundred and ninety dollars us if you combine the motherboard and memory prices your stand to save at least $40 u.s. and that's starting it up there however with the memory being by far the biggest expense I'd still opt for a Zed 370 board and pair that with the more affordable memory thankfully their sous does offer an extensive range of tough motherboards so those wanting to buy a cooler masters master box MB 500 tough Edition have a number of options for an AMD themed build there is also the new tough x4 70 plus gaming and with a rise in process so you really could take full advantage of those air master tough Edition CPU coolers unfortunately right now I only have local pricing information for the new cooler master tough Edition products over at PC case key the master box is selling for one hundred and twenty-five dollars Australian which is $20 more than the standard model the master what 750 also comes in at one hundred and twenty five dollars while the coolers cost $80 each as expected you are paying a premium for the tough Edition gear but if you want that unique almost custom modded look then this new lineup will appeal and that is going to do it for this one if you enjoy the video between it the like button subscribe for more content you can appreciate the work we do here at harbor unbox then can see that supporting us on patreon thanks for watching I'm your host Steve and I'll see you next time
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.