Gadgetory

welcome back to harbor unboxed today we're finally comparing the gaming performance of aim DS Verizon 7 and Intel's Core i7 coffee-like CPUs although we've done this in the past we're still yet to carry out extensive gaming benchmarks using the GTX 1080 TI previously all testing was done with the RX Vega 64 graphics card not only that but all previous comparisons are also made before the meltdown and Spectre security flaws were discovered as it stands we're still the thick of it with these security issues for those of you unaware a team of security analysts employed by Google tasked with finding a zero-day vulnerabilities found three CPU related security risks last year the team is known as project zero and the vulnerabilities that they discovered are now commonly referred to as spectrum meltdown variant 3 which is technically known as rogue data cache load is now commonly referred to as meltdown and this variant only impacts Intel processors and I wasn't mitigated via Windows Update AMD believes that its processors are not susceptible to this variant due to the way their CPUs are designed and therefore no mitigation is required that said the windows update is still being applied to AMD systems and the same performance hits suffered when testing the Intel systems can be seen thankfully though the meltdown patch had little - basically no impact on gaming performance though it does degrade storage performance just a little bit the real problem for both AMD and Intel is the spectra vulnerabilities variant one being bounced check bypass and variant to branch target injection starting with AMD they believe anyone can be contained with an operating system update and see Microsoft's working on rolling those updates right now variant 2 is also an issue for AMD processors AMD has stated they believe their CPU architectures make it difficult to exploit variant 2 but they'll continue to work closely with the industry on this threat they go on to say that optional microcode updates will be available to their customers and partners for Eisen and epic processors starting this week so that was actually meant to commence last week as it stands AMD nor their board partners can give me any information these updates from what I can tell nothing's actually in the works for be 350 and X 370 boards at the moment but no one's really telling me anything so I have no additional information on the situation so right now for eyes and users you can update using the latest Windows patch but there will likely be more updates to come as for Intel it seems to be a much messier situation meltdown has been mitigated by the Windows Update but spectra still remains a big problem all the major partners have now issued BIOS updates intended to mitigate variant to branch target injection we tested this a few weeks ago now and found that for the most part games were only around three to five percent slower while storage performance took a massive nosedive since then countless users and media outlets have confirmed those findings so we need to do more testing to see what this ultimately means for desktop users as for variant 1 which aim they think they are mostly safe from Intel is yet to address at all as I understand it they are more vulnerable than AMD is worse still apparently even after erasing variant 2 with the BIOS update that destroyed storage performance it's reported that doesn't actually solve the issue Google has since come out and said they have a solution for variant 2 that doesn't impact performance their method uses software patches rather than disabling the affected CPU features according to Google their patch codenamed ret Pauline has a negligible impact on performance particularly when compared to Microsoft and Intel's fixes this is obviously something Intel arts clients will no doubt want to take advantage of so we can probably expect to see more updates in the not-too-distant future okay so now that we've gotten all that out of the way it's quite clear that neither AMD or intolerant of the woods yet and that performance we're seeing today could likely change in the near future what many of you or rather a whole heap of you have been handing me for over the last few weeks is a comparison between Rison and eighth generation core processors so today we're comparing the Rison 7 1,800 x with the core i7 8700 k and for this test the focus is on stock out-of-the-box performance the core i7 87 ROK processor has been tested on the gigabyte z 370 horas gaming 7 motherboard using version f 5l which includes the updated CPU microcode intended to mitigate variant 2 of Spectre meanwhile the risin 718 hundred x has been tested on the asrock X 370 Taichi and so far ambi hasn't issued any BIOS updates both the AMD and Intel CPUs were tested using ddr4 3200 CL 14 memory with the latest version of Windows fully patched the graphics card of choice this time is the GeForce GTX 1080 I'm testing takes place at both 1080p and 1440p using a range of quality presets we have a number of modern CPU demanding titles to look at so let's get to it first up we have Assassin's Creed origins and here we see at 1080p using the medium quality settings at the 87 RK is 18% faster or 23% faster if you look at the 1% low data the margin closes up as we increase the quality settings once we hit ultra the 87 RK is just 13% faster that's still a decent margin but with both CPUs maintaining over 50 FPS at all times the experience was much the same using either CPU of course 1080 T Ione is a more likely to game at 1440p and here the 87 or ok was just 6% faster on average though it was 11% faster for the 1% low result obviously the reason why the margin closed from 23% down to 11% for the 1% low result is because as we increase the quality settings the GPU continued to limit the performance further moving on to ashes of the singularity we find a similar story using the medium quality settings the 87 RK was as much as 22 percent faster but that margin was more than half at the same resolution using the maximum visual quality settings then at 1440p the 87 RK was just 6 percent faster testing with battlefield 1 provides some interesting results regardless of resolution and quality settings when paired with the gtx 1080i the Rison 7 processor allowed for a 1% result of between 100 and 110 fps in fact we only saw an 8% variance meanwhile the variance for the Intel CPU was as much as 38% and while it was at least 20% faster at an EP it was 2% slower at 1440p that said at 1440p the 87 ROK was still 8 percent faster when comparing the average frame rate right so next up we have Call of Duty World War 2 in this title unfortunately doesn't play that nicely with Verizon CPUs of course the resulting performance is still excellent as we saw well over a hundred FPS at all times even at 1440p however here the 87 arcade was still at least 17 percent faster it's obviously a pretty significant margin moving on to Dawn of War 3 we find the rise in 7 1800 X is good for around seven years 72 FPS for the minimum at 109 FPS for the average this meant the 87 RK was up to 20% faster but by the time we reached the 1440p resolution that margin was reduced to just 10% debut 6 mankind Nevada was testing using the DirectX 11 API because we're using a GTX 292 I hear the 87 ROK was up to 33% fast using the medium quality settings and then 19% faster with the ultra quality settings though it was just 9% faster for the 1% low result oddly moving to 1440p allowed the Rison 7 CPU to hit the lead beating the 800 K by a slim margin dirt 4 is one of the few recently released triple-a titles that was designed with rise and CPUs in mind although the 1800 X gets stomped using the medium quality settings at 1080p using the ultra quality settings brings rise and back into play then at 1440p we again see slightly stronger 1% lower results from the AMD processor here we have another racing game f1 2017 and this time we find things mostly going in favor of the Blue Man Group the 87 arcade was after 22% faster though with the ultra quality things that margin was reduced to 10% for the average frame rate but 16% for the 1% low then at 1440p we see very similar performance though the 87 era case still offered a better minimum result the project cars franchise has never been a friend to AMD and we see that here with the 87 ROK which was around 30% faster throughout the 1080p testing thanks to a GPU bottleneck things closed up at 40 and 40 P and here the Intel CPU was just 5% faster when testing with rainbow succeeds we see that the 87 era K was up to 23% faster though it was just 12% faster when using the ultra quality settings at 1080p then once we at 1440p the margin is reduced to 2% while the 1,800 X provided a better 1% low result based on a 3 run average Star Wars Battlefront 2 is the most recently-released game I've tested with them while there will be those that complain about it being included because of the loop box controversy and all that it's still a good quality game that's very well made and for now you can't actually pay to win so I figure we should probably just hold fire anyway as I said the game's very well made and we see that it does a great job of utilizing Rison when compared to many of the games looked at already here the 87k is at most just 14% faster and that's pretty impressive when you consider that was clocked at least 13% higher in this test so that is a really good result for AMD once we jump to 1440p the 87 or okay is still 10% faster but as I said given the difference in clock speed that's really not a bad margin finally the last game we're going to look at is total war Warhammer 2 and here we find some very competitive results indeed although the 1800 X was around 4% slower for the medium and high quality tests it was 6% faster for the ultra quality test and 14% faster once we hit 1440p so again another title where the rise in CPU manages to come out on top at 1440p okay so before I summarize all the data we just went through let's quickly look back at the numbers found previously when testing with Vegas 64 liquid here we have the low resolution 720p results using the ultra type quality settings and here the 87 okay was on average 19% faster than the 1800 X or 21% if we look at the 1% low results those margins were reduced to 7% at 1080p 10% for the 1% low result and again this is with the ultra type quality settings enabled then finally at 1440p the 87 RK was only 3% faster when comparing the average frame rate I happen to think the testing done previously with Vegas 64 is actually quite interesting and gave a different perspective to the usual you know testing done with an NVIDIA GPU for example the behavior is quite different when testing games using DirectX 12 with a radeon GPU and this does often play more into the hands of horizon still I got pretty slammed by quite a few people for testing with Vegas 64 liquid claiming that my results were misleading and that I should have tested exclusively with the GTX 10 atti I'd always plan to revisit these tests with the 1080i and I made note of that in the original video but given the Vega benchmarks included 720p results I wasn't quite sure what all the fuss was about anyway I've now done the 1080i testing we've looked at that so let's go over all the data or again and summarize that quickly and see what the picture looks like now here's a full breakdown providing the average results across the dozen games tested using the medium quality settings the 87 okay was seen to be 21% fast in the 18 hundreds when comparing the one percent low data even I was a bit surprised when I realized that this is the exact figure we saw we're comparing these CPUs with Vegas 64 at 720p what would you know I'll just wipe away that smug look and move on the electoral quality results at 1080p showed the 87 Academy 15% files for an average and 13% for the 1% lower result for the average framerate this is a bit different to what we saw with the slower Vegas 64 graphics card but at 1080p that's not entirely surprising and the 1% lower margins were actually very similar previously with Vega we saw a 10% margin then finally at 1440p the 8700 a was 5% faster on average and 2% for the 1% low result previously when testing with Vega 64 we found the 87 okay it was 3% faster on average and 2% for the 1% low result so again very similar findings this time around of course this time we do also have the meltdown spectral patches applied and this will be slowing the 87 arcade down by 3 to 5% so I'm aware that this is helping to reduce the margin right so there you have it not a lot has changed since we last did this kind of test for a no compromise type gaming solution just maximum frame rates best performance you can possibly get the 8700 K is really the way to go simply the best option available right now but if you plan to play at 1440p or at possibly an even higher resolution or use a lower end graphics card the difference between the 87 RK in AMD 6 and 8 core eyes and CP is going to be negligible for productivity type work it really depends on the application but we found mostly that when comparing the CPUs and they're out of the box formats to say before in your overclocking takes place the 1,800 X is generally faster with both CP you ever clocked that means the 1800 exit for you hurts and then the 8700 K at five gigahertz they are quite similar in most heavy workloads price-wise the 1800 X is about $20 cheaper but most smart shoppers will probably pick up the r7 1700 for just $290 and then just overclock it to somewhere between 3.8 and 4 gigahertz and that price the eight core eyes and CPU is really an awesome Buy so if you can wait till April I'd certainly do so because at that point we will have these second-generation rise in parts and we'll get to see what they bring to the table along with the 400 series chipsets getting back to the testing this video I didn't cover overclocking but it honestly doesn't really make that much difference the 1800 X is only a few percent faster on overclocked and then the 87 are okay well that does see a nice boost we're not GPU limited but of course for the most part you will find yourself GPU limited with ultra type settings and therefore you will see a little to no improvement even at 1080p in fact I've already shown hundreds of benchmark runs on the channel using these very games with the 8700 K clocked it between 4.8 and 5.2 gigahertz you can expect about a 7% boost at 1080p using the medium quality settings with just a one to two percent boost at an EP and 1440p using the ultra quality settings so for those of you who are going to beat your chest claiming at the 8700 K has to be overclocked and when it is it will just devastate the rise in CPU well that's certainly not true when using ultra quality settings with the GT X 1080 Ti anyway that's gonna do it for this one not a lot has changed but honestly we weren't really expecting it to Intel remains the king for performance while AMD still provides the most bang for your buck I hope you guys enjoyed the video if we hit the like button and be sure to let me know what you think below in the comment section Armijo Steve see you next time you