Gadgetory

our previous 777 100k coverage focus on game streaming ultimately ruling that least 1,700 be best suited for a task like playing games while simultaneously live-streaming them today we're looking at the other end of the spectrum which one does better with 1440p 144 Hertz gaming or 144 Hertz in general and because we know that 7700 K is a leader in gaming performance from our earlier CV vowel night 1080p testing what we're looking at is can the 1702 also achieve that sort of frame rate we've put these chips against one another before in VR testing where our conclusion was that the GPU choice mattered far more since both CPUs can deliver 90 FPS equally well and this newest test is a less of a competition and more of a 10 the 1,700 do it to scenario the top of scene how it has features that make an attractive for casual streaming or rendering that doesn't mean customers that want to sacrifice smooth 144 Hertz and pure gaming scenarios there are applications or both before getting to those that this coverage is brought to you by EVGA and they're 1080p is c2 which we've recommended fairly highly for its build quality and the icx sensors which are kind of fun to play with you can check our full sc2 review for the 1080i if you're curious to learn more or you can click the link in the description below to find the product page for the 1080i sc2 so look at this test less of as a versus scenario and more of as a can the 1700 do it also because we know that the 7700 K weeds in unconstrained 1080p gaming scenarios we know that already we know that it's frame times are at least the same if not better in a lot of the games that we've tested maybe not appreciably so but objectively so and so now the question is to verify what we and other media outlets have been saying which is the 7700 K theoretically could be 144 Hertz better than the 1700 just like we validated our own previous statements about the 1700 doing streaming while gaming better turns out that was the case so we'll look at and see if the 144 Hertz scenario favours Intel list athena so for this test we're using the review sample r7 1700 all the testing methods the specs for the machines are in the article linked in description below and that includes the latest BIOS update for the crosshair motherboard and then we're using the i7 7700 K that we purchased from Silicon Lottery which actually is a cool site so commodity calm if you're interested so we bought one of those to go along with our other engineering sample which just was an engineering sample so it needs to be replaced this one can technically go up to 5.2 gigahertz for this benchmark because we haven't deleted it yet we are only going to 4.9 gigahertz when we overclock for the review sample 1700 were going to 3.9 gigahertz because that was the most stable with all the games and then otherwise running stock between the two CPUs so that's what we're looking at for the two processors and for the specs and the rest is in the article I stated for game if we chose five popular games that we expected to run at 144 frames at some level of graphics as in actually achievable at 1440p because that was ultimately the goal as 1440p 144 Hertz when we asked on Twitter and in this quarters patreon backers what frame rate they preferred and what resolution they preferred it on it was 1440p 144 that's the growing one 1080p 144 was second place but not anywhere close to leading so even at high resolutions we tested dual Mendota to rocket League overwatch and battlefield one those are games where you want the higher framerate again success here for the 1700 isn't defined as victory over the 7700 K it is defined as keeping up with the 7700 K just like success for the 7700 K and the streaming benchmarks would have been keeping up with a 1700 since we know that each one is already advantaged in its discipline against the newcomer which in this case is going to be AMD and the streaming one was the 7700 K Intel CPU so before diving into the results one final disclaimer here the point of this series we're doing these are follow-up tests from the reviews the point of these is to illustrate with a bit deeper testing on each specific topic that there are disciplines where each CPU as use-cases it's not quite as binary as by this one it's better for these two especially these two are 5s i5s gets a bit murkier but these two it's it's really they've each got good applications so we're trying to show here's where each one of them stands out and then hopefully you can take all that data and make your own assessment because we're not in a position to judge if the 70 to 100 K or the 1700 better fits your uses as a user that's up to you to determine but we're giving you all the data for each of the we'll call them a stereotypical use cases that people have brandished for each of these CPUs overwatch starts us off we use a special testing method for overwatch that's detailed in the article links below and was originally found when we did our overwatch performance optimization guide overwatch caps at 300 fps so we've got a decent lot of headroom for testing but we're not unlimited starting with Penna DP at max settings we're hitting 246 FPS average on both B stock and overclocked 7700 K and as detailed in the article below we're using a 1080 TI FTW so we're up against other limits here these two are effectively identical in performance and fall within our error bars indicated on the chart the r7 1700 is capable of achieving our 144 hurt to go although the low-end frame times do dips down below 144 Hertz if that matters to you this becomes a game of perception and subjectivity and speaking subjectively we don't much notice the difference there are certainly folks who think they can see one and if that's the case it's up to you to determine whether you fit into that crowd if so take note and by the appropriate CPU both of these CPUs can sustain 200 Hertz displays at 1080p is desired though the 7700 K is a much better option if higher quality settings at 240 Hertz are desired that is an insanely small market right now let's just be clear but the few who are truly fanatical enough about frames to really actually want 240 Hertz but wants out for these 7700 K in this particular title note of course that you can only really do this append ATP with these types of settings let's move on to 1440p have this resolution everything levels out to perform within a couple percentage points an average frame rate the 7700 K is tactic leading but it's close enough to be within our margins for this particularly long test the r7 1700 is consistently lower in frame times measurably and repeatedly though not in a manner which is appreciable both CPUs are capable of sustaining 144 Hertz at 14:40 P and so both past that bar minimally anyway here's for K just to show it we're completely within test margins here and can reliably state that this is completely GPU limited as you'd expect this has become a GPU benchmark at this point and is no longer a processor comparison we're using the same card for all tests so we're seeing the same for Florence the differences here are statistically insignificant and should not be read into further than effectively identical and within test variants moving on to dota 2 the core of our most recent 1,700 versus 7700 case streaming test was dota 2 so it's only fair that the game makes return here as a side note because of this 144 Hertz testing we discovered a typographical error in one of our streaming benchmark graphs pertaining to the 1,700 baseline performance the FPS that is that's been corrected in the article with a note though the conclusion remains unaffected basically the r7 1,700 baseline performance that when not streaming was lower than we originally thought but streaming performance was unaffected and remained identical since again just a typo on the baseline item anyway that cleared away now that we have a new test to look at let's start with 1080p and Ultra settings where we've manually maxed the game dota 2 shows clear favor to the i7 77 hard K and both its stock and overclocked configurations will later turn down some settings to try and achieve that 144 FPS marker as you can see the 7700 K is clearly ahead in this title under current conditions and again this is tested differently than our previous dota 2 test because we're using a different duration for the scene tested so you can't really compare the numbers we're at 174 FPS average on the 4.9 gigahertz 2700 K with a stock CPU at 164 FPS average one percent point one percent low metrics are always distance in dota 2 but are higher with the intel part than the 1700 the 1700 is clearly limiting GP performance here where we're seeing a 117 FPS average when overclocked to 3.9 gigahertz and 106 when and stock clocks those places the over 7700 que 49% ahead of the overclock 1700 with the spark 7700 K about 55% that of the 1700 stock let's move on to the star of the show 1440p at 144 Hertz the stock r7 1700 performs at about 106 FPS average roughly the same as we saw at 1080p indicating that we're still choking overclocking gets us roughly the same within test variants I 117 FPS average the i7 CPU is haven't really changed here either we'll see more of a change at 4k once the GPU bottleneck is instantiated but for now we're clearly Seaview bottleneck on all these parts we want to know what settings would be required to get the r7 1700 CPU up to 144 FPS and achieve our 1440p 144 goal so we drop the settings as indicated on the screen now with our low settings we're able to free up enough of the CPU to ascend to 149 FPS average one overclock and nearly 140 FPS average stock again this is by dropping basically every setting down to low or it's near level point that's getting about where our target is to dropping settings to low and overclocking pushes beyond 144 Hertz territory on the 1700 just to close this one out here's a 4k charged GPU bottlenecks finally enter play choking the 7700 K down to 156 FPS average the r7 1700 sits where it has been for the past two charts and the CPU is clearly the limiting factor here zoom is up next and stands as one of the lightest workloads we benchmark presently despite looking pretty good the game is well optimized to a point of being difficult to benchmark at times particularly given its 200 FPS physics bug that causes framerate to lock the closer we get to 200 FPS the less accurate our top end results will be as the frame rates are capped so that will drag the average down for this reason testing 1080p is pointless at 1440p with ultra settings and asynchronous compute on because they sync can be worse with anti-aliasing disabled these days before point 9 gigahertz 7700 K is chart-topping at 181 FPS average 141 fps 1% lows and 1 24 fps your bond recent lows the CP is beginning to bump it into the 200 FPS limiter in some scenes so this is truncated a bit the stock CPU runs at 178 FPS average with the overclocked 2.9 gig or at 1700 at 1:07 fps Ridge Emilie's which are proportionately scaled this place would be overclocked 7700 K about 8.4 percent ahead of the overclocked r7 1700 not accounting for the FPS cap the stock 1700 runs at 166 FPS average so we're really not gaining much from the overclock in this particular title at least with limited upward scaling regardless it's clear either CPU could reasonably achieve ultra settings at 1440p 144 Hertz and doom and so we meet our objective on both the r7 1700 and the i7 7700 k despite the latter is 8% lead both are capable of achieving the goal so depending on your uses you may well be ok with a 1,700 at 4k resolution just because that's what doom requires to become stressful all the CPU has become more bound by the GPU than anything else getting stuck at 92 to 97 FPS average the 77 hard K technically still holds a lead here but not an appreciable one the r7 1709 777 hard K are effectively identical and for Florence with regard to appreciable differences and neither is capable of holding 144 hurt we'd have to lower settings a resolution to get that rocket League is next n is an extremely undemanding game at least without modification its maximum frame rate is capped at 250 FPS out of box just like doom anything beyond 2 - the FPS will not be reflected in averages since it's not recorded so our numbers will be dragged down as we approach that cap at 1440p the stocks coming tomorrow DK was already at max - 49 FPS average let me dance usefulness in the comparison since there's no telling what it would be running at without that artificial cap the stock 1700 man is 203 FPS which indicates a cpu limitation further proven by our improvement in average FPS by 12.4% when overclocked to 3.9 gigahertz at 1440p we're able to achieve 144 Hertz playback actually 200 Hertz playback on the r7 1700 the 71 hard K carries tighter frame times for overall greater frame delivery consistency and so there's an argument to be made for keeping even low-end performance towards 140 FPS but the averages are passing on both the 1770 700k particularly after overclocking and are largely not really appreciate kind of higher than that anyway we've in the 4k the cat becomes less of a concern we stray further from 250 FPS but the CPUs are still about 144 average given the low intensity of rocket League with a 1080i the stubby semi-hard K is holding its lead by about 10 percent on average whether or not this is significant and advantages is up to you as the buyer and depends on your other use cases if you're not going to use it for anything else but this type of gaming it may be by the CPU that would best fit the scenario that be the 7700 K if you're planning to do other things consider the 1700 battlefield 1 is the last benchmark these tests are conducted differently from our standard battlefield benchmarks just like the dota 2 ones were here and so the data is not at all comparable to pass data this test is run using more intensive scenes involving heavier combat that's recorded over a longer period of time it's not quite what you'd get in the 64 player server so account for that but it is as close as we can reliably get without introducing a million uncontrollable variables by 64 of meeting players at 1080p the 77 hard K leads at 174 FPS average overclocked or 172 FPS average stock this positions the CPU beyond 144 Hertz territory with the rs7 1700 falling just below at 132 FPS average overclock and 123 FPS average stock low in frame times that sit behind the 7700 K I have 1440p ultra settings the 7700 K becomes more GPU bound and limits to 133 to 134 FPS average with results between the stock and overclocked variants outputting effectively equally link to the GB limitation the r7 1700 now operates in the range of 122 averages to 125 average regardless of vendor neither CPU is hitting 144 hurts at 1440p so to get that we have to drop down to 1440p and high settings with our lowered settings the 7700 cup to 152 152 FPS average now achieving our 144 Hertz gold and the r7 1700 sits around 139 FPS average when overclocked or 133 fps hours when stock it's tough to get much more performance out of this given that most graphics options in game center around GPU bound items but lower in geometric complexity and other led type effects would help boost framerate a bit where Intel can run a mix of high and ultra for 144 hurts at 1440p in our test environment again your mileage will vary and he needs to drop to a mix of medium and high and again the point of this coverage is to provide a look at the reality of the situation of these two CPUs they're both good CPUs but depending on what you're doing one of them may be less good for your use case so this hopefully along with the streaming benchmark and the VR benchmark helps show where each one shines in VR we saw no affective difference really at all even objectively looking at the numbers with really narrow margins you just you don't see a lot of difference with streaming there is a measurable difference certainly and the 1701 handily over the 7700 K with this test depending on the game we're seeing scenarios where well generally as a rule across all these titles the 77 RK is basically always in the lead how much of a lead depends on the game how much that lead matters depends on the target framerate in cases where we're going north of 200 FPS like rocket League or overwatch in some scenarios doesn't really matter if you're targeting 144 Hertz maybe not but if all you're doing is gaming the seventies when hard K gets you that much further past whatever FPS it is you want so it's entirely up to you whether that matters some people are frame rate nuts more than others and I mean that's really all there is to it if you're doing stuff alongside this high FPS gaming for example maybe you're a streamer and you don't want to use an external box for one reason or another and you don't want to use that envy encoder for one reason or another then consider the 1700 or something like that because as we've tested it is clearly superior in CPU encoding tasks while gaming and streaming simultaneously if you're not doing that options include things like blender animations or Premiere encoding or something like that where if you're doing something that's CPU accelerated it would probably be better to get the rise in CPU if you're doing something that's very heavily CUDA accelerated and your gaming otherwise the 7700 K is still good by so as you can see this is not me waffling between a decision between these two CPUs I'm not trying to appease everyone by saying both are good at different things because the reality is they are it's not an attempt to make friends with colleges on the Internet these CPUs are both good at different things is very complex which one is better and it depends entirely on what you're doing so the point of this is to say that despite what you see in comments they're both good at different things figure out what things you do take our numbers and others hopefully look around the web and find others and combine them to figure out what you should buy so at this point spectators in the comments are probably getting whiplash from crying chill for Andy and chill for Intel depend on which benchmark it is because as you can see it sways from AMD to Intel who wins the benchmark depend what task we're testing and wins there's and scare quotes because victory really depends on what else the CP is going to do if it's only gaming well I guess Intel wins here if it's live streaming and gaming and the wins there so it boils down the wins and losses there's some of your wins and losses though is that you can find more information in the article below hopefully this helps paint the picture that yes things are quite complicated with silicon and semiconductors and processors and at this point you should have enough data to put together some thoughts on whether you need the 1700 or the 72 100k for this type of workload and which one benefits you more for other types of workloads you can check our full reviews or our other follow up coverage for benchmarks outside of just gaming subscribe for more patreon.com slash gamers next ourselves not directly you can go to Garren's access at squarespace.com and pick up a shirt like this one and I will see you all next time let's move on to the start of the short start of the shores