Gadgetory

and our recent trend of enraging one-third of the Internet we're back today to review the AMD it rise in r7 1,700 CPU we delayed it by a few days so that we could really take in the positivity of some comments stemming from our first 1800 X review now that we're back in this 1700 review were primarily looking at the price to performance of the 330 ish dollar and the r7 1700 and comparing that in positioning with the company's marketing materials and with the competition of course which would be the 7700 K at a similar price though we have plenty of other CDs in the stack before getting to that this coverage is brought to you by Thermaltake makers of the new compact silent 12 cooler that's priced at $25 and includes a m-- for mounting supports the contact silent 12 offers budget performance with low at noise and a low price link in the description below if you want to learn more about this cooler their eyes on r7 1700 cpu is an 8 cores 16 thread unit with the same CCX layout as the 1800 X and 1700 X it's really the same in most regards the cache the architecture of course is the same so you're mostly looking at the same thing other than the frequency and the TDP TDP is lower by about 30 watts looking at 65 versus 95 on the to higher skewed chip and then the price $330 versus about 400 for the 1700 X and about 500 for the 1800 X for the unfamiliar you can read or watch our original reviews for a straight-shooter recap of the architecture the r7 1700 ships with a stock frequency of 3 gigahertz and boosts to 3.7 under the right conditions mostly limited thread scenarios the CPU we received it did not include the race cooler though we just purchased our own 1700 and are awaiting its arrival so that we can independently test the cooling solution on the topic of cooling the 1700 requires a much less powerful solution than the 1800 X largely thanks lower voltage and frequency on the 1700 TDP again is 65 watch the 1700 to 1800 X and 700 X both at 95 watts but with higher native frequencies memory support is the same on paper just with the caveat that we had greater memory clock in limitations on these 1770 1800 X even with the same platform we were only able to hit 26 66 megahertz out of box even using these and kit and motherboard as in the 1800 extra view this indicates that there's some sort of memory controller limitation on the r7 1700 with our model though we've spoken with other reviewers and board partners and found out that it seems to just vary depending on which CPU you have and by that I mean individual CPU to not skew an hour later overclocking tests we were able to jump back up to 29 33 megahertz without issue further validating that this is likely a stability and bandwidth issue at the lower CPU frequencies we're still testing on the ACS crosshair motherboard still with the latest stable EFI release and that's version of five seven zero four at time of shooting we've begun testing on gigabyte boards as well and have largely seen no difference in performance that will publish those numbers in a separate content piece so that we can have some comparisons between boards and platforms again so far between the two not looking like a huge difference if you're looking to learn more about the 1800 X and our thoughts on it links in the description below and we also have a link to this review the written version in the description below that will include the testing methodology for this if you're curious about what components were used for each of these CPS on the bench its detailed there we also have some thermal test notes their power blender Premiere gaming all that stuff is detailed in the article starting first with temperature over time we see temperatures drastically reduced versus the r7 1800 x with both CPUs using our kraken x 62 at max fan and pump RPM the r7 1700 sustains a completely flat clock rate under load other than one hitch and the 1800 x by comparison had some fluctuation although small during the torture testing this is largely due to heat and seen as b-17 100 it constantly sits below 50 C with this cooler generally closer to 45 it's easy to see why clock fluctuations aren't happening on the r7 1700 skew CPU far easier to keep this thing thermally controlled and again you can learn more about those thermal tests in the article looking now at total system power draw and voltage stats we observed the r7 1700 operating at 133 Watts peak draw for the complete system when under stock settings and with a blender workload that's compared to 185 watts total system power draw of the r7 18 that that said note again that our 1800 exit did run 29 33 megahertz memory at all times during testing whereas the current 1700 CP that we have was limited to 20 666 unless overclock the 1700 push is a core averaged core clock around 3200 mega Hertz when left to auto in blender with a V core of 1.0 six eight volts Cinebench single-threaded we see power draw at 80 watts total system versus the compare the total system powers of 92 us on the 1800 X with CPU speed bouncing between thirty two hundred and thirty seven fifty megahertz this is because again boost and XF are on the AMD CPUs only kick in when running limited thread workloads it does not enable for workloads at leveraging all threads like blender voltage is around one point two eight six to 1.30 eight V core for this particular test but pov-ray rounding us out at 37 50 megahertz clock rate for a single threaded run with voltage stable at one point two eight six this is lower than the 1800 X contributing to lower temperatures overall we run our power tests in balance mode for a few reasons but the performance test that would be production blender premier and gaming are run in performance mode as explained in our first article there is a limitation of performance because of some core parking issues when running and balanced and so we run performance for the gaming and workload tests and for power draw volt it seems like that we just decided to do balance because really running performance in those looks a bit extreme in some cases so we're kind of hoping that you can pick through the numbers and figure out how you're going to use the CPU generally we would hope that you could use balance mode for the most part because does consume a lot less power let's your clock rate drop lower when idle but there needs to be some more updates before that's the best option just because it does limit gaming by about 4% an hour testing moving on to production our blender benchmark uses an in-house render scene that renders at 400 samples per pixel with a 4k resolution and 16 by 16 tile sizes for CPU benchmarking this scene is used across all CTS we test and represents a real-world use case for a release candidate seen as created by GM's animator as shown here the r7 1700 stock SMT enabled completes the scene in about 33 minutes this is between the 77 100 K overclocked and the r7 1800 X or 69 hundred KS stock CPUs that's about where we'd expect these 1702 Falls given the 1800 X's performance but what's interesting is the overclocking aspect assuming you're willing to put in the five minutes to overclock and run a little higher temperature the 1700 is able to complete the scene in the same time as the overclocked 1800 X as both at land at around 3.9 to 4.0 gigahertz in our production test we were able to sustain a 4.0 gigahertz clock in gaming for the 1700 but production workloads that stress stability enough and forced us to drop to 3.9 which is where the 1800 X overclocked again if you're willing to overclock and aren't concerned about IT approval the 1700 beats of the stock 1800 X and effectively ties the overclocked variant note that our OC on the 1700 bumps memories of 29:33 since we can only support 2666 stable without on EOC this more or less invalidates the existence of the 1800 X if you are okay with overclock in and of strictly considering its for CPU accelerated Bunder tasks though a GP will still get you further faster in most these cases particularly the blender benchmark we use just like we said in the 1800 X review and the deserves praise for its production positioning versus Intel fame or double cost alternative also just like we said in the 1800 X review we'd recommend the 1800 X over the 6900 k4c be targeted or software accelerated production and again that's nearly a direct quote from the last content piece but now we would recommend the 1700 over the 1800 X it's an even better value for CPU accelerated production if you do any but does need an overclock to kick it into gear the overclock is trivial and the cooling requirement isn't all that high otherwise for the most part blender and premier are happening on the GPU though other workloads like compiles should you work with them would benefit from CPU acceleration that said most of our audience is going to be the gaming type and if they do any production at all it's probably YouTube content creation or streaming so we're keeping those audiences primarily in mind when working on this content our Adobe premier benchmark is conducted by rendering our EVGA icx review representing a real-world workload filled with color Corrections warp stabiliser transforms things like that and this benchmark the sixty nine hundred K leads one over box with the 18-hundred accent leading these 5900 case talk marginally again both stock now as stated at last time the marginal lead is made less marginal when considering that we have cost positioning that said if we look at the our 717 one hundred we're performing ahead of these heavy 700k by about thirty minutes which is a considerable gain when looking at the near price equivalents slightly favoring a and b this is finally starting to look like the Rison processor that should have champion to launch despite initial shipment and branding of the 1800 x overclock at least 1700 again outperforms the 1800 x or equals the overclocked 1800 x with equal memory speeds given the $170 price reduction from the 1800 x we see no reason for our audience or mixed workload users to buy the 1800 x the 1700 is far and away a better deal now again premier is cuda accelerated so you generally want to do those workloads on a GPU that said the price positioning is much stronger versus adjacent intel competition and finally starts to build a case for Rison in our reviews users who do cpu accelerated workloads should of course check benchmarks for their relevant software as we don't test all of it but could generally postulate that the 1700 would outperform an i7 7700 k stock in software accelerated tasks again check reviews for the actual software you use it doesn't all behave the same way with regard to multi-threading and frequency we've got a few other synthetic workloads in the article like Cinebench fire strike and pine spy you can check that link below if you want those let's move on to games this is where the 1800 X fell short in our last review it ultimately performed on par with the i5 CPUs some from older generations than others depending on the game and was priced about two times what an i5 is so for pure game performance it's a bad Buy and we say the same about the 69 hundred K by the way given me seventeen hundred's it's significantly lower price the stack up should be a lot more interesting now of course it's not going to outperform the 1800 X when both our stock but the price is still $170 lower so the competition should be more comparable before getting to that this is a slide from Andy's own marketing materials for our X 480 and Polaris launched as the slide suggests 95% of PC gamers use 1080p or below so for that reason and for methodological reasons like actually finding these CP limitations in gaming we are primarily using 1080p for benchmarking now we do have a few 1440p tests just like we had for the 1800 x but you'll see that that basically makes all of the CP is the same including the 1700 and the 1800 x so based on this slide most of the market is still on 1080p or below as stated by AMD as always for full methodology the links in the description below we're starting with battlefield 1 and battlefield 1 the and the r7 1700 SMT on operates at around 129 FPS average or 85 1% lows with 68 fps 0.1% Lois this is sandwiched between the i5 4690k stock and I 535 70 K stock all three of which have more or less identical loads at with regard to user perception now again that's not great price to performance release 1700 but it is a whole lot easier to justify as a product than the 1800 X for gaming if we're looking at pure gaming workloads we'll get to that more in the conclusion though overclocking the r7 1700 to 4.0 gigahertz which was stable with games but not production on our model and with an increased memory speeds of 29 33 the CPU is now outperforming the stock 1800 X or performing equally to the 1800 X overclock the 1800 X overclocked 3.9 gigahertz was able to sustain a more stable clock rate with a 1700 fluctuated more frequently and so places higher and it's low values at 109 and 94 verses 89 and 75 on the 1700 the higher sustained clock rate is primarily a benefit of the additional power throughput higher voltage and higher baseline of the 1800 X if you okay it with the higher frame times on the low-end higher frame times being a bad thing the 1700 overclock is clearly significantly better in value because although I our frames arms are not desirable it's not a huge difference in this instant the average at frame rate is imperceptibly different as well furthering the argument for the 1700 overclocked over the 1800 X stock or overclock that makes the 1700 a whole lot better than the 1800 X in terms of price to performance for gaming but it's still on par with Intel i5 CP is last generation comparing against the i-5 60 600k the 1700 stock performs measurably worse across all three metrics but the perception isn't different as we said last time so despite lower values overall you probably won't detect them as a user one potential argument here is that the 1700 includes of stock coolers so if you're ok with using that it has some value that we haven't tested it yet we're waiting on our retail model to come in and in theory the motherboards should also be cheaper once they aim for market saturates we wouldn't recommend using any motherboard for overclocking because some of these VRMs are not that great but keep an eye out on our channel and the website for recommendations on which boards are ok for that so between the boards and the stock cooler that brings the price closer to an i-5 but we're not quite there we're close but the significant gain is in production and we absolutely would not recommend an i5 CPU for production rendering because the thread limitations are just too great the 1702 helps fill that mixed workload gap where the 1,800 exa fell short for gamers at its price last time let's move on to total war hammer as a reminder total war Warhammer has some issues with measurement of 0.1% low values a game is more variable and its output to begin with but it's also a lot more sensitive to Glocks ability on CPUs and given the CPU bottleneck that's why it's a great TV benchmark it just so happens that over clocks will really show their limits here if you're on the border of instability and the game is also just generally less reliable in its low-end frame latency that stated the results placed the r7 1700 stock at about the AI 370 to 50k performance range sandwiched between the i5 2500 k overclocked to 4.5 or if you prefer stock the i3 6300 and the i3 is 22 K again stock its measurably slower than the 800 X given the lower base and boost frequencies and generally not that competitive in terms of price to performance for a pure gaming build of course it's also got that production benefits that's much easier to justify for streaming and content creators at its price point and we still don't recommend the 70 to 50 K at its price just like we didn't when we reviewed it but the point stands at least 1700 isn't all that impressive in terms of raw FPS throughput the 1700 does have it better 1% low values which are a bit more mad be consistent in this game then its neighboring Intel CPUs so that is an upshot the overclock that variance is a lot more interesting particularly because again this is a trivial overclock for our CPU your mileage may vary but ours hits 4.0 easily for gaming workloads the performance is now at 136 average with high sustained one present values the TD plants itself just under an overclocked 72 stood K ahead 5.0 gigahertz and out matches the 1800 X stock CPU so we're around the 70 to 50 K overclock and an 800 X which is a reasonable improvement for the increase the extra 100 megahertz on the clock also helps overcome the overclocked 1800 X since ours was limited to 3.9 gigahertz this again pushes the 1800 X further out of recommendations for gaming builds with the 1700 d SERP II and any potential it had for most of our audience we'll go ahead and highlight as some key off performance before moving on for does significantly improve overall throughput with this particular game we're not focusing in on it as much as the initial review watch starts to is heavily multi-threaded but does push more GPU load then in total war Warhammer without an overclock the stock r7 1700 performs ahead of the i5 6600 K in average frame rate matching the 1800 x stock and it's measurably but imperceptibly worse frame Layton sees this performance puts the $330 r7 1700 under the $240 i5 7600 K when both are stock momentarily ignoring other cost factors in the builds we don't yet know how good that stock cooler is overclocking pushes the 1700 past the i-5 7600 K and close to the 7700 K with hyper-threading off hyper threading on we see a big performance advantage for Intel's architecture and $340 7700 K which is now performing at 113 FBS average to Andes 86 FPS average overclocked or 80 FPS average without the overclock regarding the 1800 X were able to effectively match that CPU and overclock in the 1700 and we outperform z stock 1800 X as expected for a few more games including Metro last light and GTA 5 there in the article below as you'd probably guess but we have two benchmarks from 1440p before moving on just like we included 1440p benchmarks in our previous contents will include them again keep in mind that pushing it more pixels inherently creates more load on the GPU and with these higher settings we're looking at a GPU bottleneck that makes all CPUs look more or less equal including Andy's own a product stack these charts are limited as we just added them specifically for the 1800 X and for actual CD performance limitations you'll want to look at those 1080p benchmarks as those gives the best idea for Raw perform from scaling within gaming as GPS advanced regardless let's look at a few 1440p numbers following and these arguments of representing CPUs at higher resolutions and thus introducing GPU bottlenecks we see that the 1800 X is effectively irrelevant as a gaming option when GP limited since the 1800 X and 1700 produced results which are perceptibly equivalent again this was a MV suggestion so following the path of imposing external bottlenecks e 1700 is equal to an 1800 X in gaming when bumping against the choke point and with the 7700 K outperformed both by about 10 FPS average and in low values the 6900 K as it has been remains bad value for anyone who only games now with battlefields one at 1440p we see again that the stock 1700 and stock 1800 X are equal in performance thanks to the imposition of that GP limitation with both slightly tailing the Intel CPUs the and the r7 1700 is priced in a way that makes it a lot better at the cost than the 1800 X in the case of the r7 1700 we see mixed workload use cases at shine at the $330 price point where the CPU definitely outperforms the 7700 K in premier and blender rendering tasks when the CPU accelerated if you're pushing renders the CPU and doing some gaming it's not a bad buy it just depends on how much production you do versus how much gaming or how much you care about hitting high refresh rates and games in which case you'd clearly be better off with the 7700 k if 144 Hertz at 1080p is what you care about this positioning further reinforces our stance on the 1800 X you are far better off buying an r7 1700 for about $330 or if you can find it without a cooler even better it might be a bit cheaper and applying a five-minute overclock with a half-decent cooler than buying the 1800 X you will net a better product in turn of price of performance and really perform about the same in gaming and in production as long as you are okay with overclocking which isn't difficult with these and sustaining that overclock for whatever lifespan of the system you are planning to use just means you need a halfway-decent cooler and you need a few minutes looking at guides to make sure you're applying correct voltages if the correct voltage manual tuning is required which you should generally do because auto often goes a bit overboard so there's really no reason to buy an 1800 X if you are okay with that scenario now we normally assume that most folks aren't overclocking that's generally my assumption is we see a lot of people who pick something like a 7700 K because it's the best not necessarily overclocking it so that may be the case here as well but with a 1700 it is actually worth doing it's not hard it doesn't change really a whole lot versus Intel if you're used to Intel overclocking same concept the only point of hesitation and saying this is the future bending of these CPUs we're not sure if the early 1700 models will OC hire or not or if they're modified 1800 X parts that retain the same Headroom our sample overclocked to 3.9 or 4.0 depending on a workload and that's what puts it on par with the 1800 X where outperforms it depending on how IV 18 or Dex clock that also allows us to clock the memory higher which is not great in the stock settings so another point in favor of overclocking if yours is limited and we know some other folks like Paul from Paul's hardware and bit wit were able to achieve similar overclock on their review sample r7 1700 CPUs we're still waiting it for our 1700 to come in the purchased one from Amazon and that one of course will try to overclock see if it does the same and it's still from the first round of production so obviously keep that in mind we'll also be able to look at the stock cooler at that point and see how good the cooler is if it's actually worth using or if you should purchase a standalone one anyway in the past with the Wraith cooler was genuinely a good effort by AMD we gave it a great review so if it's the same thing here it might be just fine of course the aftermarket ones do tend to be a bit better in terms of price performance but if you're trying to save 30 bucks now they have not been that in the past we'll see how this one is because the rise in series is a is new so wait for that so again we strongly advise against the 1,800 x as a cpu for a gaming machine being gamers Nexus that is what we focus on if you're doing zero production you're not doing any content creation or anything like that then you're really still better off in terms of price to performance for gaming with an i5 or an i7 CPU particularly at high refresh rates with 1080p that said for people who are combining content creation like we do with a gaming or maybe considering streaming the r7 1700 is actually a viable chip and far more so than the 1800 X given the price and the overclocking potential to equalize it to and then in terms of 1440p performance the two chips are the same assuming you have something like a GTX 1080 and you're bottlenecked it looks the same now of course they're not actually the same because we have a GPU bottleneck so when the higher end cards come out that may change but that's what it looks like now and that's also why we favor attended yuuki benchmarking we still have to look at these 1700 X but from early results it looks like the r7 1700 is the hero of AMD's lineup this looks a whole lot better price to performance for our audience even other audiences considering overclock potential if you can overclock not all use cases allow it it's far better than the 800 X in that regard so it's easier to argue this CPU just know that you're still going to be limited in terms of total frame rate throughput in games versus compared Lee price Intel CPUs so if you're okay with that trade-off it's really not bad we can actually stand behind the r7 1,700 under the correct usage conditions and those are going to be mixed workload conditions where you do some production that is CPU accelerated maybe streaming in addition to gaming if you're just gaming is still going to be a cheaper CPU or an equivalently price i7 as your better bet in terms of raw frame throughput but in those mix workloads the 1700 it really does thrash the i5 and even does a pretty good job beating the i7 though isn't as they're threshing in production workloads that are secu accelerated so that's all we have to say for this one thank you for watching subscribe as always I'll see you all next time you