AMD Ryzen 7 1700X Review

Power Consumption And Temperatures

Direct Comparisons of Power Consumption

Similar to our performance numbers, we re-ran the 1800X review's power data to reflect the continual improvements being made to motherboard firmware.

Let's start by looking at idle power consumption. Intel's overclocked Core i7-6900K turns in a better result than the stock configuration because we also reduced the one-core Turbo Boost frequency.

The 95W AMD CPUs clearly use less power in our combined CAD benchmark. When we weigh average performance against power consumption, however, the eight-core CPUs from AMD and Intel are fairly similar.

The same story applies to our measurements in games, where Ryzen 7 blows away the Core i7s. It's been a long time since AMD was at least equal in terms of efficiency.

The 3.8 GHz Ryzen 7 sucks down more than 140W , but the Core i7-6900K is even worse at 166W. The Core i7-7700K down-clocked to the same frequency uses a conservative 86W or so.

Based on AMD's technology briefings, we know it has more granular control over clock rate. And it's notable that the 95W Ryzen 7 1700X we're reviewing today uses less power under our stress test than Intel's 91W Core i7-7700K. It takes a significant underclock to put the Kaby Lake flagship in first place.

Temperatures

We optimized our CPU cooler for Socket AM4 by using two nuts between the spring and bracket to increase force on the package to 0.4Nm. That is why these results differ from those in our launch article, where we only used washers.

The temperatures we recorded for the FX-9590 are a bit uncertain, since AMD’s older Bulldozer CPUs don't measure with 100% confidence. Moreover, the Ryzen 7 and Core i7 CPU readings aren't exactly comparable; both companies employ different sensor approaches.

Intel's Core i7-7700K is the only processor in our test field handicapped by cheap thermal paste between its die and heat spreader. Thankfully, AMD solders Ryzen's heat spreader, which results in good thermal transfer. This naturally shows up in the relationship between power converted to heat.

Enthusiasts should be happy with Ryzen's power consumption and its resulting waste heat.

This thread is closed for comments
122 comments
    Your comment
  • Ergosum
    Whoever runs PR at AMD needs to reevaluate their methods. The 1800's and 1700's are very solid products, but for some reason were shadow marketed--letting rumor define the target application sets.

    AMD should have had a strong positive campaign on where these chips do well. Specific. Timely. They would have gathered some gamers who wanted to brag about handbrake performance or some-such. Instead they let the market build fairy castles in the sky about gaming-specific performance, and so (again) lost a great deal of goodwill and trust.
  • ykki
    Tom's please update the win 10 pro version number. "All updates" doesn't tell crap.
  • hannibal
    The problem is that the market made those cloud castless, not the AMD... it is hard to AMD to say users not to speculate. Every information that AMD did give out was confirming that Ryzen was going to be really good multicore performer, and we all know that very few game use more that teo or three threats, so the extra 4-6 cores that Ryzen has normally Are useless I. The games, so any Intel 2-4 core prosessor wa going to better in the games if They would run in higher freguences that Ryzen and AMD very clearly tell us that 3.8 was the very near the top of the prosessor speed.

    If and when games start supporting 8 cores the 1700 is super good perfomer in the games too, but if and most propably because the situation stays the same. 2-4 more powerfull cores is always better in games that having more of them.
    It seems that people now know games really poorly if They expected the Ryzen has any chance in those.
    Ryzen 1500 (four cores) is as fast in the games than 1800X is and 1500 is much cheaper.
    http://www.techspot.com/review/1360-amd-ryzen-5-1600x-1500x-gaming/
  • TechyInAZ
    Good review. Can't wait until software devs and motherboard manufactuer's get better optimized softwares and BIOS's so Ryzen isn't constantly dealing with this problem anymore.
  • Conclusion: Ryzen still sucks.
  • dstarr3
    2164959 said:
    Conclusion: Ryzen still sucks.


    It's not that Ryzen sucks, it's that Ryzen was meant to compete with Xeon CPUs, which it does very well. Gamers should look elsewhere, but for some reason, gamers are the ones that got most excited about these CPUs.
  • elbert
    I wouldn't suggest either a 1070 nor the 1080 on 1080p. Probably should have tested with a more normal 1060 6GB at that resolution. Those willing to pay high prices for both the CPU and GPU should be running at 1440p. The game benchmarks are unnaturally skewed.
    The biggest thing to look at in the game benchmarks is how unutilized the Ryzen CPU's are. We will probably see the Ryzen R5 quads running the same FPS for a few hundered less then all but the old 8350 on April the 11th.
    Also it looks like AMD is gearing up a new socket with 8, 12, and 16 cores on an X390 motherboard. At the current pricing the 16 core could be lower priced than Intel's over priced 6900 8 core.
  • Achaios
    ?ll this talk about games supporting more than 4 cores is really a COMPLETE WASTE OF TIME.

    46.19% of Steam Gamers own 2-Core CPU's and 47.74% of Steam Gamers own 4 Core CPUs. No Game Studio is going to waste resources and money on optimizing games for more than 4 cores in the foreseeable future, if ever.

    According to the same survey, only 0.24% of gamers own octacore CPU's. A little less than this is the presence of AMD Ryzen CPU's in the gaming market. Almost nonexistent.

    I wish I was trolling.
  • Oranthal
    Why did you only test 1080p performance? Most people who would consider these CPUs run an RX-480 or greater GPU if they are gaming ($300+ is still the premium market). Other reviews show the gaps shrinking or disappearing with greater resolutions as the bottlenecks are at the GPU where most die-hard review reading gamers will actually be limited.
  • Oranthal
    45049 said:
    I wouldn't suggest either a 1070 nor the 1080 on 1080p. Probably should have tested with a more normal 1060 6GB at that resolution. Those willing to pay high prices for both the CPU and GPU should be running at 1440p. The game benchmarks are unnaturally skewed. The biggest thing to look at in the game benchmarks is how unutilized the Ryzen CPU's are. We will probably see the Ryzen R5 quads running the same FPS for a few hundered less then all but the old 8350 on April the 11th. Also it looks like AMD is gearing up a new socket with 8, 12, and 16 cores on an X390 motherboard. At the current pricing the 16 core could be lower priced than Intel's over priced 6900 8 core.


    100% spot on. I would of skipped my comment had I seen this when I started writing it.
  • Oranthal
    1328515 said:
    ?ll this talk about games supporting more than 4 cores is really a COMPLETE WASTE OF TIME. 46.19% of Steam Gamers own 2-Core CPU's and 47.74% of Steam Gamers own 4 Core CPUs. No Game Studio is going to waste resources and money on optimizing games for more than 4 cores in the foreseeable future, if ever. According to the same survey, only 0.24% of gamers own octacore CPU's. A little less than this is the presence of AMD Ryzen CPU's in the gaming market. Almost nonexistent. I wish I was trolling.


    Not a waste of time, core counts will steadily increase over time in nearly all devices. AMD is pushing the price down on x86 CPUs and making it happen sooner. Most Steam users cant buy the R7 CPUs and most people are not playing with max settings on even Dota 2. Performance costs money that most gamers don't have. The benchmark games are almost always the most demanding not the most played to give us an idea of what the max is for hardware. Witcher 3 was probably played more on console at 720p looking like crap. Still doesn't change the point of testing max performance. Which takes me to the worst part of this review. Tom's stuck to 1080p with $1,500+ test rigs so they failed to properly test it.
  • dstarr3
    1328515 said:
    ?ll this talk about games supporting more than 4 cores is really a COMPLETE WASTE OF TIME. 46.19% of Steam Gamers own 2-Core CPU's and 47.74% of Steam Gamers own 4 Core CPUs. No Game Studio is going to waste resources and money on optimizing games for more than 4 cores in the foreseeable future, if ever. According to the same survey, only 0.24% of gamers own octacore CPU's. A little less than this is the presence of AMD Ryzen CPU's in the gaming market. Almost nonexistent. I wish I was trolling.


    First of all, gaming isn't the only reason people buy computers.

    Secondly, I've got an i7, four cores, eight threads. And looking at AIDA64, never are any of the threads at 0% usage. Computers are already always using every resource it has to at least some extent, and software is only going to leverage high thread counts more and more. There is zero reason to stagnate and stop introducing a higher core count, because each new core and thread you introduce, it will be used.
  • KAILOONE
    If only Ryzen sold in Mexico, I'm very sure Ryzen will get better over time, optimizations will help alot and some bug fixes and better RAM support will make this chip shine in a couple of months.
  • Brian_R170
    The problem with waiting for the software ecosystem to catch up is that it takes a long time. Both the releasing of new apps and optimization of existing apps will be slow. The change to the ecosystem takes years, not months. Buying a CPU based on what might change is a risk that buyers need to consider because by the time the change happens, newer CPUs from both AMD and Intel will be available.
  • dusty13
    this article would be more meaningful if it had cpu-usage statistics along with the fps.

    that would actually give us some indicator on actual performance and headroom.


    generally though:

    none of the games tested here really fall below 50 fps.

    rise of the tomb raider is lagging behind 35-50fps ... yeah. do tell how that is relevant with a minimum fps of 73 and the max being over 100 though.

    what articles like these constantly fail to address is that they are talking about gaming performance in current games on 1080p to construct a picture with visibly different graphs.

    on that level however you will not see a difference. for all real life intents and purposes a 7700k or a 1700 will give you the same gaming experience, especially since 1080p gaming usually means entry level to lower midrange and that still mostly happens on 60fps monitors.

    on the other end of the spectrum if you go for a 1440p/144Hz freesync/gsync or 4k monitor and pro gaming setup, you will end up with identical performance due to gpu bound situations.


    so bottom line is, the differences you manage to produce in graphs are not noticeable for the average joe who plays on 1080p and do not exist for highend gamers.


    that leaves only other workloads - but there ryzen usually roflstomps the 7700k and what? future games? not really useful since that can only be speculation, so we can not really say anything meaningful either.

    so for today you will not experience any difference in acutal gameplay no matter what system you use and we can only speculate about optimizations and future games ... although amd hardware has one thing going for it. it is the basis for the two main gaming consoles, one of which runs windows10 and will be refreshed for 4k. even with intel being definitely the bigger player, this should at least somewhat level the playing field as game studios will in essence have titles from xbox who are already heavily optimized for amd on windows 10 with 4k at hand in many cases in the coming year.
  • 10tacle
    2277213 said:
    Why did you only test 1080p performance? Most people who would consider these CPUs run an RX-480 or greater GPU if they are gaming ($300+ is still the premium market). Other reviews show the gaps shrinking or disappearing with greater resolutions as the bottlenecks are at the GPU where most die-hard review reading gamers will actually be limited.


    You answered your own question: because a lower resolution shows the true power of the CPU. But besides that, well over half of PC gamers still run 1080p monitors, with an increasing popularity of users preferring faster frame rates at 1080p via FreeSync and Gsync over slower frames on a higher resolution monitor.

    The 1440p QHD users like me are still a minority niche market, and 4K UHD users an even a smaller market. And moving up to a 1440p 144Hz Gsync/FreeSync monitor will require a *lot* of GPU money to run most games today - even some older games like W3 and Far Cry 4 - even remotely above 100FPS to take full advantage of its capabilities (1070 SLI minimum).
  • spdragoo
    45049 said:
    I wouldn't suggest either a 1070 nor the 1080 on 1080p. Probably should have tested with a more normal 1060 6GB at that resolution. Those willing to pay high prices for both the CPU and GPU should be running at 1440p. The game benchmarks are unnaturally skewed. The biggest thing to look at in the game benchmarks is how unutilized the Ryzen CPU's are. We will probably see the Ryzen R5 quads running the same FPS for a few hundered less then all but the old 8350 on April the 11th. Also it looks like AMD is gearing up a new socket with 8, 12, and 16 cores on an X390 motherboard. At the current pricing the 16 core could be lower priced than Intel's over priced 6900 8 core.


    http://www.techspot.com/news/68407-tackling-subject-gpu-bottlenecking-cpu-gaming-benchmarks-using.html

    1080p may be the most common resolution that gamers use, but it's not the most demanding. By using a top-of-the-line GPU, the testers know that the GPU won't be holding the performance back (same reason why they've tested with 16GB of RAM for some time, even though you still really only need 8GB on 64-bit Windows for gaming). They may vary that somewhat depending on the game -- Techspot's Dark Souls III testing, for example, had to go up to 1440p for their CPU benchmarking (http://www.techspot.com/review/1162-dark-souls-3-benchmarks/page5.html), because the PC version was artificially frame-capped at 60FPS, & their GPU testing showed that you had to go down to a 4th- or 5th-tier GPU (like a GTX 780 or R9 380X) to start dropping below that cap with the i7-6700K.

    By using a powerful GPU at a resolution where it's going to idle along without any trouble, the testing allows them to definitively show how a given CPU (in this case, the Ryzen CPUs) perform against other chips.
  • InvalidError
    1328515 said:
    46.19% of Steam Gamers own 2-Core CPU's and 47.74% of Steam Gamers own 4 Core CPUs. According to the same survey, only 0.24% of gamers own octacore CPU's.

    And what is the reason for that? Because quad-core CPUs that performed well in games used to cost over $200 and octo-core ones were out of most people's budget. This may change in the near-future with Ryzen putting AMD back on the map. If AMD hadn't dropped out for over four years, it would have happened sooner.

    While the majority of games may never scale much beyond two cores, the number of games that do will increase as PCs with 4+ cores become more affordable, just like the number of games that scale to 8+ threads has been slowly increasing.
  • none12345
    If you dont overclock, the 1700x and 1800x make sense. If you do, there is little sense over the 1700 right now.

    I have a 1700x and i dont regret buying it. This thing is an utter beast. BUT, if i had to do it over i would have bought a 1700. I thought there would be a larger difference in chips, so i preordered a 1700x when preorders opened. That was the risk with preordering, but i was done waiting.

    That said, i ended up getting my 1700x for $350 instead of $400. Credit card price protection for the win(they matched a sale price). So, really i didnt spend much more buying a 1700x instead of 1700.

    Tho, i coulda also saved on the cooler i bought. The stock cooler on the 1700 seems to be very good. The noctua nh-15d i bought is utterly overkill. But, at least i can run the fans very slow on this thing and be really silent, so it wasnt a waste.

    If you do any content creation, and need more cores. Ryzen is the utter king right now, intel just cant compete. It can compete on performance, but not on price, you can get the same or better perf here on ryzen for much less cost, so its a no brainer choice to get an 8 core ryzen over 8 core intel.

    In real resolutions, its a great gaming chip too. Its not the utter fps king at low resolutions tho, that is still owned by the 7700k(no chip can match its single thread clock speed, the intel 8 cores cant do it either). But for everything else, ryzen is a great gaming chip.

    Ive had butter smooth gameplay since upgrading, tho im also 100% gpu limited, ryzen doesnt even break a sweat.

    If you only game tho, i wouldnt buy a 8 core chip in the first place. Tho the 6 and 4 core R5s will be great gaming chips for the price when they come out in 2 weeks. Those chips will pretty much completely obsolese the intel i5 line. i3 would only still exist for its price point.
  • 10tacle
    As a long time Intel builder, I am happy to see AMD finally take a real swing back at them. Just from architecture previews the past several weeks I figured Ryzen would be a better performer in certain productivity apps. I'm also pleasantly surprised at the lower power use and thermals. Which brings me to:

    "Intel's Core i7-7700K is the only processor in our test field handicapped by cheap thermal paste between its die and heat spreader."

    WTH is Intel doing? De-lidding classes are now open and available for Intel 7700K chip buyers! Good grief. Is this only for certain Intel 7xxx models and others get better quality compound?
  • spdragoo
    1328515 said:
    ?ll this talk about games supporting more than 4 cores is really a COMPLETE WASTE OF TIME. 46.19% of Steam Gamers own 2-Core CPU's and 47.74% of Steam Gamers own 4 Core CPUs. No Game Studio is going to waste resources and money on optimizing games for more than 4 cores in the foreseeable future, if ever. According to the same survey, only 0.24% of gamers own octacore CPU's. A little less than this is the presence of AMD Ryzen CPU's in the gaming market. Almost nonexistent. I wish I was trolling.


    Steam is far from the only platform out there. But there are a number of games out there that won't launch at all on a dual-core system, & plenty of other games that are "playable" only in the barest sense of the word (i.e. they can maybe reach 30FPS, or have such low minimum rates and/or near-100% CPU usage that they're constantly stuttering). The Dark Souls III results I mentioned in my prior post, for example, indicated that the dual-core Pentium G3258 "doesn't look too bad with a minimum of 37fps, but unfortunately with a constant 100% load on the CPU there was quite a bit of stuttering."

    Nor are all Steam games created equal. For example, my Probook 4520s has no trouble with Earth Defense Force: Insect Armageddon (available on Steam), because its requirements are fairly low (so a 2C/4T CPU & integrated graphics do OK)...but I wouldn't expect it to even launch Fallout 4 (which I also have through Steam). Nor would I expect its ability to play Starcraft: Brood War or Halo:CE to provide me with any kind of confidence in its ability to play BF4.
  • elbert
    587530 said:
    45049 said:
    I wouldn't suggest either a 1070 nor the 1080 on 1080p. Probably should have tested with a more normal 1060 6GB at that resolution. Those willing to pay high prices for both the CPU and GPU should be running at 1440p. The game benchmarks are unnaturally skewed. The biggest thing to look at in the game benchmarks is how unutilized the Ryzen CPU's are. We will probably see the Ryzen R5 quads running the same FPS for a few hundered less then all but the old 8350 on April the 11th. Also it looks like AMD is gearing up a new socket with 8, 12, and 16 cores on an X390 motherboard. At the current pricing the 16 core could be lower priced than Intel's over priced 6900 8 core.
    http://www.techspot.com/news/68407-tackling-subject-gpu-bottlenecking-cpu-gaming-benchmarks-using.html 1080p may be the most common resolution that gamers use, but it's not the most demanding. By using a top-of-the-line GPU, the testers know that the GPU won't be holding the performance back (same reason why they've tested with 16GB of RAM for some time, even though you still really only need 8GB on 64-bit Windows for gaming). They may vary that somewhat depending on the game -- Techspot's Dark Souls III testing, for example, had to go up to 1440p for their CPU benchmarking (http://www.techspot.com/review/1162-dark-souls-3-benchmarks/page5.html), because the PC version was artificially frame-capped at 60FPS, & their GPU testing showed that you had to go down to a 4th- or 5th-tier GPU (like a GTX 780 or R9 380X) to start dropping below that cap with the i7-6700K. By using a powerful GPU at a resolution where it's going to idle along without any trouble, the testing allows them to definitively show how a given CPU (in this case, the Ryzen CPUs) perform against other chips.

    Yes but like in the old days everyone had to keep saying 640X480 was unrealistic. No one was getting information they could count on with such an unnatural benchmark. I think we are starting to fall back iinto that rut. The 1060 6GB is faster than the old 980 so its not that great of a bottleneck.
    From the Ryzen side tho the memory is bottlenecking as most are 2933 which effects the CCX greatly. AMD has show some 3466 RAM tests that really ups the game play. This is why the Ryzen 1500's will do so good in games. The CCX is taken out so the RAM speed doesn't hold back the game performance.
  • Papacheeks
    Well seeing as you guy's didn't use higher bandwidth memory outside of 2933 you wont see that ryzen is on par with a overclocked 5ghz i7 7700k in games.

    https://www.youtube.com/watch?v=RZS2XHcQdqA&t=392s

    This shows that the disparity goes away with 3200mhz ram or 3600.

    In some test's the ryzen beats the 7700k.
  • Cos70
    "If you do any content creation, and need more cores. Ryzen is the utter king right now, intel just cant compete. It can compete on performance, but not on price, you can get the same or better perf here on ryzen for much less cost, so its a no brainer choice to get an 8 core ryzen over 8 core intel."

    This is true if you are doing video encoding, sure. Although the story flips if you use Autocad or Photoshop, 2 widely used workstation applications.

    I agree that amd does outperform the 7700k in some markets, but those markets are select, and narrow down the target audience for these chips even more. Not ALL workstations will benefit from the extra cores. Again, it depends on your needs. If your applications are core friendly, you will see better performance. Unfortunately, it's a very small slice of the pie as of now, and by the time applications catch up, better 8 core chips will be on the market, including a amd-crushing intel cpu.

    Intel has been on minor revision tweks for 3 cycles now awaiting the release of the Ryzen chipset, so they could respond with another crushing blow. Amd is behind the game, while intel is holding onto it's gems awaiting a reason to release them.