AMD 2nd Gen Ryzen 2 2700X Zen+ CPU Review @ [H]

Nice review! Extra points for "Frag Harder Disco Lights". I also actually LIKED that you just "threw" it on the motherboard and did a "quick and dirty" overclock, as, I think, that's what most people will be doing.

My opinion? This is AMD going toe-to-toe with Intel for the first time in GENERATIONS of processors! AMD has always been looked at as being somewhat inferior to Intel, and this sounds like it's a real game changer! I have been leaning AMD for a while now, given Intel's attitude and their handling of the Meltdown/Spectre issue. This chip seals the deal. I will be either using a Ryzen 2 or Threadripper 2 in my next system! (soon, sooooooon!)

Now I just have to wait to see what AMD has up their sleeves for graphics cards this fall! I am really looking forward to going back to "Team Red"!
 
  • Like
Reactions: N4CR
like this
Kind of cool that the chip can manage itself so well that a all core overclock that is less then the 4.3 it can boost itself to on less cores will show better results then the all core overclock. Kind of bummer that it might kill the need to overclock anymore for us enthusiasts. AMD must really know the limits of what their chips can do.
 
Awesome! Thanks for the review.

Couple of points. On the conclusions page, you guys state that the 1800x is faster in Civ 6, however the graph shows the 2700x as being faster :)

Also on the first page . discussing the new x470 board

And of course there is a news X470 chipset

all in all, it looks like the 2700x out of the box is a great CPU.
 
Last edited:
Great review Kyle! Intel should really take a page out of AMD and learn how to do a proper upgrade release.
 
Awesome. I dig the 8700k's sacrifice.

Some wonky results though, as mentioned in the conclusion. I'm especially stumped by that Civ 6 chart on the gaming benchies page.. that's a pretty CPU intensive title, and I'm fairly sure it's multi-core aware, so it should have been flying. Of course, most core tests show a pretty huge scaling dropoff past two cores anyways, so maybe it's not all that weird.
 
Glad I pro-ordered.
Now to figure out what memory and motherboard for my 2700X.
 
Great review Kyle! Intel should really take a page out of AMD and learn how to do a proper upgrade release.
Proper upgrade release? The bump in performance is generally in line with what we got with Intel over the last 8 years. Granted it is better then a 8700k but don't try and make it out to be some godly upgrade over lasts years release.
 
Last edited:
Proper upgrade release? The bump in performance is generally in line with what we got with Intel over the last 8 years.

You are correct after I think about it, just feel jaded how Intel handle the Lake series upgrades, seems rush in comparison to AMD handling of Ryzen to Ryzen 2, though Ryzen initial release was plague by memory issues.
 
Great review! Glad to see AMD back in the game, so to speak.

I'm curious about the dead CPU; was it a test bench CPU that only ran benchmarks @ 5GHz @ some voltage, or was it living in a desktop which it was borrowed from as needed? I haven't seen a confirmed kill on a newer intel CPU in awhile, so I'm curious what you think may have nuked it.
 
-1 Internetz for lack of [H]ard. I fought my 1700X for weeks to get 4.0 out of it. Because.

When the manufacturer has access to the tiny bits to out [H]ard you, the [H]ardest thing you can do is let it go. I though OC'ing was all about squeezing the most performance out, by any reasonable means? Which means [H]ard cooling and letting Precision Boost work at a granularity we don't have access to. (Yet or perhaps ever)
 
Color me suitably impressed, though as some have said, the real news from this post is that an Intel Core i7 8700K just went tits up doing standard (and reasonable) OC shenanigans. You said you bought it used on eBay, I think? Perhaps the former owner pushed it into "Forbidden Voltage" territory? Maybe even ESD damage? Sadly, I guess we'll never know...

edit: He got it used from Microcenter, my bad... Still, my question stands.
 
Great review! Glad to see AMD back in the game, so to speak.

I'm curious about the dead CPU; was it a test bench CPU that only ran benchmarks @ 5GHz @ some voltage, or was it living in a desktop which it was borrowed from as needed? I haven't seen a confirmed kill on a newer intel CPU in awhile, so I'm curious what you think may have nuked it.
In the review he said he got it used from microcenter last week.. my guess is the person before tried to overclock it like an amateur and failed so they exchanged it but it worked just enough that microcenter thought it was fine to resell.

Edit by Kyle: THE *700K CPU I PURCHASED WAS BRAND NEW SEALED!
 
Last edited by a moderator:
Yeah, it was open box. Makes sense sirmonkey.

Edit by Kyle: THE *700K CPU I PURCHASED WAS BRAND NEW SEALED!
 
Last edited by a moderator:
Is there any chance that there were problems with the numbers generated by a damaged 8700k, or am I misunderstanding something here? I'm not a "this brand or that brand" fanboy, but it certainly would seem irresponsible to publish a review based with numbers from a CPU you knew to have faults at a given frequency.
 
Is there any chance that there were problems with the numbers generated by a damaged 8700k, or am I misunderstanding something here? I'm not a "this brand or that brand" fanboy, but it certainly would seem irresponsible to publish a review based with numbers from a CPU you knew to have faults at a given frequency.

What are you getting at? The benches were in line until it gave up the ghost. Results TERMINATED after blue screens..
 
In the review he said he got it used from microcenter last week

In the article, he said "but I bought the 8700K used in this review at Microcenter last week for $300." The way I read that, it means "I bought the 8700k <i>that I</i> used in this review...". Unless he specifically said elsewhere it wasn't new and I missed that, it sounds like a new one. $300 is their online price for a new 8700k.

Edit by Kyle: THE *700K CPU I PURCHASED WAS BRAND NEW SEALED!
 
Last edited by a moderator:
would there much of a performance difference between x370 and x470 ? i got a really good x370 board.
 
Glad that AMD really looks like it is pushing hard on CPU front, I wish that next year on Ryzen 2 will go all out to hurt bad intel to the throat, making cluster of 8 CPU Vs 4 and offer the first 12 core on mainstream level, connecting 2 dies and disabling a pair for cluster , and offer a 24 core on top threadripper for $1k , and 14 core for $500 on full 64 PCIe lanes .
Ok can accept also cluster of 6 core and go to offer a 10 core for mainstream , and 20core for $1k on threadripper. and moving the wall at 4.6-4.8GHz.
If intel cannot push out the 10nm in 2018, can be in trouble in 2019 play from low performing fab tech for the first time in a long time.
For sure the competitions is finally here to stay, and finally, we like user can get more for the same money year after year, We all were very tired to be milky for years from intel with minimal performance upgrade, higher prices, no discount, no prices lowering with the time like used to be before.
Personally I'm looking to await 2H 2020 to buy the next decade high-performance system, wishing memory prices will crash, new generation of HDDs of 20TBs are out and the PCIe 4/5.0 is finally out.
Hardocp are always the best review to read for sure no bias, telling you the hard truth, no marketing. and I read Hardocp even before the times of Pentium 1.13Ghz Fiasco that Kyle unleashed. Way to go!
 
yeah anandtech's numbers seem to be far different from everyone elses.. only thing i can think of is that it's related to their cooler choice and they're not seeing the same boost clocks on their intel processor as other reviewers are using custom loops.

The difference appears to be Anandtech's initial testing going with stock setting, recommended memory and air cooling, consistent with what the majority of buyers will actually do which is a more honest initial review IMO.

From Anandtech's setup page - "It is noted that some users are not keen on this policy, stating that sometimes the maximum supported frequency is quite low, or faster memory is available at a similar price, or that the JEDEC speeds can be prohibitive for performance. While these comments make sense, ultimately very few users apply memory profiles (either XMP or other) as they require interaction with the BIOS, and most users will fall back on JEDEC supported speeds - this includes home users as well as industry who might want to shave off a cent or two from the cost or stay within the margins set by the manufacturer. Where possible, we will extend out testing to include faster memory modules either at the same time as the review or a later date."

The takeaway is for the majority of buyers the 2700 and 2700X provides far better out of box performance across nearly all workloads than the best Intel has to offer.
 
Last edited:
What are you getting at? The benches were in line until it gave up the ghost. Results TERMINATED after blue screens..

What am I getting at? Were my questions not clear? I suppose I could have quoted Kyle so he knew it was a question meant for him.

If the CPU was damaged at 5ghz, why count on the results generated by it at stock. Do you understand now?
 
Possibly, things were already close enough that a couple percent down on Intel and the Zen+ improvements could easily flip the tables

All I know is the whole herd of Intel shills high frequency gamers will definitely shutup and make the switch ;)

Their absence is one the most notable and welcome differences compared to Ryzen's launch last year.
 
would there much of a performance difference between x370 and x470 ? i got a really good x370 board.
Not worth upgrading just the CPU or MB. Upgrading from 1700 to a 2700x is barely a side grade. It's great to get if building a new system tho.
 
Nice review Kyle! I love the peanut butter TIM jab.

My FX 8320 encoder/plex box needs an upgrade I think. Someone feel free to buy a new 2700x and sell me their gen 1 Ryzen 1700/1800!
 
When the manufacturer has access to the tiny bits to out [H]ard you, the [H]ardest thing you can do is let it go. I though OC'ing was all about squeezing the most performance out, by any reasonable means? Which means [H]ard cooling and letting Precision Boost work at a granularity we don't have access to. (Yet or perhaps ever)

We can still out [H]ard them. Guru3d got 4.4 out of theirs.
 
Steve Burke observed lower voltages per clock than on first gen , sometimes substantial and for me that is major, that is ultimately going to be the 12nm refined node working better and leaves me encouraged for Matisse, it will be nice to see what H get in their review on power side of things.
 
Those synthetic benches were weird, thermal problems?

Still, staying over 4ghz on 4+ threads is temping me hard.
 
If I were replacing my 7820x by no means would I gain zilch but... if I were replacing an 8600k or 7000 series Intel it's a no brainer and since I have an 8600k as my other machine it's rather tempting to get an 2700x. However... it CAN'T replace my venerable 7820x.
 
Sounds pretty impressive. It looks like an impressive competitor to the 8700k.

Another reviewer, who I won't name but you can probably guess, decided to run their review with XFR 2 and Precision Boost Overdrive turned off, and thus came up with very different results. (Of course, I quit reading their site a while back, and only took a look because a friend mentioned this issue.) Funny how that works.
 
I was hoping for more. The 8700K is $300 at Microcenter, so if my new build works out in the next few months it looks like I'll be sticking with that seeing as it's solely for gaming. I don't mind settling for 95% of the performance if it's cheaper, but not when it's the same price or more.
 
I was hoping for more. The 8700K is $300 at Microcenter, so if my new build works out in the next few months it looks like I'll be sticking with that seeing as it's solely for gaming. I don't mind settling for 95% of the performance if it's cheaper, but not when it's the same price or more.

Man I miss living near a MC. For people without access to one, it really seems like the 2700x might be the way to go.
 
The takeaway is for the majority of buyers the 2700 and 2700X provides far better out of box performance across nearly all workloads than the best Intel has to offer.

Yeah it's kind of an interesting question too, when you consider that Kyle is not taking it lightly to say what he did in his conclusion:
from Conclusion page said:
I actually asked myself, should I even consider overclocking this CPU if it was my daily driver? Unless you are crunching core and thread intensive workloads all day long, I would have to say no. I feel dirty saying that, but I feel as though it is true.

I've been always insistent on the DIY experience, and whatever I upgrade to next, it's likely I still do it manually, as I have been since 1999. But if Precision Boost 2 is as good as advertised - and Ryzen 2000-series is a testament to that, it certainly helps the more layman get the same performance that many of us at [H] have been getting for nearly two decades through manual overclocking. I wonder if it's too early to start getting nostalgic about it, to see effective boosting mechanisms now on *both* CPU and GPU. What is left to manually tweak and change?

From AMD website said:
https://community.amd.com/community/gaming/blog/2017/11/27/asdasd
Precision Boost 2
Precision Boost 2 for the Ryzen Processor with Radeon Graphics is also based on the reliability triangle, but with a major difference: Precision Boost 2 does not impose a lower clock speed limit if more than two CPU cores are being used. Precision Boost 2 only assesses whether the processor is within specifications, and continues to boost—on any number of cores—until reaching the maximum clockspeed printed on the box, or bumping into a boundary on the reliability triangle (whichever comes first).


This new, more flexible boost can have a big impact on applications that spawn many lightweight processing threads. Even collectively, light threads often don’t demand much of the processor, which means they don’t require much energy or generate much heat to process. Where an AMD Ryzen desktop processor with Precision Boost would move into the lower frequency all-core boost for such workloads, Precision Boost 2 can allow a Ryzen Mobile Processor with Radeon Graphics to keep cranking the clockspeeds as high as possible on however many cores the workload requires.
 
Back
Top