Review: AMD Ryzen 7 2700X processor

^ I've got an AM4 A9500 dual core chip just in case, for future Ryzen builds.

 

I use that board with my 1600. Love it! clocked at 3850 mhz. Had no issues with this board! I hope it runs the ryzen2 well. A lot of folks what to know!

 

I really want to see a review on this board, Asrock x470 taichi vs a Asrock AB350 pro4. Comparing CPU boost, XFR, and memory speeds with the 2700x and or non "x". AsRock states the ab350 pro4 can handle the 105 watt 2700x. I'd love to see how well it can handle it. My guess is, just fine!

 

yea^....that would be cool review, there's quite a few boards on sale from last gen.
the savings could go toward Ram$

$64.99
MSI b350 tomahawk


$79.99
GIGABYTE ab350 GAMING 3

https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&IsNodeId=1&N=100007625 601292786

 

Oh yes you make a good case there i did not even consider the coolers at all!

 

That board seems to have some very meh VRM cooling and lacking in power phases. That being said, if you're planning on running it at stock and money is an issue, why not? Ask yourself a few things before you buy a motherboard and if the board has everything you need, it's a contender. I myself look at these few things before buying a motherboard, in no particular order...

• Power phases
• VRM cooling
• Sufficient SATA ports
• Overal design (position of PCI-E slots, M.2 placing, fan header count and placing, number of USB ports and such)
• Price
• Reputable brand (I refuse to buy anything from Gigabyte for example)
• Warranty
• Bios (overclocking features, ease of use, manufacturer that updates their bios regularly etc)

If you're about to drop a 2700x in that board and leave it at stock and you have good airflow inside your case (because those VRM heatsinks will need some airflow over them), why not?

 

That Tomahawk looks like a nice board. Built and two PC's with the Mortar which is the Intel version of this board. Great value right there.

 

Not sure if this article has been updated for the latest AGESA updates, but it seems there are some significant performance gains. Might be worth revisiting some of the tests.

 

Uhh XFR2 is nothing like an overclock or the "MCO" modes on many Intel Z370 boards. XFR2 is identical to XFR 1 in having a 100MHz max addition, but can be applied to any number of cores at once instead of just one.

Ryzen 2's "MCO"-esque auto-overclock feature is called "Precision Boost Overdrive" and like MCO allows it to ignore the chip's normal TDP limits (and also like MCO, is enabled by default on some boards), but unlike MCO doesn't outright apply the single-core boost clock to the all-core boost, but rather just removes the normal thermal & power boundaries based on the chip's TDP for Precision Boost 2 and instead let's PB2 crank things as high as your cooling capacity & board's power delivery can handle. And even if a reviewer was too dumb to make sure this wasn't enabled just like they would for MCO on Intel, most all boards didn't even get this option in their BIOS till the AGESA 1.0.0.2a update which came out after most everyone had done all of their testing to begin with.

XFR2 otoh, is just AMD's Turbo Boost 3.0 equivalent, and part of the chips normal operating specifications, the exact same way XFR1 was with OG Ryzen.

 

Most of those I found personally came from enabling the "Precision Boost Overdrive" option that it added (allows Precision Boost 2 to ignore TDP limits, and clock as high as thermal & power delivery headroom will allow) but Gigabyte's f23d BIOS that added it was full of major problems on my Aorus AX370-Gaming 5 mobo (identical to K7 aside from lacking BCLK generator). Most notably the use of any Corsair AIO on any fan header causing the board to read ridiculously false/incorrect tctl/tdie temperatures.

As in it thought the tdie temp was ≈95°C at idle, when the socket was 20-30°C, meaning even the slightest load would push it over the 105°C thermal shut-off point and crash my PC. This was a huge bummer as with Precision Boost working off the normal core temp like always and not tctl/tdie, I was able to test & taste the awesome performance boost from turning on PCO but was unable to actually use it in practice (i.e. from 1765cb with it off to like 1850cb with it on, both otherwise stock).

According to the GB forums, people on air or custom loops are having no issues with f23d, but I'm not sure about non-Corsair AIO's though. Regardless it leaves me stuck waiting on Gigabyte to either fix f23d and re-upload it or god fobid have to wait for the inevitable f24. Gigabyte makes fantastic high end hardware (the G-5 & G-K7 were the best value X370 flagships by a mile, and the new X470 G-7 is in a league all it's own at just $240 with that insane 10+2 VRM setup + true heatsink cooling), but their BIOS dev team is just not on the same level.

They also have a reeeeeaaaallllly bad habit of releasing gimped models disguised in flagship garb that average Joe wouldn't realize were gimped till it's too late, like the atrocious G-K5, which was pretty much the much cheaper K3 (which was basically just their B350 board's VRM setup, but with an X370 chipset for the added I/O & multi-GPU) dressed up in seriously misleading & unwarranted Aorus design/branding. Near everyone that got suckered into that board regretted it with the functionally near identical K3 being much cheaper and the minimally more expensive G-5 being on a completely different quality level entirely.
Gigabyte's just seemingly incapable of not overfilling their product stacks with boards that their own models in said stack make utterly pointless (though to be fair, X470 so far seems to be a MAJOR step in the right direction for them on this front, assuming there aren't more coming). ASUS is really bad with this as well. IMO, only ASRock seems to actually avoid overstuffing/overlapping their product stacks like that, but like GB have their own problems with BIOS development (though in a totally different way, with their bad habit instead being a notable lack of BIOS depth/control vs the other's equivalent models, rather than GB's tendency of public releases before they're actually ready/ consistently bug free, lol tbh I dunno which bugs me more).

 

Which is exactly how an actively competitive capitalist market should look. The fact that AMD's able to go toe-to-toe with Intel AT ALL considering the massive orders of magnitude sized gulf in R&D funding, along with Intel's (though rapidly shrinking) fab process advantage is pretty much a miracle in and of itself. AMD is managing to do nearly as much with exponentially less AND while using pure-play fabs. After a boring as molasses 2/3 of the past decade, thanks to Zen's far exceeding of nearly everyone's expectations, the game has finally well and truly changed, and I dunno about everyone else, but I'm loving every damn second of it!

 

I like your attitude Cooe^...your Cool

Did I spell that^ right? ...

 

Indeed you did. And just in case you're curious I've always pronounced it Koh-E, not Ceww-ee, double oo ≈ eww rule be damned.

Nickname actually originated on Christmas Day 1998 as I tried to come up with a unique file name to start up my shiny new TLOZ: Ocarina of Time cartridge that wasn't simply the same old boring "Link" yet again as I'd always done before (with a LttP & L'sA). So I spent like an hour before even starting the game I'd been waiting years for trying to pick something that visually looked & sounded super badass to my late 90's grade-school self hahaha. Hence the non-standard pronunciation. I thought Cooe looked really freaking cool/unique, but "Ceww-ee" sounded lame. But the much harder, 2-sharp syallable "Koh-E" pronunciation otoh felt like putting on a glove the first time I said it out loud vs the rest of my crap ideas till then, but then "Cohe" looked even lamer than "Ceww-ee" sounded. So still being a kid where easiest = bestest solution, I just combined the the spelling of the former with the pronunciation of the later and voila! it's stuck for all these 20 years since hahaha. (During which I've never seen it used anywhere else either, so mission "make a unique screen-name" so far, so accomplished for 20 years and counting at least. Not bad for something a 2nd grader pulled out his ass xD).

Still have to correct some people here & there in chat's & game's and whatnot, but much to my surprise, at least as many people naturally pronounce it how I do, as don't. Haha it's prolly bc "Ceww-ee" happens to sound as much like the noise a derping chicken makes to them as it does to me hahaha.

 

How there is no i5 7600K which was very popular CPU and its not yet date. it's big mistake imo....

 

you get a lot of numbers here...http://www.guru3d.com/articles_pages/core_i5_7600k_processor_review_desktop_kaby_lake,14.html

 

Looks like Anandtech completed their analysis of their original Pinnacle Ridge review. Some surprising results for Intel CPUs regarding HPET.

https://www.anandtech.com/show/12678/a-timely-discovery-examining-amd-2nd-gen-ryzen-results

 

They delivered reasonably good read. So, we have new moto: "Noob, don't force HPET."

 

Thanks, an interesting read, I didn't know about HPET timers - made me check my system to make sure HPET wasn't forced on in my OS, luckily & as expected HPET is not forced on in my OS:

"bcdedit /enum" in the command line & then checking to see that "useplatformclock Yes" is not listed amoung the output of lines - courtesty of the following thread:
https://www.reddit.com/r/Planetside...d_off_hpet_in_windows_10_now_i_have_an_extra/

 

You miss my point, there is little numbers on yet quite popular CPU, at least numers from review you pointed should be put in to graphs of this review ...

 

ANy word on the review for the 2700 (65w chips)?