Last month, we shared the specifications and the first official gaming benchmarks from AMD for its upcoming 2nd generation Ryzen CPUs. AMD’s new Ryzen CPUs are currently available for pre-order, will launch on April 19th, and even though the first reviews will go live on April 17th, it appears that the first third-party gaming benchmarks have surfaced.
El Chapuzas Informatico has benchmarked the AMD Ryzen 7 2700X and used the latest Promotory X470 motherboard alongside an MSI GeForce GTX 1070 Gaming Z and 3200Mhz memory modules. The website tested five games: Battlefield 1, Doom, Resident Evil 7, Rise of the Tomb Raider and Total War Warhammer 2.
What’s really interesting here is that at 1080p, the AMD Ryzen 7 2700X was able to come really close to the Intel i7 8700K. While AMD claimed that the Ryzen 7 2700X CPU would be around 7.7% slower in a variety of games than the i7 8700K, El Chapuzas Informatico claims that AMD’s new CPU is faster in Battlefield 1, Total War Warhammer 2 and Resident Evil 7 than the i7 8700K. On the other hand, Rise of the Tomb Raider appears to be GPU-limited even at 1080p and DOOM appears to be running better on the i7 8700K by 2%.
These are definitely interesting results so we are really looking forward to more third-party gaming benchmarks from other publications!
John is the founder and Editor in Chief at DSOGaming. He is a PC gaming fan and highly supports the modding and indie communities. Before creating DSOGaming, John worked on numerous gaming websites. While he is a die-hard PC gamer, his gaming roots can be found on consoles. John loved – and still does – the 16-bit consoles, and considers SNES to be one of the best consoles. Still, the PC platform won him over consoles. That was mainly due to 3DFX and its iconic dedicated 3D accelerator graphics card, Voodoo 2. John has also written a higher degree thesis on the “The Evolution of PC graphics cards.”
Contact: Email
Wow. Those performance gaps between AMD and Intel have really marginalized, especially when you factor in multitasking.
Wow.
Nope. Just bad test settings. Ultra settings and 1080p in everything.
I think you’re missing the point.
Testing at 1080p is the best way to eliminate GPU bottlenecks from showing up in the test, skewing the results. You test at 1080p to make sure the results are only CPU dependent. You don’t want the GPU to have a impact on the results. It’s standard procedure. Go higher and you’re just dealing with GPU variances and such.
Good to know my 6700k will still see continued use.
I’d love to see them benchmark more CPU intensive games like Planet Coaster and the upcoming Jurassic Park game.
I’d be interested in seeing better ram considering the Intel parts were running 3600mhz ram vs 3200mhz on AMD.
https://uploads.disquscdn.com/images/25affa7a84cf402a56fb3405976740d272c74191d90d93daa62330cb71fb8274.png https://uploads.disquscdn.com/images/660a22e3ea9bfbf80324cf8c4f2f044c8b736a9e8d91e77c4b8dca01d3f1d772.png https://uploads.disquscdn.com/images/0625f6881d83984751daebcfec7922163c67dde3415cf234f9791246a9217ce8.png https://uploads.disquscdn.com/images/7563b95f8f8af1f59095102df8ad4d8bafe6c584ead369b92bad3dafa57df11a.png
In certain games it absolutely matters.
This kinda makes no sense . We have no idea what GHz the cpus runing . But how it looks i would say the AMD is OCd and Intel not .
Stock vs stock , 3200 mhz memory on the zen vs 3600 on the intel
Well then this test is useless if its stock .
+ where is the 7700k ? Its a rly good cpu .
The test is not useless , it is a representative of the cpu perf vs the 8700k and the old zen.. the 2700x in this review mind you has beaten the 8700k in 7 out 10 games while having slower memory.. and zen loooioves fast memory
Yes . That is true . BUT ! But but buuuut there is a big butt . …
Why would you test it at stock clock . We both know that intel have way more horsepower if its OCd . And who would buy a K cpu and dont overclock ? This test is just there so AMD can look good. ( dont get me wrong i like AMD more then intel ) but i just want the truh and until we have real time in game tests , this images above are useless .
And again where is the 7700k ?
Dunno…
It is good to see out of the box perf , and zen is crippled by slow ram vs the intel
“Why would you test it at stock clock ”
Why wouldn’t you test it at stock speeds? Mate not everyone who buys a PC overclocks.
There’re people who are just happy or prefer to leave their system at stock speeds.
Stock speeds equal stock voltages and potentially bad ones at that. I personally learned to OC just by making sure my stock settings were in decent parameters in regards to my voltages after reading now stock settings can let them voltages just fly all over.
In doing so would truly degrade your CPU if your standard voltages were remotely above the higher recommended values for that particular piece of hardware.
Stock isn’t always the best in those BIOS settings and that has been proven many times.
A i7-8700 ( not the 8700k ) have a base clock 3.2 ghz . And it auto boost it self up to 4.6 GHz without anyone OC it …….
So we kinda need specific numbers if someone is dooing a benchmark .
um dude both are 1700x… look at the chart more closely
Test on stock CPUs are the most relevant because they represent what people will buy in the shop. Tests with OCd CPUs are just good extensions because the most of people will not overclocked their CPUs. Such tests are just for enthusiasts. All practical (not specialiazed) tests should be based on HW with default settings and behaviour.
“Test on stock CPUs are the most relevant” oh rly ? If its stock then why they dont test with the not K 8700 ? You know the CPU what cant be OC . . . . Cuz the 8700 k and 8700 are the same just the 8700 k can be OCd .
“most of people will not overclocked their CPUs” yeah. + “tests should be based on HW with default settings and behaviour”
That is the biggest lie waht you call tell . Its like saying ppl buy ferrari to go slow . GENIUS . ppl OC everything Ram Gpu Cpu . . . . . Only precial ppl dont oc a K cpu . . . . . And if you will say well you need good cooling for it then i say if you have a buget for a K cpu and MB then you have money for cooling as well
You are living in disillusion. The most people even don’t reading HW or SW websites. They just buy CPU and leave it be on its default settings. This matters only for some enthusiasts. When I bought i7 3770K in 2012, it run 4 years on default clocks because I didn’t have to OCd it. I did it only because I bought GTX 1080 for 4K resolution in the end of 2016. Not everybody is like you. You are big minority here. 🙂
As I said, the test on stock settings are most relevant because that is the case how the most people will use it. You will learn that lesson when you grow up.
I live in a disillusion ? You are 100% wrong . I dont live in a disillusion . Thats why i ask for numbers and do t just accept the benchmark . You live in a disillusion .
I am minority ? You say that while playing on 4k ? Most ppl still use 1080p . You are the minority . Again …you are wrong .
Sotck ? A stock I7-8700 have a auto oc up to 4.7ghz . So yeah
Oh you are a big boy now cuz you OCd you CPU.
“Oh you are a big boy now cuz you OCd you CPU”
You should learn to understand of written text. I wrote I did not OCd my CPU because I didn’t need it and the most of people just don’t OCd their CPUs at all. That’s how it is and if you belong to the group which do it, then you are the minority.
“A stock I7-8700 have a auto oc up to 4.7ghz”
Yes and this “auto OC” is not turned off during the tests. Just exactly how the most people will use it. And that is why each test of CPU should use it in this way too. I am not against testing of OCd CPUs, just it should not be the basic way how test them. They should be tested in stock settings.
I am the minority ? A Person say this who OC his cpu + play on 4k belong to the minority more then me .
Why would they turn off a out of the box settings ? You make no sense . . . . Moat ppl dont turn off a setting in a bios . . . Not mst ppl dont do this NO ONE DOSE . . . You make no sense . . . A normal ppl just buy a box in a shop and let the cpu do its best and let it boost it self up . . . .
Default turbo is ON On stock . . . . .
Man you make no sense
“Why would they turn off a out of the box settings ?”
“A normal ppl just buy a box in a shop and let the cpu do its best and let it boost it self up”
Exactly the same what I was talking all the time. You obviously have problems to read and/or understand written text. All people here are telling you, that it is the best to test CPU on stock settings (it means with turbo too). You were talking about testing of OCd CPUs which mean CPUs overclocked beyond default settings (so higher default and turbo frequencies). Nobody (including me) was telling you about turning off turbo. Read the whole discussion again.
“I am the minority ? A Person say this who OC his cpu + play on 4k belong to the minority more then me .”
Minority in area of overclocking CPU. Minority of people is overclocking their CPUs. They just use them with default settings (default and turbo frequencies). Turbo is overclocking by the box. You obviously misunderstood the difference between turbo and out of the box overclocking.
Find out for yourself what “stock” means. Because there is the whole problem. You obviously thought that it stock CPU is without turbo. And that is wrong. It means CPU with default settings as you buy it (so with turbo too).
If you actually were bothered to click on the link they were testing at stock speeds..
If its stock why they dont put the stock clocks into the benchmark ?
And what STOCK means ? A non K I7-8700 have stock boost up to 4.6 GHz and its a out of the box auto boost. . . . . SO
So this until we got some numbers this benchmark is a pile of horse S**t
Wow, did you look at the other scores? Every other game tested the 1600 was down 20-40%.
So the 6700k is faster than the 8700k? I don’t recall any of the reviews of the 8700k showing that. Even the 8400 beat the 8700k in a game.
There is also a typo on the ROTR slide, the 6700k is listed twice with two different scores. Looks a little sketchy to me, let’s wait till the real reviews come out.
This looks good Amd keep up the work! Nice to have another option in the CPU market.
8400 is one heck of a value though
These tests are all incomprehensive because of improper methodology. GTX 1070 bottlenecks the CPUs, so their full potential cannot be explored. If you use GTX 1080 or GTX 1080 Ti the results will be completely different.
Even then the graphical settings would still have to be lowered a lot.
Those games are just unoptimized. Notice how a majority of games have literally zero difference.
So, I’ll change my statement, I wasn’t completely correct, only partially correct, I didn’t consider trash developers. Nothing but unoptimized games makes any difference to FPS on Intel CPU’s with higher or lower speed RAM. Also Hardware Unboxed is trash. But if you wanna use his word as proof, you also just kinda proved yourself wrong trying to prove me wrong, that’s funny :).
The Reaper95, you’re just wrong on this point. You’re acting childish because people are proving you wrong.
I posted several pictures of testing above either on a 7700k or a 6700k showing that ram speed matters. It’s more than just some instances.
What I believe is the “old school” thought and what reviewers were saying is the amount of money needed to spend on High Frequency Ram was not worth the performance difference.
Then if you weren’t overclocking the CPU the difference wasn’t noticed as prevalent. This was more true with intel cpu’s at 4 Ghz or lower.
However, things have changed in the market and ram is all around just expensive. Newer CPU models come at high clocked default 4+ Ghz and it doesn’t for some reason look like high frequency ram is 300% more expensive than lower frequency ram.
As hardware unboxed showed, as with other reviewers, a bigger gain between several ram frequencies than other frequency differences.
My nephew needed a fast computer but was on a budget so I got the 3000 mhz ram as higher yielded less return for value. I then OC’d his computer to 4.2 Ghz to get more out of that ram speed. Even since that build 2 years ago things have changed.
Doesn’t mean there is no return in scenarios with ram beyond 3000 mhz for skylake. Just depends on budget.
margin of error – they’re all very close is the takeaway
they’re all very close because they bottlenecked the GPU.
Garbage tests, to benchmark CPU performance you don’t use max settings and 1080p. 1080p might be okay at this point, but the highest settings a game offers certainly are not. These framerates are so close because regardless of the CPU the games are bringing the GPU close to its limit.