• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Zen 2 (Ryzen 3000) - *** NO COMPETITOR HINTING ***

DragonQ said:
Significant improvement over the R7 2700X in every game they tested, but also significantly behind the i8-8700K and i9-9900K. These are all at 720p though so the gaps will be very much smaller at realistic resolutions.
Click to expand...

Nothing wrong with 720P testing, it's when you use low detail's is when it's not a real test.

A lot of those detail settings use more CPU thread's, you turn all that off a low core count CPU is just as good as a high core count CPU, yet while a lot of people might run a 2080TI at 1080P they wouldn't turn off all the details, unless they are playing a competitive game match.
 
RavenXXX2 said:
We all know how well Zen scales with faster tighter ram, I bet with tuned sub timings and faster ram that gap will close down significantly.
Click to expand...
This time around not as much gains to be made with fast and tight RAM as previously...... But still something for sure...
 
So far it seems to me that if you are an out and out gamer, then the 9900 is still among the best of choices, but if you want a good all rounder for gaming, video and photo editing, audio, rendering etc, then Ryzen is the better choice, particularly the 3900X.
 
8 Pack said:
Standard or Auto detect settings.... Maximum on Heaven, Valley etc.....
Click to expand...

With the amount of GPU grunt you have use Ultra Settings at 1080P or even 720P, you will still bottleneck the GPU but you will also distinguish between low core count CPU's and High.
 
I would take 1080p, 1440p & 4k, medium, high & ultra settings where applicable (as that is generally where most people seem to game)

Use 3 different cards from each vendor so Vega 56/Vega 64/Vega VII & 980, 1080, 2080 and just see what they do.

Then do both out box RAM as they have and profiles to suit so 3600 C16 or something realistic.

That gives what real world performance is for gamers at their resolution and graphic settings.

The 720p tests are waste of time because we just don't game at that and I don't care about theoretical, I care about what it actually does at what I game at.

If the CPU is then showing that the FPS are within 5% then I am not worried.
 
Kelt said:
Like I said earlier, 720P testing means absolutely nothing to me, utterly meaningless.
Click to expand...
Yes, 720p CPU testing is basically an artificial benchmark, like Sysmark or Geekbench. Personally I want to see 1440p results but obviously a lot of folks will care more about 1080p.
 
Curlyriff said:
I would take 1080p, 1440p & 4k, medium, high & ultra settings where applicable (as that is generally where most people seem to game)

Use 3 different cards from each vendor so Vega 56/Vega 64/Vega VII & 980, 1080, 2080 and just see what they do.

Then do both out box RAM as they have and profiles to suit so 3600 C16 or something realistic.

That gives what real world performance is for gamers at their resolution and graphic settings.

The 720p tests are waste of time because we just don't game at that and I don't care about theoretical, I care about what it actually does at what I game at.

If the CPU is then showing that the FPS are within 5% then I am not worried.
Click to expand...

Agree.
 
OO I see the AMD FineWine brigade is here. Worry not lads once you slam 3600cl14 You will see 1:1 performance with Intel :D
 
humbug said:
With the amount of GPU grunt you have use Ultra Settings at 1080P or even 720P, you will still bottleneck the GPU but you will also distinguish between low core count CPU's and High.
Click to expand...
For some efficiency testing I matched up number of cores, mem speed and clock frequency of the two CPUs being tested.... Just to get an overall IPC picture....
 
boom

uY7bUtz.png
 
DragonQ said:
Yes, 720p CPU testing is basically an artificial benchmark, like Sysmark or Geekbench. Personally I want to see 1440p results but obviously a lot of folks will care more about 1080p.
Click to expand...

Nothing wrong with 1080P testing, according to Steam it's what most of the world is gaming at, but not 720P.

I used to game at 640x480 but that was almost thirty years ago!
 
DragonQ said:
Significant improvement over the R7 2700X in every game they tested, but also significantly behind the i8-8700K and i9-9900K. These are all at 720p though so the gaps will be very much smaller at realistic resolutions. Also interesting to note that even at 720p, the 12-core is barely beating the R7 3700X in gaming. Definitely reinforces my gut feeling to go for the R7 3700X.

The chips are killing it in productivity.
Click to expand...
Maybe MAD should run benchmarks on 8k then ?? :D
 
Looking at the archived source, out of the 6 games they test, the R7 3700X has better minimums (yes, at 720p) than the i9-9900K in 4, and worse in 2. Guess which 2 results were first posted in this thread... :D

FPS Minimums:
RotTR 9900K 8% faster
FC5 9900K 16% faster
Wolfenstein 2 9900K 4% slower
AC: O 9900K 3% slower
Cities Skylines 9900K 4% slower
KC: Deliverance 9900K 9% slower
 
Zeed said:
OO I see the AMD FineWine brigade is here. Worry not lads once you slam 3600cl14 You will see 1:1 performance with Intel :D
Click to expand...

Yeah it's probably not a good idea to use what was just posted to gleefully troll who you might term "The AMD Brigade" it's far too early for that much confidence. it could easily backfire in a couple of days.
 
8 Pack said:
For some efficiency testing I matched up number of cores, mem speed and clock frequency of the two CPUs being tested.... Just to get an overall IPC picture....
Click to expand...

That's pretty good, looking forward to seeing your benchmarks :)
 
You must log in or register to reply here.
Back
Top Bottom