• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD launches Zen+ 12nm Ryzen and X470 motherboards

Page 29 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Udgnim said:
only looked through Anandtech gaming charts but the impression is that this is much better than expected
Click to expand...
PingSpike said:
https://www.anandtech.com/show/12625/amd-second-generation-ryzen-7-2700x-2700-ryzen-5-2600x-2600

First thoughts: I'm pleasantly surprised with the gaming performance
Click to expand...

Wasn't what I expected but it does look like AMD upped the game.

Udgnim said:
only looked through Anandtech gaming charts but the impression is that this is much better than expected
Click to expand...

Better than I expected also..
 
Shmee said:
Very interesting if they are actually that good, but hard to believe. coercitiv, how do you figure that? I could not find where AT specified it's timings / speed.
Click to expand...

It is at the top of its Test Setup page. Although the chart is missing.
 
Eh, the performance gaps certainly level out @ 4K...

3b03db22-7222-448c-a4f5-a869cbb55422.png


This makes the decision a lot easier in my case...
 
Timorous said:
It is at the top of its Test Setup page. Although the chart is missing.
Click to expand...
I saw they specified JEDEC settings only for a basic test, but how do we know what that speed/timing that is for their RAM? Unless I am missing something.

I would like to see more online play of intense FPS @ 1440p 144Hz. BF1, BF4, SWBF2 etc.
 
Shmee said:
Very interesting if they are actually that good, but hard to believe. coercitiv, how do you figure that? I could not find where AT specified it's timings / speed.
Click to expand...
Timing/speed info is not posted (although I think a table will be added momentarily) but description above is quite clear on the subject:
As per our processor testing policy, we take a premium category motherboard suitable for the socket, and equip the system with a suitable amount of memory running at the manufacturer's maximum supported frequency. This is also typically run at JEDEC subtimings where possible.
Click to expand...
 
I don't think spectre had anything to do with gaming performance. Although I did few tests myself, I saw little performance drop and that on UWP games.

I'm very happy with the new Ryzens. Really good job AMD, kudos.

Hardware.fr is also up. Still shows the coffeelakes as faster in gaming, but it does not matter really. From where I stand Intel and AMD are now equal.

https://www.hardware.fr/articles/974-19/indices-performance.html
 
coercitiv said:
Timing/speed info is not posted (although I think a table will be added momentarily) but description above is quite clear on the subject:
Click to expand...
It seems very unclear to me. I read over that paragraph again, and it seems that they just let it set by auto most likely, as most users would? In which case, it will default to the what the motherboard will default to, which will depend on motherboard, BIOS, and RAM used. This tells us their methodology, but IMO it tells us very little about what speed the memory is set to and what timings. I don't think that how much memory was even mentioned.
 
krumme said:
As The stilt explains in other thread the memory latency difference is more pronounced at lower mem freq. At 3200 its partly nill.
Let me guess. Gamers nexus used 3200c14 for both the 1800x and 2700x?
Jedec and timings explains a lot.
Click to expand...

PCPer did the same things and their results are different as well.

Civ 6 at 4k over at PCPer is ~76fps for the 8700k and ~ 79fps for the 2700x with a 1080Ti, Anand is getting ~ 76 and ~ 86 fps respectively with just a 1080?
 
krumme said:
As The stilt explains in other thread the memory latency difference is more pronounced at lower mem freq. At 3200 its partly nill.
Let me guess. Gamers nexus used 3200c14 for both the 1800x and 2700x?
Jedec and timings explains a lot.
Click to expand...

Steve ran 16-18-18-36 3600
 
Looks like certain review sites are "forgetting" to patch their systems for meltdown and spectre. While more reputable sites like Anandtech are running with all the patches and the results are clear: intel has lost its gaming lead.



I guess that puts the last nail in the coffin of the 8700k. No point in spending $340 on 6 core 6 thread CPU if it performs worse even in games than the cheaper 8 core 8 thread CPU.
 
Shmee said:
It seems very unclear to me. I read over that paragraph again, and it seems that they just let it set by auto most likely, as most users would? In which case, it will default to the what the motherboard will default to, which will depend on motherboard, BIOS, and RAM used. This tells us their methodology, but IMO it tells us very little about what speed the memory is set to and what timings. I don't think that how much memory was even mentioned.
Click to expand...
He probably set the timings manually, but in such a manner as to abide to JEDEC standards and obtain similar latencies for both 2666 and 2933 (ex: 2666 @ 17-17-17 vs. 2933 @ 19-19-19). That having been said, we need the actual numbers to better understand what happened, let's hope they get published.
 
I would just hold off on proclaiming anandtech results as the only ones to believe, there is a massive difference between them and every other review I have read...your telling me all the other professional review sites never thought about updating the firmware before days of painstaking testing?

Something is very off with anandtech numbers, besides using a gtx 1080 for 1080p ultra benchmarks? ...come on AT, your better than that.
 
Termie said:
For anyone interested in IPC numbers, my Tech Buyer's Guru review provides a look at IPC increases by clocking Zen, Zen+, Broadwell and X-Lake at 3.8GHz.

And I hate to say this, but AT clearly goofed up its gaming benchmarks. Results like those should have led to a publishing hold, but NDA deadlines called.
Click to expand...


I'm guessing they goofed and the numbers are swapped in the graphs.
 
FIVR said:
Looks like certain review sites are "forgetting" to patch their systems for meltdown and spectre. While more reputable sites like Anandtech are running with all the patches and the results are clear: intel has lost its gaming lead.



I guess that puts the last nail in the coffin of the 8700k. No point in spending $340 on 6 core 6 thread CPU if it performs worse even in games than the cheaper 8 core 8 thread CPU.
Click to expand...

GPZ fixes shouldn't make any difference in gaming.
Meanwhile, in e.g. compiling tasks the penalty is BRUTAL.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top