Chinese site shows extensive benchmarks results of AMD Ryzen

[kladzen]

Source: http://www.overclock.net/t/1624243/cpu-z-ryzen-1700x-single-multi-thread-benches-also-1700x-vs-6800k-benches & http://bbs.pceva.com.cn/thread-137649-1-1.html

 

Tested with AMD Ryzen 1700x - NOT with the 1800X... so maybe even better results... These results are with STOCK speed's and not overclocked
1 Thread = 2327

http://valid.x86.fr/bench/rjmzdu/1
Multi Threads = 19874
 

http://valid.x86.fr/bench/rjmzdu/16
AMD 1700x vs Intel 6800k
PCEVA：http://bbs.pceva.com.cn/thread-137649-1-1.html
100x @3.4Ghz vs 6800k@3.4Ghz

Intel：
Intel Core i7 6800K ：3.4G
ASUS STRIX X99 GAMING
16GB DDR4-2400（dual channel）
RX 480 8G 1266/2000 8GB GDDR5
BIOS：1401
OS：Win10 64bit
Crimson 17.2.1
2017.2.23

AMD：
AMD Ryzen 7 1700X ：3.4G
ASUS PRIME X370-PRO
16GB DDR4-2133
RX 480 8G 1266/2000 8GB GDDR5
BIOS：404
OS：Win10 64bit
Crimson 17.2.1
2017.2.10



Translated by CortexA99

3DMark Firestrike Physic score
3DMark11 Physic score
Ludashi 5.15 CPU score
Ludashi 5.15 GPU score
Ludashi 5.15 Memory score
Ludashi 5.15 Disk I/O score
Blender 2.78A (second, lower the better)
7zip 15.05 beta 1min
LuxMark 3.1 CPU Opencl
Fritzchess
Cinebench R15 ST
Cinebench R15 MT
Platform power consumption Full-load (lower the better)
Platform power consumption Idle (lower the better)
Platform power consumption Gaming (lower the better)
Platform power consumption Office-usage (lower the better)
Ashes of the Singularity DX12 - Average FPS
Battlefield 1 Average FPS
Call of Duty 13 Average FPS
Counter Strike Global Offensive Average FPS
Counter Strike Global Offensive Minimum FPS
CrossFire 4.1.8 Average FPS
CrossFire 4.1.8 Minimum FPS
Deus Ex: Mankind Divided Average FPS
Deus Ex: Mankind Divided Minimum FPS
DOOM-VULKAN Average FPS
DOOM-VULKAN Minimum FPS
H1Z1 | King of the Kill Average FPS
H1Z1 | King of the Kill Minimum FPS
Hitman 6 Average FPS
Tomb Raider DX12 Average FPS
Tomb Raider DX12 Minimum FPS
Civilization 6 DX12 Average FPS
Civilization 6 DX12 Minimum FPS
Tom Clancy's The Division DX12 Average FPS
World of Tanks Average FPS
World of Tanks Minimum FPS

[Komadi]

Ryzen really seems promising

[kladzen]

2 minutes ago, Koenyie said:

Ryzen really seems promising

It sure does!

[Simon771]

It just makes me sad that single core performance still isn't that great.

150 in cinebecnh ... hopefully we will be able to OC it and get at least 180 out of it.

[kladzen]

4 minutes ago, Simon771 said:

It just makes me sad that single core performance still isn't that great.

150 in cinebecnh ... hopefully we will be able to OC it and get at least 180 out of it.

Sure it can be overclocked... at least to 4.4ghz.. we know that from MSI...  I would guess that we will likely have 4.5 4.6ghz average overclock's with water

[MageTank]

1 minute ago, kladzen said:

Sure it can be overclocked... at least to 4.4ghz.. we know that from MSI...  I would guess that we will likely have 4.5 4.6ghz average overclock's with water

You cannot use the words "at least" when talking about overclocking. The sample size is too low. I don't care if MSI overclocked to 5ghz, we don't know anything yet. 

[LAwLz]

Those are very promising results.

Would be interesting to see what the average turns out to be. Seems to be like ~3% for single core stuff and like 35% for multicore things.

[Guest Darth Revan]

Why do people still say oh too bad single core performance isn't that good?

How many programs are still limited to single core in 2017?

[kladzen]

1 minute ago, MageTank said:

You cannot use the words "at least" when talking about overclocking. The sample size is too low. I don't care if MSI overclocked to 5ghz, we don't know anything yet. 

Well when MSI wants to make a marketing stunt about their "boost" function... i'm fairly sure that they are confident that most people will be able to do such a OC... sure some will be getting some shitty cpu's that wont hit that mark... and then the average.. propperly will ( i assume since MSI uses it as marketing ) and then the godlike cpu's that hit 4.7-4.8 maybe even 4.9 on water.... beyond that its LN2 i would assume

[Ashiella]

5 minutes ago, MageTank said:

You cannot use the words "at least" when talking about overclocking. The sample size is too low. I don't care if MSI overclocked to 5ghz, we don't know anything yet. 

Yeah, they coulda cherry picked the chip.

[MageTank]

Just now, Darth Revan said:

Why do people still say oh too bad single core performance isn't that good?

How many programs are still limited to single core in 2017?

Most of the time, your minimum framerates are tied directly to single threaded loads. No matter how well you try with parallelism, some aspects simply fall back to single threads. Amdahl's Law isn't obsolete just yet.

 

Just now, kladzen said:

Well when MSI wants to make a marketing stunt about their "boost" function... i'm fairly sure that they are confident that most people will be able to do such a OC... sure some will be getting some shitty cpu's that wont hit that mark... and then the average.. propperly will ( i assume since MSI uses it as marketing ) and then the godlike cpu's that hit 4.7-4.8 maybe even 4.9 on water.... beyond that its LN2 i would assume

I don't care what MSI is saying, or what they are marketing. You simply cannot mislead people by using the words "at least" in regards to overclocking. No overclock is guaranteed. As I've already stated, we know nothing of the sample sizes being tested, and we don't know if they received binned chips in their tests. I bet if you check MSI's marketing, they use two magical words. "Up To". Every single company uses these words to avoid legal repercussions. "Up to 60Mbps". "Up to 4ghz". See the trend here?

[Simon771]

3 minutes ago, kladzen said:

Sure it can be overclocked... at least to 4.4ghz.. we know that from MSI...  I would guess that we will likely have 4.5 4.6ghz average overclock's with water

I'm still not sure about that. R7 1700, 1700X and 1800X are all the same chips. Just slower clock speed, and for the last one they even decreased TDP.

But what's the point of buying 1800X for 500$, if you can buy 1700 for 320$ and OC it to 4,4GHz (10% better than R7 1800X) ... Something just doesn't feels right here.

I think that they made million of those 8 core CPUs, and then test them to see how far can they clock them. So it's something like silicon lottery, but AMD already sorted them.

If any R7 1700 can get up to 4,2GHz at least, I will be buying that instead of 1700X. But I will wait for reviews and benchmarks from other people to see how good does 1700 OC compared to 1700X and 1800X.

2 minutes ago, Tedny said:

Vulkan and Dx12 advantage from multicore performance and you know, they are future, not dx11 or OpenGL(and others OpenSomething) 

Yes Vulkan and DX12 (together with great game engine) will benefit from more cores even at lower IPC.

But right now there aren't many games that support DX12/vulkan. And the game that I play (BDO) is depending on single core performance and it can't utilise all cores anyway.

So yes, I do agree that this is the future, but right now not so much.

[Simon771]

8 minutes ago, Darth Revan said:

Why do people still say oh too bad single core performance isn't that good?

How many programs are still limited to single core in 2017?

Because 99% of games still depend on single core performance rather than overall CPU power.

I wish all game developers would update their games to DX12 and Vulkan so we would all benefit from it, but it's not that simple update. It would take them way too much time, and they will just stick to DX9 or DX11.

1 minute ago, Tedny said:

I am talking not only about games engines, those APIs will greatly improve others application in Windows and others OS 

That's true, but some of us only care about good gaming experience  

[Bouzoo]

Nothing shows your FPS like benchmarking CS:GO, COD13 (I had to google what exactly that was), and WoT. And I'm surprised they used an RX480. 

[79wjd]

11 minutes ago, huilun02 said:

I'm still not sure about that. 7700K, 7700, 7600K and 7600 are all the same chips. Just slower clock speed and HT disabled, and for the last one they even decreased TDP. But what's the point of having a 7700K for 330+$, if the 7600 was allowed to OC with HT for 200+$ and could also get to 5GHz... Something just doesn't feels right here.

I think they made a million of those 4 core CPUs, and then test them to see how far can they clock them. And then only the best gets unlocked with HT enabled, followed by unlocked without HT and then totally locked without HT. So it's something like the silicon lottery, but Intel deliberately limits the potential of lower end SKUs.

The difference is that there is since incentive to buy the -k SKUs as the non-k SKUs don't offer HT. Sure it's possible that AMD is charging 50% more just for a stock clockspeed bump (1700 to 1800x), and it certainly wouldn't be unprecedented (just look at how the 9590 was around 300% more expensive; and Intel also does it with the 7600 being 20%~ more than a 7400), but there is still a big part of me that thinks that there will be some form of limitations -- power limited or just much lower quality chips.

[TOMPPIX]

[TigerHawk]

I'm one of the ones that foolishly preordered zen and I have to say these benchmarks reek of BS. It would be great if they were true, but I'm not even close to believing any of it, outside of it probably beating most of intel's desktop lineup in multi-threaded at least.

 

Seriously this CPU has to be the most hyped up thing in recent memory. Holy cow. I'm sure it'll be a decent cpu and not a letdown like bulldozer, but you'd think it was the second coming of Christ or something with all this hype going on.

[zMeul]

I want to see benchmarks with equivalent Kabi Lake

There are Kabis that shit on 6800K

 

And that shithead accused Intel of foul play

[goodtofufriday]

Source: http://valid.x86.fr/bench/rjmzdu/1

 

 

 

We don't know What sort of cooling this validated result was using, but it is rather promising. I'm still holding out for some 1800x benches on the x300 chipset personally. 

[Guest Generallee]

OUHH BOY i haven't been this excited since the release of the Phenom X3

 

AMD IS BACK, AWWWW YISS 

[Bouzoo]

People are still believing that a stock 1700X will beat a stock 7700K...

[goodtofufriday]

4 minutes ago, Bouzoo said:

People are still believing that a stock 1700X will beat a stock 7700K...

I don't think I've heard or even have read of anyone that believes that. The TITLE says it's OC'd

[TigerHawk]

repost (the link contains these results)

 

 

[VagabondWraith]

Just now, Bouzoo said:

People are still believing that a stock 1700X will beat a stock 7700K...

Can't wait for Friday for all the shitposting to cease.

[iRileyx]

There's been benchmarks on userbenchmark for the last like 4/5 days  at stock speeds its not close to a 7700k on single/quad performance, so they must have gotten a pretty good overclock.

 

http://cpu.userbenchmark.com/AMD-Ryzen-7-1700X/Rating/3915