Some people are still not getting the point - I am referring to both AMD and Intel camps.

7700K is an obvious winner for gaming-only cpu. Before down putting Ryzen, Intel camp also need to realize that 7700K also game better than their more expensive 6900K and 6800K. In multi-threaded applications and synthetic benchmarks, R7 matches well with Intel HEDT system, but still loose at gaming. Why? It's all about the clockspeed.

This is the day of GPU being the main performance defacto of your system and gone are the days when MMX and 3D Now were the in-thing. You need a fast enough cpu to feed data to the gpu back and forth.

In gaming, raw clockspeed still has the most advantage in today's games. Ryzen did not lose to Skylake or Kabylake because of IPC, it's because of clockspeed. The gap in clockspeed is significant that extra cores do not give much advantage to overall performance. We can see the 1080p gaming comparison video by Joker Production (Youtube) where the 1700 while utilizing all-cores, the load is very low which means the games are not taking full advantage of the extra cores (a.k.a lazy cores).

What if Ryzen has the exact same clock speed as Skylake/Kabylake? But of course it's not the current situation here because it is hard for an 8-core Ryzen to match the clock of Kaby even with watercooling. But at 3.9GHz overclock, the performance is near to Intel 5ghz. The problem now is that with its base clock, some reviewers are reporting that Ryzen has issues with boost clock. Hardware Unboxed claimed that their 1800X is somewhat stucked at 3.7GHz on his Asrock Taichi board. And I have yet seeing any reviewer that actually analyzes how the boost behavior is on Ryzen. How many active cores are at specific boost clock?

It was different with the old Bulldozer architecture when its high clockspeed was not an advantage due to the FPU resource sharing. In layman's term you can say the performance of clockspeed advantage is cut into half (not literally).

There was another argument regarding gaming resolution. In real world, those with 4K monitors and capable GPU will game at resolutions higher than just 1080p. At 4K the cpu advantage is diminished. 1080p testing is only relevant ONLY IF you want to measure the cpu performance. 7700K wins, hands down, but is it an ideal real world situation? No. If we're being practical... if you have a 4K capable system you will game at 4k, not 1080p. 1080p testing is only if you're stucked with 1080p monitor and it's capable of outputting very high refresh rate. Otherwise, it just serves another bench and bragging right. At 4K gaming and very high end GPU, even the old FX can perform as well as the newer generation CPUs.

This is just some of my thoughts at the moment, will write again later once I have more ideas.

IMHO, since the fps is not very big. 8c/16t is more future, without any optimisation Ryzen still run as good Intel. It's just matter u want risk 3-5 fps for better future gaming.

Now we just have to wait for that awesome fps game that can take advantage of 16 threads. Battlefield with 128 players... XD

Yes, thing is consumer like me dominate the market, not content creator. I'm paying to game. So despite the benchmark shows a very large performance for content creator. That isn't what I want. What I'm expecting for is just a cheap quad core which has better single thread performance. Heck I don't even want to buy a CPU for over MYR 2k price.

Feel free to correct if I'm wrong.

And that's the problem. You had your expectation, so it's not the product that failed you but rather you had expectations.

If you look at the presentation past to present, AMD didn't market much on the gaming aspect.

Fact remains that gamers isn't their main market, they look at overall - gamer, content creator and servers.

goldfries, did you pair ryzen with amd card? vortez side socre "higher" with rx 480. Lastly, disabling smt does it make a differences in testing?

https://www.vortez.net/articles_pages/amd_r..._review,19.html

I did but I so no point doing benchmarks on RX 470.

Among the 3 Ryzen models, I can't see the point of buying the 1800x. 1700 can overclock just as high as 1800x (maybe just 100mhz lower). Same 8c 16t and same amount of cache. Perhaps AMD traditionally likes to give 'free' performance to end users, like some Phenom II that can be unlocked to gain extra cores.

Among the 3 Ryzen models, I can't see the point of buying the 1800x. 1700 can overclock just as high as 1800x (maybe just 100mhz lower). Same 8c 16t and same amount of cache. Perhaps AMD traditionally likes to give 'free' performance to end users, like some Phenom II that can be unlocked to gain extra cores.

Any idea on power consumption? from many review site, power consumption higher than advertised.

Any idea on power consumption? from many review site, power consumption higher than advertised.

Depend on your workload, i think the power consumption is really good for 8c/16t processors.

i mix up tdp with power consumption. Thought that tdp = power consumption.

Yeah, including myself sometime.

TomsHardware - Various loads | Processor load read from Mainboard sensor [1800x]


Goldfries - OCCT Linpack with AVX | Full system load with RX-470 [1800x]


PCPer - Cinebench r15 | Full System load [1800x]


HotHardware - Full system load [1800x, 1700x, 1700]

The power consumption is better than Intel 8 core parts and even 7700k in certain case which is great. But from the graph can be seen once overclocked, the power consumption shoots 

The overclocking headroom for the higher-end Ryzen models is rather slim. This was expected due to the relatively high stock frequencies, high-density orientation of the design and the low power targeted manufacturing process used for the Zeppelin die (Samsung 14nm LPP).

As indicated by the Vmin-Fmax curve, Zeppelin's voltage scaling is perfectly linear until 3.3GHz (25mV per 100MHz). The first deviation ("Critical 1") from this linear behavior can be seen at 3.3GHz. The second and the final deviation ("Critical 2") can be seen at 3.5GHz. Beyond this point the voltage scaling is neither linear or recovers even temporarily, and the CPU is requiring higher voltage in increasingly larger steps to scale further.

The ideal frequency range for the process or the design (as a whole) appears to be 2.1 - 3.3GHz (25mV per 100MHz). Above this region (>= 3.3GHz) the voltage scaling gradually deteriorates to 40 - 100mV+ per 100MHz.

This means that at ~3.8GHz pushing further usually becomes extremely costly (power / thermal wise).

In comparison, the "critical" points for the two previous AMD desktop designs were at:

- Orochi Rev. C aka Vishera, 32nm SHP SOI - (1 = 4.4GHz, 2 = 4.7GHz)

- Kaveri / Godavari, 28nm "SHP" HPP Planar - (1 = 4.3GHz, 2 = 4.5GHz)

The voltage scaling indicated by the Vmin-Fmax curve (above) can be also clearly seen in the default voltages for the different frequency states (PStates) of the CPU.

On the high-end models the actual (effective) voltage for the base frequency (e.g. 3.6GHz on 1800X SKU) can be anything between 1.200 - 1.300V. Meanwhile the actual (effective) voltage for the highest single core boosted PState (XFR, e.g. 4.1GHz) can be as high as 1.47500V.

In the tested sample the actual default voltage for the base frequency (P0, 3.6GHz) was ~1.25000V, while the highest single core boost state (XFR, 4.1GHz) defaulted to 1.4625V.

The power is better than Intel 8 core parts and even 7700k in certain case which is great. But from the graph can be seen once overclocked, the power consumption shoots