Official AMD Ryzen Benchmarks, Reviews, Prices, and Discussion

[unseenmorbidity]

The wraith spire is junk... It's not going to cool an 8 core, even at stock.

According to a "leak", you could buy the cpu without the cooler.

 

[CHADBOGA]

It was created as an amd bashing thread by a troll that devotes countless hours of his life to hate amd.

I bet if I go look at that old thread I will see that Fritz benchmark that was proven fake almost immediately after they were posted.

The perpetrators mostly went to troll another forum, and it is a far better environment because of it. I imagine the shame will keep them away for awhile.

EDIT: Ohh look there it is...

PS: idk why you want to stir that pot though. Just let it go.

The previous thread was just using what info was around at the time.

 

[unseenmorbidity]

The previous thread was just using what info was around at the time.

I just proved that wasn't true.

 

[lolfail9001]

The wraith spire is junk... It's not going to cool an 8 core, even at stock.

Fairly certain it will. Even my old Zalman flower could cool a 95W TDP CPU just fine.

 

[krumme]

Ouch.

But keep the 6900 at 4GHz and its 153w or so even with a serious fpu load.
The last 100MHz is expensive.
My bet is most will run their 8c zen at max 4.4. Above 200w you are into serious cooling.
The water cooling stuff might for a change actually give a serious benefit because of w/die size challenge.
An interesting and cheap enthuaiast cpu to play with.

 

[unseenmorbidity]

Fairly certain it will. Even my old Zalman flower could cool a 95W TDP CPU just fine.

Intel and amd aren't using the same metrics to define tdp.

 

[shabby]

We have to remember this is 8c cpu the 6900 at 4.3 uses 210w

How did amd outdo intel in the power efficiency game? The 1700 @ 3ghz has a 65w tdp while a 3.2ghz 6900k is 140w... mind boggling i tell ya.
Currently the 6900k is $1050 on newegg with a $50 discount, lets see how fast this will drop.

 

[lolfail9001]

Intel and amd aren't using the same metrics to define tdp.

I know, except that Ryzen has no obvious critical situation when it's power consumption would skyrocket like AVX on Intel. And for typical power consumption, 95W is a fine ballpark for 6900k and Ryzen.

 

[krumme]

How did amd outdo intel in the power efficiency game? The 1700 @ 3ghz has a 65w tdp while a 3.2ghz 6900k is 140w... mind boggling i tell ya.
Currently the 6900k is $1050 on newegg with a $50 discount, lets see how fast this will drop.

Naa. Lets wait and see where it lands. It should be compared with similar load. And the 6900 have some taxing fat 256 bit wide fpu units. For desktop i am not sure.

But granted. Where efficiency plays a major part at servers to me it looks like amd have a winner. I think 14lpp will shine at lower freq. 2.6 GHz 32c at 180w tdp?

Perhaps we should try some undervolting with zen ?

 

[Doom2pro]

How did amd outdo intel in the power efficiency game? The 1700 @ 3ghz has a 65w tdp while a 3.2ghz 6900k is 140w... mind boggling i tell ya.
Currently the 6900k is $1050 on newegg with a $50 discount, lets see how fast this will drop.

My guess is Intel didn't give a damn for so long due to no competition so they got complacent and just half-a**ed their designs.

AMD however absolutely needed to improve in every single area, and they did.

 

[sirmo]

I had a feeling AMD was going to top Intel in efficiency. Anyone paying attention could have seen what a huge progress they have made on the dozer cores with Carrizo.

 

[Absolute0]

Honestly I can't make sense of AMD's TDP numbers. The 65W sku's ship with 95W coolers; the 95W sku's ship with 140W coolers.

The 4C4T 1100 is rated 65W; the 8C16T 1700 is rated at 65W

Then the 6C12T 1600X is rated 95W

And the only leak we have re: power consumption (from Canard PC December leak) shows the 3.3 Ghz ES drawing 93W under load.

 

[psolord]

I am think of getting a d15. If that cannot keep it cool, then that's insane!

I have a D14 myself and does a pretty good job on my 4.8Ghz 2500k. What kind of power draw does a 4.8Ghz has and what kind of heat dissipation, I don't exactly know, but I could measure the delta from stock and get a ballpark figure.

Now these things (the D14 and D15), can dissipate up to 220W. That seems like a lot of headroom from the stock 95W.

 

[CHADBOGA]

I just proved that wasn't true.

No you didn't.

 

[IEC]

Honestly I can't make sense of AMD's TDP numbers. The 65W sku's ship with 95W coolers; the 95W sku's ship with 140W coolers.

The 4C4T 1100 is rated 65W; the 8C16T 1700 is rated at 65W

Then the 6C12T 1600X is rated 95W

And the only leak we have re: power consumption (from Canard PC December leak) shows the 3.3 Ghz ES drawing 93W under load.

I would guess they are optimizing for quiet performance at stock clocks. Because end users are a lot more sensitive to noise these days versus the days of 7000 rpm Delta screamers to get that last 100MHz OC.

 

[JDG1980]

Honestly I can't make sense of AMD's TDP numbers. The 65W sku's ship with 95W coolers; the 95W sku's ship with 140W coolers.

I can think of at least two good reasons for that:
(1) An oversized cooler can run the fan at a lower RPM, resulting in quieter operation.
(2) An oversized cooler allows the user to overclock and/or use XFR (which can go above the specified TDP) without overheating.

 

[Mopetar]

How did amd outdo intel in the power efficiency game? The 1700 @ 3ghz has a 65w tdp while a 3.2ghz 6900k is 140w

The Intel cheap has some beefier hardware for certain tasks. For example, Ryzen doesn't have any 256-bit AVX registers, but that doesn't matter for a lot of workloads. Ryzen also has larger L1 and L2 caches (though Intel's L3 cache is larger) which likely contribute to better performance in certain workloads as well. Also, I wouldn't use raw TDP numbers to compare efficiency as that's merely what the chip can draw up to under normal circumstances. Here's one article that looked at actual power draw for the system and came away with Ryzen being only slightly more efficient because the Intel chip was drawing well below its 140W TDP.

Also, AMD is releasing a new architecture that has the advantage of being able to learn from previous and current architectures while Intel is pushing refinements to their Core architecture that's about decade old at this point. Early on they could make double digit gains from generation to generation, but I think they've nearly reached the full potential of the design and run out of low-hanging fruit as evidence by the recent Kaby Lake release. They're also at a point where they're near the end of their 14 nm process, but haven't moved to the next process yet, which puts them on more even footing with the competition, when historically they were typically a node ahead.

Much like AMD, Intel is going to need a new design. They've probably learned some things over the years or have ideas for better designs, but they may be so radically different from the current design or necessitate further changes that those ideas can't be implemented. Ever since Bulldozer, Intel hasn't had a lot of reason to create a brand new architecture and they may have underestimated AMD's ability to have anything competitive so instead of starting on a new architecture they've just been refining the existing one and focused their efforts into trying to break into new markets.

 

[imported_jjj]

In those Cinebench slides, looks like the 6900k runs ST at 4GHz given the MT score.

 

[looncraz]

Me too. If Ryzen absolutely requires water cooling to push it.. well, we'll see. If not I already have a decent air cooler (TRUE Spirit 140m Power, 6x8mm heatpipes) that I'd have to mod the backplate and bracket (it's just drilling 8 holes) to make it compatible with AM4.

That cooler has a delta-T of ~50C at 150W coming from an i7 920 D0 @ 4.2GHz. That die is 263mm^2, so has an easier time transferring heat than Ryzen will at its measely ~192mm^2.

So, given that 8-core Ryzen uses ~65W at 3.4Ghz (CPU-only) and ~150W at 4.4Ghz, has a likely safe temperature limit around 80C, and 37% less surface area for transferring heat, I suspect you will find yourself running about ~75C at 4.1GHz on 8-cores with that heatsink. Accounting for the added power draw for the higher temperature, you will likely "creep" upward from about ~72C a few degrees, which is how I come to ~75C.

A Noctua NH-D15, from my math, based on theory and very little real world data (so a few salt mines are needed to satiate the requirements here - as with the above), will allow about 4.25Ghz at the same temperature. Be mindful: these are equilibrium temperatures with assumed wattage outputs... lots of assumptions, in fact, such as 22C ambient temperatures (my house temps).

Standard water-cooling, however, should be able to keep the case(IHS) temperature low enough to keep the delta at the die-IHS interface high, which increases thermal efficiency, which reduces power needs, and allows for better results. The actual amount of heat is, by no means, a problem. It's getting the heat out of the die through its small interface with the IHS that is problematic.

A heatsink with a larger thermal mass will take longer to reach equilibrium temps, so you could probably run 4.4Ghz with gaming loads, if the CPU can handle it. Overclocking should improve with BIOS updates, so keep up to date

 

[JDG1980]

For example, Ryzen doesn't have any 256-bit AVX registers, but that doesn't matter for a lot of workloads.

My understanding is that this is similar to how Sandy Bridge was designed, as well:

Sandy Bridge allows 256-bit AVX instructions to borrow 128-bits of the integer SIMD datapath. This minimizes the impact of AVX on the execution die area while enabling twice the FP throughput, you get two 256-bit AVX operations per clock (+ one 256-bit AVX load).​

Zen has two FPU ADD units and two FPU MUL units, each 128 bits wide. Presumably AVX will be handled the same way as on Sandy Bridge, and then should provide the same throughput. Haswell and up have 2x the theoretical AVX throughput of Sandy Bridge, but that doesn't mean double speed in real applications, not even x264.

 

[IEC]

@looncraz That's reasonable. If the die size is indeed that small then it looks like I have my excuse to go water!

 

[HurleyBird]

The Intel cheap has some beefier hardware for certain tasks. For example, Ryzen doesn't have any 256-bit AVX registers, but that doesn't matter for a lot of workloads. Ryzen also has larger L1 and L2 caches (though Intel's L3 cache is larger) which likely contribute to better performance in certain workloads as well.

Although the L3 deficit is only against HEDT. Ryzen has the same amount of L3/Core as non-enterprise Intel parts eg. 7700K.

 

[JDG1980]

In those Cinebench slides, looks like the 6900k runs ST at 4GHz given the MT score.

Tom's Hardware observed this during their review of the 6900K - apparently, in at least one instance, a single threaded workload did hit 4GHz just through boost without overclocking.

 

[esquared]

Folks, stop talking about that other thread and the reasons that it was or was not updated.

Its done. It's locked. Focus on this topic. There are multiple people already in this thread,
that are derailing it. All in less than 4 hours.

Continue to do so and Markfw will be levying infractions.


esquared
Anandtech Forum Director

 

[Ryzenenthusiast]

Noctua Nh D15 is just 1-2 degrees higher than the best of best AIO's under the same loads. So I think if Ryzen 7 chips can hit 4,5 Ghz with realistic voltages, then Noctua Nh D15 won't have a problem to keep up with that clock.