AMD Ryzen 5 3600 Beats Intel’s Core i7-9700K in Cinebench

[mrb]

Since 2000, AMD has often beaten Intel on multi-threaded workloads, but never single-threaded (at comparable clocks). This bench shows AMD beating Intel on both! That's what's incredibly notable here. The last time it happened was 20 years ago with the AMD K7.

Now Zen 2 seems to overall beat Intel on every metric: single-threaded perf, multi-threaded perf, perf/dollar, perf/watt. No matter how you look at it, Zen 2 comes out on top.¹ Very impressive.

Man the folks at Intel must feel the heat.

¹ Except perf/socket when competing with the Xeon 9200, but that's just a PR stunt no one cares about: https://mobile.twitter.com/zorinaq/status/113576693566724096...

[nolok]

You want notable ? AMD is currently beating Intel on price, single threaded perf, multi threaded perf, TDP/power usage at equivalent performance AND on many of the "little pluses on the side" (more PCI-e lanes, ECC supports, ...).

When Zen first came is was a huge deal, but it merely put them as a real competitors, with a decent advantage on many cases, reinforced with Zen+. But Zen 2 put them ahead in almost every category, and in all markets; threadripper and EPYC are just as strong in their areas.

Either Intel has something strong about to appear, or they're going to face a truly difficult few years with customers going AMD now that it's not merely "one generation of chip" that was good. It feels like getting their 10nm working will not be enough by itself.

[kllrnohj]

> The last time it happened was 20 years ago with the AMD K7.

K8, too. AMD's IPC was so far ahead of Intel's at that time it was crazy. 2.2GHZ Athlon 64's were keeping up with or beating the 3-3.2ghz P4's.

It was so strong & competitive Intel resorted to straight up bribery to compete, resulting in many anti-trust judgements against it as a result. But they succeeded in preventing K8 from hurting their market share, and kept AMD down despite a vastly superior product. Here's hoping that doesn't happen again this time around, but maybe Intel will decide the wrist slap is worth it.

[neogodless]

There's one thing left that I have both doubts and excitement for:

How good are AMD's laptop chips? The improvement in IPC and efficiency in Zen 2 can go a long way in improving this, and then, of course, they must improve perception.

Anecdotally: I've owned almost exclusively AMD chips in my desktop builds for the past 15 years. I've never once owned an AMD laptop. When Intel built the Core line of chips, they seemed to nail laptop first, and then apply the efficiency to their desktop line with higher clocks. It worked wonders.

In my opinion, AMD really needs to nail laptop CPUs/APUs now more than ever. I hope they do!

[kteofanidis]

Chances are I'll be upgrading to a Ryzen 3000 this year. Now that I think about it the last time I ran an AMD CPU was indeed the K7.

[leetcrew]

iirc, amd was beating intel on single thread in the pentium 4 era with way lower clocks.

[notacoward]

> Man the folks at Intel must feel the heat.

Well, their processors are very effective space heaters, so yeah.

[hurrdurr2]

This is great news for consumers. I've always wanted to build another pc based on AMD chips and it looks like Zen 2 is going to be it.

[chrisdfrey]

I'd argue AMD and Intel were pretty close in single-threaded performance until the Core 2 Duo (released in 2006).

[criley2]

I'm happy that AMD really seems to be surging again. It was finally time for the ol' mobo+cpu+ram rebuild for me recently, and I ended up replacing my second generation core i5 2500k with a new Ryzen 5 and I could not be happier. The price was spot on compared to how much more I would have paid for an intel chip and board. I believe I got the Ryzen 5 2600, so one generation back from this 3600 I assume.

[Sileni]

Did a 2600x / Vega 56 build recently, and I'm honestly impressed with the value. Total under $1k after tax. I would have spent that before getting a GPU if I had gone Intel.

[noir_lord]

I went from a 2500K with HD6950 to a 2700X and a RTX2080.

That was a slight jump.

Yep the 2600 is Zen+, 3600 is Zen 2, decent uplift in IPC (about 10%) on its own not worth upgrading.

That 3900X looks tempting though to replace my 2700X.

50% more faster cores than an already very fast processor, AMD are on fire at the moment.

[127]

Since nobody yet mentioned it, just going to remind everyone that a 16 core, 32 thread desktop chip for Zen 2 will exist, and be priced under 800 dollars.

[teolandon]

Source on that <800 dollar figure? Is it all just leaks?

[vardump]

Sure. But two memory channels is not amazing. I'll hold out for ThreadRipper, with 4 channels. Although AMD Rome Epyc got massive 8.

[baybal2]

They will surely push 12 cored ones first. They glue 2 downgraded dies and get ones CPU that costs more than 2 defective dies sold separately.

Just brilliant

[wil421]

Anyone have any experience with AMD and ECC memory? I was going to do an AMD build for a NAS but the support for ecc was kinda like:

“We enabled it but didn’t test it and we don’t guarantee it will work.”

[juergbi]

Works fine here with 1st Gen Threadripper on ASRock X399 Taichi, running Linux. Some points:

* Not all motherboards support ECC, make sure to check before buying.

* Only unbuffered DIMMs are supported by Ryzen/Threadripper.

* According to motherboard vendors, Ryzen APUs do not support ECC. Ryzen Pro APUs do support ECC but those SKUs are typically not available in retail.

[tiew9Vii]

Check the motherboard specs, they are a bit hit and miss when I was looking.

I am using an ASROCK - B450M Pro4, Crucial - CT16G4WFD8266 16GB ECC with a Ryzen 5 2600.

[zrobotics]

I'm genuinely curious, for a Nas does ECC RAM provide any benefit? I've always just used old desktop hardware, typically from about 4 generations prior (when I upgrade the desktop, the old hardware goes into the lab PC, and the lab PC goes into the NAS). I know there are power consumption benefits to the specialized NAS boards, but would I see any benefit from ECC? Considering the NAS is a RAID array of spinning rust drives, a flipped bit in RAM really shouldn't affect anything seriously. Or am I grossly mistaken?

[std_throwaway]

Usually it works without problems. Check with the mainboard manufacturer if it supports ECC memory. Most do, some don't.

[starman100]

That's our impression, too. There are AMD Motherboards that take ECC memory, but we've never seen them act on ECC errors that were uncorrectable (the correctably errors are handled, but uncorrectable errors aren't reported!).

We will only use Intel Xeon for our work because of this. You'll get about 1 bit flip/GB/year. With 128 GB or more in our standard builds, this would be more than 2/week. We just can't have that uncertainty in the data we provide.

And while Cinebench is a useful benchmark, all our heavy number crunching is done on NVidia 2080 architecture so the fact that AMD may have an advantage on some cases isn't that interesting for us. Perhaps if you're a gamer, who doesn't care about an occasional bitflip, looking to squeeze the last drop of value out for his dollar....

[koolba]

Any plans for AMD to release first party NUC form factor devices with Ryder chips? I bet I’m not the only one that both loves that form factor and also amazed at the perf numbers and core counts of these CPUs.

[std_throwaway]

I'll wait till the embargo is lifted and detailed benchmarking will be available.

The date is rumored to be the 7th of July.

[flyinghamster]

I must say, I've been quite happy with my now "old" 7/1700. Between it and the NVMe SSD, it's been quite a snappy machine. Even before Zen, I preferred AMD for my builds just for the bang-for-the-buck. It's great to see that they're cranking up the performance per dollar to new heights.

[pingec]

Are the Intel scores from these benchmarks with meltdown, spectre etc. mitigations enabled or disabled?

[akbT]

Is it known yet whether the 32-core Threadripper will be discontinued or whether a potential 32-core 7nm Epyc will run at higher clock rates?

I'm searching for the fastest C++ compile machine, that includes linking (which is single core).

[ivl]

Seeing that screenshot and how high well the i7-6900k hangs with much more recent hardware impresses me. I'm quite glad I went with Broadwell E. Probably means I'll be looking to upgrade around Zen 2+.

[jcelerier]

as the owner of a 4.3ghz 6900k which I already find way too slow... cries

[xupybd]

This is good news but how many daily tasks come close to being represented by Cinebench?

Is Ryzen a better choice for a daily driver?

[studmuffin650]

Jim Keller strikes again. I swear every place this guy goes, groundbreaking chip architecture work is done.

[Nursie]

So it looks like the 3700X should be reasonably expected to be on a par with or better than the i9-9900K?

And that's very much in the middle of the zen 2 offerings... and at a significantly lower price.

[andy_ppp]

I want to see performance between Zen 2 and Intel with security mitigations enabled on Intel (AMD don't need any with Zen 2).

Should be interesting and I suspect on even games AMD will be 10%+ ahead.

[vkaku]

I'm quite excited with what AMD has pushed on socket AM4 over the last three years.

Come September they're going to release the 3950x,and if folks can wait a bit, you get 16 cores/32 threads on one socket.

But even on a budget, the 3600G looks like a great buy with Navi 20+3600 on chip.

The deal is that they have reasonable spectre mitigations working since last few gens, and Intel has to fix this if they need to be competitive. Not to mention standardizing a chipset.

[kitchenkarma]

I need single core performance and I am maxing out 9900K @ 5.1. If AMD proves to be faster than that, show me where to pay.

[harry8]

Anyone know where to find an inter-thread latency benchmark comparison?

[fibers]

does this mean they broke the embargo for 7/7??