The AMD 2nd Gen Ryzen Deep Dive: The 2700X, 2700, 2600X, and 2600 Tested
by Ian Cutress on April 19, 2018 9:00 AM ESTImprovements to the Cache Hierarchy
The biggest under-the-hood change for the Ryzen 2000-series processors is in the cache latency. AMD is claiming that they were able to knock one-cycle from L1 and L2 caches, several cycles from L3, and better DRAM performance. Because pure core IPC is intimately intertwined with the caches (the size, the latency, the bandwidth), these new numbers are leading AMD to claim that these new processors can offer a +3% IPC gain over the previous generation.
The numbers AMD gives are:
- 13% Better L1 Latency (1.10ns vs 0.95ns)
- 34% Better L2 Latency (4.6ns vs 3.0ns)
- 16% Better L3 Latency (11.0ns vs 9.2ns)
- 11% Better Memory Latency (74ns vs 66ns at DDR4-3200)
- Increased DRAM Frequency Support (DDR4-2666 vs DDR4-2933)
It is interesting that in the official slide deck AMD quotes latency measured as time, although in private conversations in our briefing it was discussed in terms of clock cycles. Ultimately latency measured as time can take advantage of other internal enhancements; however a pure engineer prefers to discuss clock cycles.
Naturally we went ahead to test the two aspects of this equation: are the cache metrics actually lower, and do we get an IPC uplift?
Cache Me Ousside, How Bow Dah?
For our testing, we use a memory latency checker over the stride range of the cache hierarchy of a single core. For this test we used the following:
- Ryzen 7 2700X (Zen+)
- Ryzen 5 2400G (Zen APU)
- Ryzen 7 1800X (Zen)
- Intel Core i7-8700K (Coffee Lake)
- Intel Core i7-7700K (Kaby Lake)
The most obvious comparison is between the AMD processors. Here we have the Ryzen 7 1800X from the initial launch, the Ryzen 5 2400G APU that pairs Zen cores with Vega graphics, and the new Ryzen 7 2700X processor.
This graph is logarithmic in both axes.
This graph shows that in every phase of the cache design, the newest Ryzen 7 2700X requires fewer core clocks. The biggest difference is on the L2 cache latency, but L3 has a sizeable gain as well. The reason that the L2 gain is so large, especially between the 1800X and 2700X, is an interesting story.
When AMD first launched the Ryzen 7 1800X, the L2 latency was tested and listed at 17 clocks. This was a little high – it turns out that the engineers had intended for the L2 latency to be 12 clocks initially, but run out of time to tune the firmware and layout before sending the design off to be manufactured, leaving 17 cycles as the best compromise based on what the design was capable of and did not cause issues. With Threadripper and the Ryzen APUs, AMD tweaked the design enough to hit an L2 latency of 12 cycles, which was not specifically promoted at the time despite the benefits it provides. Now with the Ryzen 2000-series, AMD has reduced it down further to 11 cycles. We were told that this was due to both the new manufacturing process but also additional tweaks made to ensure signal coherency. In our testing, we actually saw an average L2 latency of 10.4 cycles, down from 16.9 cycles in on the Ryzen 7 1800X.
The L3 difference is a little unexpected: AMD stated a 16% better latency: 11.0 ns to 9.2 ns. We saw a change from 10.7 ns to 8.1 ns, which was a drop from 39 cycles to 30 cycles.
Of course, we could not go without comparing AMD to Intel. This is where it got very interesting. Now the cache configurations between the Ryzen 7 2700X and Core i7-8700K are different:
| CPU Cache uArch Comparison | ||
| AMD Zen (Ryzen 1000) Zen+ (Ryzen 2000) |
Intel Kaby Lake (Core 7000) Coffee Lake (Core 8000) |
|
| L1-I Size | 64 KB/core | 32 KB/core |
| L1-I Assoc | 4-way | 8-way |
| L1-D Size | 32 KB/core | 32 KB/core |
| L1-D Assoc | 8-way | 8-way |
| L2 Size | 512 KB/core | 256 KB/core |
| L2 Assoc | 8-way | 4-way |
| L3 Size | 8 MB/CCX (2 MB/core) |
2 MB/core |
| L3 Assoc | 16-way | 16-way |
| L3 Type | Victim | Write-back |
AMD has a larger L2 cache, however the AMD L3 cache is a non-inclusive victim cache, which means it cannot be pre-fetched into unlike the Intel L3 cache.
This was an unexpected result, but we can see clearly that AMD has a latency timing advantage across the L2 and L3 caches. There is a sizable difference in DRAM, however the core performance metrics are here in the lower caches.
We can expand this out to include the three AMD chips, as well as Intel’s Coffee Lake and Kaby Lake cores.
This is a graph using cycles rather than timing latency: Intel has a small L1 advantage, however the larger L2 caches in AMD’s Zen designs mean that Intel has to hit the higher latency L3 earlier. Intel makes quick work of DRAM cycle latency however.
Facebook
Twitter
RSS
545 Comments
View All Comments
danjw - Friday, April 20, 2018 - link
They Anandtech used the rated speeds that the processors were stated to support by the manufactures. Anandtech, is using everything at stock. Anandtech ran all the processors through fully patched systems (both bios and OS). Not every website other tests to these same methodology. So, there will be differences in their results. None the less, Anandtech, is auditing their results to double check them. I really don't think they are going to see anything wrong. Toms, ran their Intel parts without the latest bios updates. Others overclocked their systems.Most users do not overclock their systems. Sure, a lot of us readers do, but not everyone. I overclock my systems, but, my two brothers who are both just as technical as I am, do not. It is a choice some make and others do not. The majority of users do not overclock. So, Anandtech does not overclock in their most reviews. They have at times in the past and may in the future include overclocking results in reviews, but they have are always broken out the overclocking results in a separate section and/or labeled the overclocked results to differentiate them from the standard clocked results. These are editorial choices that Anandtech makes, I don't see any problem with that.
Luckz - Monday, April 23, 2018 - link
Intel for some reason have 4 memory sticks. Weird idea.werpu - Friday, April 20, 2018 - link
Well the main difference is they tested against fully meltdown and specte patched systems, which in fact is the norm, while all other reviewers simply tested against bare metal. It is known that Intel took a pretty serious hit especially with Meltdown and a more serious hit with Spectre compared to AMD which did not have meltdown at all and to a lesser degree Spectre than Intel did.I would say Anandtechs tests are spot on.
And this reflects the sad state of nowadays performance testing which seems to be done to 99.9% by incompetent idiots or fanboys (especially the youtubers are the worst)
However in extreme situations Intel again wins since the 8700k can be oced by decapping and good cooling to 5GHz while the OC capabilities of the 2700x are basically non existent. It really depends, which is better. But the performance gap is closing and in non OCed system it is not existent anymore. It will be interesting next year when AMD has moved to 7nm while Intel still will be stuck at 10nm which they currently try to pull it but not have yet managed. Then the game might be entirely reversed.
Alphasoldier - Friday, April 20, 2018 - link
Unfortunatelly, you are the only idiot and fanboy here. Pretty much everyone stated in their reviews, the system were fully patched, all cpus were reused and everything was retested, because AMD fanboys were screaming Meltdown here, Spectre there.Now, the internet is full of this garbage review, it spreads like cancer, because AMD fanboys have nothing better to do, once again they are disappointed that 6 cores from Intel outperformed 8 cores from AMD and they are now like the Liverpool fans repeating "The next year will be ours"
But at least they got some fancy RBG cooler.
Fallen Kell - Friday, April 20, 2018 - link
Alphasoldier, I've been reading the reviews, and while many have stated they have applied the software (OS) patches, very few have stated they applied both the software and BIOS patches for the Spectre variant 2. Thew few places that I have seen which have stated both the software and BIOS patches were applied all seem to be showing much more similar results as the AT article.In anycase, Ryan stated they are looking into it, and I am certain we will see an update within the next few days. And don't come saying that I am a AMD fanboi, I havn't purchased a AMD CPU since the Thunderbird (i.e. a slot A CPU).
mapesdhs - Saturday, April 21, 2018 - link
werpu, oc an 8700K to 5GHz? Makes me laugh that a 300MHz bump over a CPU's max single core turbo is even called an oc these days. Sheesh, it's a far cry from the days of SB, oc hardly seems worth bothering with now.mkaibear - Thursday, April 19, 2018 - link
It's here, it's here!Dr. Swag - Thursday, April 19, 2018 - link
What is with the gaming benchmarks? On your tests the whole ryzen 2 series is a step above everything else, but all other reviews show it between ryzen and coffee lake...fallaha56 - Thursday, April 19, 2018 - link
This is the Spectre2 patch effectNot looking great for Intel and HFR gaming
Ryan Smith - Thursday, April 19, 2018 - link
"What is with the gaming benchmarks?"We're looking into it right now. Some of these results weren't in until very recently, so we're going back and doing some additional validation and logging to see if we can get to the bottom of this.