Here's our first look at the new Ryzen 5000 series of CPUs, starting with a full review of the flagship Ryzen 9 5950X. As a direct replacement for the R9 3950X, the new chip offers 16 cores with SMT support for 32 threads. The suggested retail price has been increased from $ 750 to $ 800. This is an expensive CPU for the mainstream AM4 platform.
Of course, we also have the Ryzen 9 5900X, Ryzen 7 5800X, and Ryzen 5 5600X on hand, and you'll be seeing reviews of these parts in the next few days. We think these new AMD processors represent a significant leap forward as each one deserves its own review.
AMD justifies the price increase across the series with an allegedly dramatic improvement in performance, especially when it comes to gaming. The highlights include boost clocks with up to 4.9 GHz and an IPC improvement of 19% thanks to a new design in which core complexes are expanded to 8 cores with a new 32 MB L3 cache topology. This was achieved using the same TSMC 7nm process.
There are a number of architectural improvements, some of which were discussed at the announcement and we could only discuss the new CPU features on paper. But instead of going through all of that, we're going to go straight to what we suspect you're really here for: the benchmarks.
Now let's go through our test setup and jump into all of the glorious blue bar charts. We used the god-like MSI X570 motherboard to test the AMD CPUs, also because it was one of four X570 motherboards that received BIOS support from AMD. Although other AM4 boards had Zen 3 support prior to release, AMD limited reviewers to four models to expedite troubleshooting issues should a problem arise.
It's quite an accomplishment that you can run Zen, Zen +, Zen 2, and Zen 3 CPUs on one motherboard, although official first-generation support isn't listed. However, most X570 cards support parts of the Ryzen 1000 series. In the case of the X570 Godlike, we could either use the Ryzen 5 1600 or the new Ryzen 9 5950X, and both worked perfectly, leaving four generations of CPUs on a single motherboard.
To test Intel processors for the same period of time, you need three separate test systems. In any case, we have equipped all AMD and Intel systems with four 8 GB G.Skill TridentZ DDR4-3200 CL14 memory modules for a capacity of 32 GB. We cooled them with the iCUE H150i Elite Capellix AIO from Corsair, as most of these CPUs don't have a box cooler.
As a side note, we recently received the new Asus ROG Crosshair VIII Dark Hero motherboard which is a modified version of the original. We recommend paying $ 400 for the Wi-Fi 6 version. Why is that relevant? This appears to be a very powerful board that has further refined from the original, removing the X570 chipset fan and upgrading the VRM to 16 90A TI power levels. This appears to be a high-end X570 motherboard that should easily match the extreme $ 700 models like the MSI X570 Godlike in terms of VRM performance, but at a cheaper price point.
Finally, note any productivity checks we run on the GeForce RTX 2080 Ti. For the gaming benchmarks, however, we've updated all of our numbers with the RTX 3090, adding a few new games in the process. So let's get to the graphics …
First up, we have Cinebench R20 with the Ryzen 9 5950X, which is the first desktop CPU on a mainstream platform to instantly break the 10,000 point mark. We expect a performance increase of 10% compared to the 3950X.
Compared to Intel & # 39; s best mainstream offering, the 10900K, we're talking about an almost 60% increase in performance, although AMD also offers 60% more cores here.
The real showstopper, however, is the single-core performance where the new R9 5950X is 20% faster than the 3950X and 17% faster than the 10900K. This type of single-threaded performance is a good sign of gaming performance.
We also looked at how the 5950X performed in each of the Cinebench R20 tests. For the multi-core test, in which all cores are heavily loaded, the 5950X clocked at around 3.95 GHz, although we saw 4.2 GHz in Blender, as this number depends on the workload. AMD announces a base clock frequency of 3.4 GHz.
AMD also announces a maximum boost clock frequency of 4.9 GHz. This should be achieved on workloads with a core or light thread. In the Cinebench single core test, the 5950X typically ran at 5 GHz or 5050 MHz to be precise. That is 150 MHz above the specified specification.
The 5950X is only 8% faster than the 3950X in the 7-Zip compression test, although this remains a nice improvement in generation performance. Note, however, that the new chip also costs 7% more, which means we don't see any improvement in value for money.
The decompression edges are a little more impressive. This time around, the 5950X is 14% faster than the 3950X and almost 80% faster than the 10900K. Based on current retail prices, the 5950X should only cost ~ 35% more.
Ryzen processors largely dominate AES encryption performance, and the SiSoft Sandra benchmark is one of the more balanced tests. The Zen 3 architecture offers little improvement over Zen 2, but the 5950X was 71% faster than the 10900K.
With Blender we only see a small improvement for the 5950X over the 3950X, 6% faster. The AMD flagship is almost 50% faster than the 10900K, which makes it the cheaper part.
Interestingly, we see a huge increase in performance in the V-Ray benchmark. Here the 5950X is 20% faster than the 3950X and 62% faster than the Core i9-10900K. This is exciting news for those who use V-Ray and don't want to spend huge bucks on a Threadripper system.
The 5950X also offers great performance increases in Corona and outperforms the 3950X by 20% and around 60% faster than the 10900K.
Here's a look at code compilation performance, where we see a slight improvement for the 5950X that is 7% faster than the 3950X. It was also 50% faster than the 10900K.
The DaVinci Resolve Studio 16 benchmark is more dependent on GPU performance. As you can see here, the CPU can still make a difference. We see a small increase of 4% with the 5950X over the 3950X.
The performance increase in Premiere Pro is also small compared to the 3950X. We only expect a 5% improvement in performance and a slight 15% increase over the 10900K.
In contrast to video editing software, Adobe Photoshop mainly relies on the performance of individual threads. As a result, the 5950X can achieve massive performance gains over the 3950X with an 18% increase in performance. The latest Ryzen 9 was 10% faster than the 10900K too, so pretty impressive stuff.
After Effects is another application that is heavily reliant on single core performance. We see a strong performance advantage with the 5950X, which is 20% faster than the 3950X and 15% faster than the 10900K.
We took a quick look at the power usage and were impressed with what we found. The 5950X is manufactured using the same 7nm process as the 3950X. However, AMD engineers were able to improve performance but also reduce power consumption to greatly improve efficiency.
Despite a 6% increase in performance, we are also striving to reduce power consumption for the entire system by 5%. Compared to Intel's 10-core Core i9-10900K, we see a 25% reduction in power consumption.
A quick look at temperatures and performance compared to the Ryzen 9 3950X and 5950X. Both were tested in a 21C room on an open test bench with the Corsair iCUE H150i AIO and MSI X570 Godlike installed in the Corsair Obsidian 500D.
The Ryzen 9 5950X ran 6 to 7 degrees cooler than the 3950X under the same test conditions, even though it ran with the same 7 nm process at a slightly higher clock rate. Part of that will be down to 7nm runtime, but AMD also claims to optimize the architecture. These improvements allow for a slightly lower Vcore, which in turn would have helped lower power consumption and thus lower thermal levels.
Time for the gaming tests where we used the fastest GPU currently available, the GeForce RTX 3090. Starting with Far Cry New Dawn at 1080p, we find a 10% increase in performance for the 5950X over the 3950X. Not a bad result, but that puts the 5950X around 6% behind the 10900K. They're identical to the 2080 Ti, but this next-gen GPU is pushing performance back in favor of Intel in that title.
If we move on to Tom Clancy's Rainbow Six Siege, we see a 7% performance improvement for the 5950X over the 3950X, again not bad, but not all that amazing either. That said, the new 16-core CPU was still 6% behind the 10900K, which will undoubtedly disappoint AMD fans.
Watch Dogs: Legion is the newest game that was added to our benchmarks. This game shows a 14% increase in performance for the R9 5950X over its predecessor. That made it only 2% slower than the 10900K, so this time it was very close, even though the Intel CPU is cheaper than the flagship from AMD.
The frame rate performance in F1 2020 is also close to between 10900K and 5950X. AMD is 2% behind the RTX 3090 and can pump out 272 fps. It's worth noting that we expect an impressive 21% increase in performance over the 3950X.
In Horizon Zero Dawn, the 5950X displaced the 10900K, albeit by a small 2% lead, so it was more of a match. To get here, the new processor received a 14% increase in performance over the 3950X.
We're only seeing a 5% increase in performance for the 5950X over the 3950X in Borderlands 3, but that's enough to put the new 16-core CPU just a stone's throw away from the 10900K. In the end, it was 4% slower than the 10900K, which is as close as possible before it is called a tie.
Intel had a decent performance advantage in Death Stranding as the 10900K was nearly 20% faster than the 3950X.
The Ryzen 9 5950X is a whopping 34% faster than the 3950X and therefore 12% faster than the 10900K. That's a win for AMD.
The 5950X also pushed forward in Shadow of the Tomb Raider. Here the R9 5950X is 4% faster than the 10900K, so basically the same performance, but that's a 24% improvement over the 3950X.
Another game where AMD is taking a big generational leap is Hitman 2. Here we see a 34% increase, pushing the 5950X + RTX 3090 combo to an average of 167 fps. That is 10% faster than the 10900K.
Another game where the 5950X performs very well is Star Wars Squadrons, which got the RTX 3090 averaging 312 fps, putting it on par with the 10900K, even though the 1% low performance was 9% stronger. Compared to the 3950X, we expect a performance increase of 26%.
We had to retest Serious Sam 4 three times to make sure everything was correct. Here the Ryzen 9 5950X was 13% faster than the 10900K and an insane 48% faster than the 3950X, which is a hell of a leap in performance from one generation to the next.
Average game performance
Here's a look at the performance averaged over the 11-game example. As you can see, the Ryzen 9 5950X and 10900K are a match in terms of performance. This also means that the 5950X was 19% faster on average than the 3950X, which is an impressive generational leap for gaming workloads.
One could also argue that Intel remains pretty competitive for gaming. Even if that means it uses more power and does poorly in productivity applications, the Core i9-10900K can hold its own against the best from AMD when it comes to gaming.
Now let's look at overclocking performance. With the 5950X, we were able to achieve an all-core overclocking of 4.7 GHz with 1.375 V. This overclocking boosted performance in Cinebench R20 by 20%, which is a very significant increase in performance for core-heavy workloads.
However, the same overclocking is nowhere near ideal for single-core workloads like the 5950X standard clocks up to 5 GHz. So we reduced the peak operating frequency by 6% and lowered the single-core power by 5%.
Back to a high core workload, the increased all-core frequency is of great benefit. Blender shows an 18% increase in performance, which is remarkable.
The disadvantage of overclocking is of course the increased power consumption and thus the heat output.
The 4.7 GHz overclocking increased the overall system utilization by 40% as the R9 5950X now used around 140 watts more. It doesn't look too bad compared to the 10900K stock and was 74% faster.
Overclocking this CPU isn't particularly useful in games as we lose a bit of performance.
Oddly enough, the performance in Far Cry New Dawn has been slightly improved. We're only talking about a few extra frames, but from what we've seen in Rainbow Six Siege, we didn't expect such a result.
Here is our look at the price / performance data, first at productivity. The 5950X should land at $ 800 and that means it will cost $ 70 more when released than the 3950X. In terms of price and performance that put these CPUs on par, they are both cheaper than the 10900K, but slightly worse than other lower-core Intel parts like the 10600K and 10700K.
But when it comes to core-intensive productivity, they're of course in a completely different league.
Despite the strong single-core performance, the 5950X is slightly more expensive than the 3950X and 10900K. However, this is unlikely to be a problem for a high-performance CPU as the "time is money" mentality applies broadly here.
For gaming, the 5950X is a pretty poor prospect, as gaming typically doesn't use nearly half of the 5950X. However, for an additional $ 70 over its predecessor, this is better value for gaming than the 3950X.
Even if we average the performance of the 11 tested games and use this data for a price-performance comparison, we see that the 5950X outperforms the 3950X, but it scores worse than anything else. This comes as no surprise as games cannot take full advantage of a 16-core / 32-thread processor.
What we learned
A year ago, AMD put Intel in a difficult situation with the Ryzen 9 3950X. It slightly outperformed the Core i9-9900K in core-heavy workloads by packing twice as many cores. When it came to maximum FPS gaming performance at low resolution, it wasn't often that much. There were times when the Intel CPU was up to 20% faster, which is a significant margin. Therefore, we recommend the 9900K for those looking for maximum gaming performance with little interest in productivity performance.
Nonetheless, the 16-core 3950X has taken over the crown of mainstream desktop performance and also eliminated the Intel Cascade Lake-X HEDT platform with the help of 2nd and 3rd generation Threadripper parts. A year later we have an even better CPU, which Intel takes the performance crown off.
Of course, if gaming performance is your primary interest, the Ryzen 9 5950X is over the top. You can achieve a similar level of performance with the cheaper 5900X and even the 5800X. In most games, even the 5600X won't be far behind. We'll be posting detailed reviews of all of these chips in the next few days.
In other words, spending $ 800 on the Ryzen 9 5950X just for gaming is pointless, but a core-heavy processor no longer degrades performance, making it a viable option for gamers with deep pockets. Where this new 16-core processor really shines is for those who want to work and play, and right now there is no better option to tackle those two tasks.
We believe we are at the end of the road for the AM4 platform, so the 5950X investment loses some of its value as the upgrade path is practically dead. However, if you are looking for the best processor for less than $ 1,000, this is it.
For those of you who are already rocking a Ryzen 9 3950X, the R9 5950X isn't worth the upgrade. If you are playing with a high-end GPU, you are likely doing so at a high resolution where you are tied to a GPU most of the time. For example, at 1440p on an RTX 3090, there is very little difference between the 3950X and the 5950X, and at 4K we don't expect any difference. When it comes to productivity performance, the profits don't justify the investment.
However, if you come from a 2nd generation Ryzen processor, the upgrade for gaming and productivity is huge. For a seriously high productivity performance, the 5950X is a great option as long as you're willing to part with $ 800. With that in mind, it will be interesting to see what happens to pricing for the 3950X over the next few months as it may be the ultimate option for high-end CPU performance.
Of course, this all depends on the actual availability of the Ryzen 9 5950X, but we hear that we should expect a strong supply for new Ryzen 5000 series processors. It's always great to see AMD take another big step forward, and we look forward to reviewing the rest of the product stack. Expect our next Ryzen test tomorrow at the same time.