AMD Ryzen 9 5900X vs. Intel i9-10900K: The Ultimate CPU Battle

For our first assessment of Ryzen 5000, we checked out the Ryzen 9 5950X, AMD’s new $800 16-core, 32-thread AM4 processor. In the present day, we’re wanting on the Ryzen 9 5900X, the 12-core, 24-thread CPU that is accessible for $550. That makes the 5900X ~8% cheaper per core when in comparison with its larger brother as AMD mannequin given it faces no actual competitors. The 12-core R9 5900X is priced to compete with Intel’s Core i9-10900Okay. Usually, the 10-core Intel processor would retail for $530, however, because of providing points, it sometimes sells for $550 or $600, which suggests the 10900Okay and 5900X are comparable by way of value.

Now we have to learn how they stack up by way of efficiency.

The 5950X is a beast, so we’re anticipating good issues from the 5900X which brings down the core rely upon as the one main change. The clock speeds are a lot identical, the increased frequency has been lowered by 100 MHz, whereas the bottom frequency elevated by 300 MHz. As a result of each CCDs are nonetheless lively, it packs a complete L3 cache of 64 MB. As soon as once more, we’re not going to concern ourselves with Zen’s three architectural enhancements on this assessment, slightly we’ll skip that and soar into the benchmarks.

Earlier than we do although, let’s take a fast have a look at one more new AM4 motherboard. In yesterday’s assessment, we showcased the Asus ROG Crosshair VIII Darkish Hero, and now we’re getting our first have a look at the MSI B550 Unify, a brand new B550 board that featuresproperly, just about every little thing. The highlights embody a 16-phase VRM utilizing Infineon TDA21490 90An influence phases, a 6-layer PCB with 2oz thickened copper, all-aluminum prolonged heatsinks with devoted VRM Heatpipe, 4 M.2 slots with double-sided M.2 Defend Frozr cooling, and DDR4 A-XMP assist for as much as DDR4-5800.

There’s additionally a Unify-X model with simply two reminiscence DIMMs. MSI claims the Unify-X can do DDR4-5800 with 4000 G collection processors and DDR4-5300 with Third-gen Ryzen. Whereas the usual Unify that we’ve can obtain DDR4-5600 with 4000 G-series although they’ve to make any Ryzen 5000 collection claims. Later this month we’ll put collectively an in-depth Zen three reminiscence efficiency information and we’ll use the usual Unify for the majority of our testing. This MSI board is priced at $290, so it’s a Gigabyte B550 Grasp competitor. We’ll return to extra motherboard testing after our slew of Ryzen opinions.

To maintain issues easy for all our new Ryzen 5000 testing, we’re utilizing the identical MSI X570 Godlike motherboard together with 4 8GB G.Ability TridentZ DDR4-3200 CL14 reminiscence modules for a 32GB capability. Cooling all methods is the Corsair iCUE H150i Elite Capellix AIO. For all of the productiveness testing, we’re utilizing an RTX 2080 Ti, however for the gaming benchmarks we’ve up to date all our numbers with the RTX 3090, so let’s get into the graphs

First up we’ve Cinebench R20 and wow, 8487 factors are fairly insane. That’s a 17% improvement from the 3900X, making the 5900X simply 8% slower than the 3950X. These are spectacular numbers contemplating the CPU packs 25% fewer cores. When in comparison with the Core i9-10900Okay which is a tough match in value, the brand new 5900X is available in 32% sooner, though it does have 20% extra cores.

Operating the single-core benchmark reveals why the 5900X is a lot sooner than the 10900Okay in Cinebench R20. Right here we’re taking a look at a 16% increase to single-thread efficiency. It’s additionally a 23% enhancement from the 3900X, and with an analogous single-core rating to that of the 5950X, we’re anticipating robust gaming efficiency. We additionally watched how the 5900X clocks in every of the Cinebench R20 checks. For the multi-core take a look at the place all cores are closely loaded, the 5900X clocks at round 4.three GHz. In the meantime, AMD advertises a base clock frequency of three.7 GHz.

Within the Cinebench single-core take a look at, the 5900X sometimes operated at 4.9 GHz, which is 100 MHz over the marketed spec. AMD on this occasion has made positive the max increase frequency is just not solely simply achieved, however could be exceeded to keep away from the drama seen with Zen 2 and its increased clocks. 7-zip appears at compression efficiency and right here the 5900X is 17% sooner when in comparison with its predecessor. It is also capable of edge out the 3950X, making it 37% sooner than the 10900Okay.

When decompressing we discover that SMT is extra helpful and in consequence, the 3950X is ready to pull forward. That mentioned, the 5900X is 8% slower and remains to be providing a considerable 18% efficiency increase over the 3900X. Because of this, it’s a lot sooner than Intel’s greatest, beating the 10900Okay by a whopping 43% margin. AES encryption efficiency is barely improved with Zen three and that is a space the place Zen 2 already excelled. Due to this fact, it’s no shock to see the 5900X dominating the 10900Okay by an enormous 52% margin.

Shifting on to Blender, right here we’re taking a look at a 12% efficiency increase for the 5900X over the 3900X and that meant it was 22% sooner than the 10900Okay. When in comparison with the 3950X it was 13% slower after which 18% slower than the 5950X. We discovered when testing the 5950X that the brand new Zen three structure completely shreds in V-Ray, and we’re seeing precisely that with the 12-core mannequin, too. Right here the 5900X mainly matched the older 16-core 3950X and that meant it was 23% sooner than its predecessor and 32% sooner than its main competitor, the 10900Okay, so a little bit of a shellacking there.

The final rendering benchmark we’ve is Corona, the place the 5900X is way sooner than the 3900X, boosting efficiency by 22%, delivering 3950X-like efficiency. That’s a giant gen-on-gen enchancment and it additionally meant the 5900X was 24% sooner than the 10900Okay. For these of you curious about code compilation efficiency we see that the 5900X is simply 9% sooner than the 3900X, so one of many smaller positive factors we’ve seen to date, in actual fact, it is perhaps the smallest. Regardless of that, it managed to beat the 10900Okay by a convincing 26% margin.

DaVinci Resolve Studio 16 utilizing the Puget Techniques Benchmark really offers the weakest positive factors we’ve seen but, the place the 5900X was simply 8% sooner than the 3900X, roughly matching the 3950X. Adobe Premiere Professional 2020 lowers the bar even additional, right here the 5900X was simply 5% sooner than the 3900X whereas it was simply 9% sooner than the 10900Okay. Nonetheless, a 9% efficiency benefit for roughly the identical cash is hardly something to sneeze at, definitely given it seems to be a worst-case situation.

The robust single-core efficiency turns out to be useful when testing with Photoshop, right here the 5900X matched the 5950X and that meant it was 10% sooner than the 10900Okay and a whopping 20% sooner than its predecessor, the 3900X. We’re seeing robust efficiency positive factors over Zen 2 in After Results, one other utility that depends closely on single-thread efficiency. Right here the 5900X was 15% sooner than the 3900X and 13% sooner than the 10900Okay, nearly matching the 5950X.

For measuring energy consumption we monitored whole system utilization in Blender. The Ryzen 9 5900X matched the ability utilization of the 3900X, despite the fact that they each use the identical 7nm course of, Zen three is ready to increase efficiency in Blender by 12%. Furthermore, when in comparison with 10900Okay, the 5900X lowered whole system consumption by 23%. By way of temperatures, the Ryzen 9 3900X and 5900X have been each examined in a 21C room on an open take a look at bench utilizing the Corsair iCUE H150i Elite Capellix AIO on the MSI X570 Godlike, put in contained in the Corsair Obsidian 500D.

We might see that the 5900X runs 7 to eight levels cooler than the 3900X below the identical take a look at situations, regardless of working at a barely larger clock velocity utilizing the identical 7nm course. A part of this might be all the way down to 7nm maturity, however, AMD has additionally included quite a few architectural optimizations. These enhancements additionally enable for a barely decrease Vcore which might have helped to cut back energy consumption and subsequently thermals.

We now have the identical battery of gaming checks as yesterday’s 5950X assessment utilizing a GeForce RTX 3090. Getting in the identical order, we’ll begin with Far Cry New Daybreak. The 5900X matched the 5950X precisely with 128 fps making it 6% slower than the 10900Okay, however, nonetheless an excellent general consequence gave the double-digit share enchancment over the 3900X. The 5900X additionally trails the 10900Okay in Rainbow Six Siege by a small 5% margin. We see the CPU matching the 5950X to offer an 8% increase over the 3900X. This was sufficient to edge out a few of the decreased core rely upon Intel processors such because the 10700Okay.

In Watch Canine: Legion we see robust efficiency from the 5900X, putting it on par with the 10700Okay and simply 2 fps down on the 10900Okay. That’s additionally a slightly massive 15% increase over the 3900X, which struggled towards even the Core i5-10400 at this low decision. We’re additionally seeing huge efficiency positive factors in F1 2020 because the 5900X is 21% sooner than the 3900X, hitting 272 fps with the RTX 3090. That’s an ideal consequence which has the brand new 12-core Ryzen processor a mere 2% behind the 10900Okay. That is shut sufficient to deem a attract our ebook.

The efficiency uplift in Horizon Zero Daybreak is stable. Right here the 5900X is 15% sooner than the 3900X and crucially is ready to match the 10900Okay, edging it out by a number of frames for what’s mainly the identical efficiency. Efficiency in Borderlands three is just barely improved. The 5900X was simply 5% sooner than the 3900X, although that was sufficient to get it inside vary of the 10900Okay, dropping by only a 4% margin. As seen earlier than within the 5950X assessment, these Zen three processors are beasts in Dying Stranding and the 5900X matched the 16-core mannequin precisely, making it 12% sooner than the 10900Okay and extremely, 35% sooner than the 3900X.

The 5900X can also be capable of match the 5950X in Shadow of the Tomb Raider, which means it is 4% sooner than the 10900Okay. A small win that is shut sufficient to name a tie. The brand new processor is thrashing the previous 3900X by a 26% margin on this title. Hitman 2 efficiency can also be stable. Right here the 5900X couldn’t match the 5950X however was nonetheless 7% sooner than the 10900Okay and 34% sooner than the 3900X. Efficiency in Star Wars: Squadrons is powerful, matching the 10900Okay, although edging it out convincingly for 1% low efficiency. Then we see a 27% efficiency increase over the 3900X, taking the typical body charge from 247 fps to 314 fps. The Critical Sam Four outcomes look good for Ryzen. The 5900X is 11% sooner than the 10900Okay and a staggering 47% sooner than the 3900X.

Right here’s a have a look at the typical efficiency throughout the 11 video games we examined. For probably the most half, the 5900X matches the 5950X in gaming, so it is no shock to see them each ship the identical 214 fps common. This additionally signifies that generally, the 5900X and 10900Okay are on par in relation to gaming efficiency, a minimum of in our small pattern of video games. On that notice, we’re wanting ahead to a 40 recreation head-to-head benchmark with these processors. It is also spectacular to see a 20% efficiency improvement over the 3900X. We all know AMD claimed a 19% IPC enhancement, however seeing that translate into real-world positive factors in video games is one thing else.

Jason Kale