Intel Core Ultra 9 285K vs AMD Ryzen 7 9800X3D: In-Depth Gaming Performance and Benchmark Comparison

Evaluating how AMD's 3D V-Cache and Intel's architectural changes impact gaming performance, power efficiency, and overall value.

Hardware by Katmin on Jun 02, 2025

Many might think that comparing the AMD Ryzen 7 9800X3D to Intel's Core Ultra 9 285K is a pointless comparison, considering the 285K loss to its predecessor. But Intel literally marketed this as "made to game" and "ready for anything."

Previous 14th-gen chips weren't separate gaming processors—you should buy only for gaming—so this is their latest and greatest CPU. That's why it makes sense to compare it to the 9800X3D. I also wanted to see just how much the 9800X3D beats out the 285K and whether Intel has a chance at a comeback here.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Specifications

Quickly glossing over the specs: the 9800X3D comes with eight Zen 5 cores and 16 threads, a 5.2 GHz boost, 8 MB L2 cache, 96 MB of L3 cache, and a $479 price tag. The 285K, on the other hand, features 24 cores in an 8+16 configuration, a 5.7 GHz boost, 40 MB of L2 cache, 36 MB of L3 cache, and a significantly higher $630 price tag.

The key changes made to the 9800X3D include its Zen 5 cores and 4 nm process, but more importantly, its underside 3D V-cache architecture. Unlike previous generations, where the 3D V-cache was stacked on top of the main CCD, this time, it's implemented directly underneath the 8-core CCD, which improves temperatures—and thus overall voltages, clock speeds, and performance—in both gaming and other applications.

The 285K represents a more radical architectural shift, featuring two new core architectures: Lion Cove on the P-cores and Skymont on the E-cores. Other changes include the removal of Hyper-Threading and a new 3nm TSMC process on the compute tile, where these cores are located.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Test Bench Specifications

For the 9800X3D, I used an ROG Crosshair X870 Hero motherboard with G.Skill DDR5–6000 MT/s CL28 RAM. For the 285K, I used the ASUS Z890-P Wi-Fi and G.Skill DDR5–6400 MT/s CL32 RAM.

Yes, I'm using faster memory on Intel, and that's because, due to Intel's radical shift in architecture, they updated the memory controller on these new CPUs. Intel requires faster RAM to maximize CPU speed, so they recommend DDR5–6400 MT/s.

We're also using a Samsung 980 Pro 1 TB SSD, an Arctic Liquid Freezer II 420 AIO, and an RTX 470 GPU. The RTX 470 is the only GPU I have available for testing due to budget constraints and because I'm a small channel. To counter this, we first ran four CPU-bound games at 1080p and 720p, followed by four more GPU-bound games in bar charts against other CPUs.

Finally, we ran benchmarks, frequency, temperature, and power analysis. Both CPUs were set to their high-performance power plans: Ryzen High Performance on AMD and Intel Application Performance Optimizer (APO) on Intel. We also left the 285K on Intel's default performance setting rather than "Extreme," since that's the default.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Side-by-Side Comparison

Starting with Red Dead Redemption 2 at mostly medium settings with no antialiasing at 1080p, the 9800X3D led by an average of 10%. However, I noted that the 9800X3D had lackluster 0.1% lows, but overall, it didn't hamper the experience much. The 285K had 11% higher 1% lows, although it also had much lower GPU utilization, hinting at greater CPU limitation.

At 720p, the 9800X3D led by 19% on average. Once again, the 9800X3D's 1% lows were low, and the 285 K's 1% lows were 31% higher—this could be due to micro-stutters, as the 9800X3D's frame graph seemed less smooth than the 285 K's. However, it didn't significantly impact the experience.

In Far Cry 6, on the high preset with no antialiasing, the 9800X3D led by 16% on average, and this time, I overlooked micro-stutters on the 9800X3D, with its 1% lows being 19% higher than the 285 K's.

However, the utilization of the 285K didn't exceed 26%, which hints at optimization or a CPU limitation; it's difficult to say. At 720p, the 9800X3D led by 49% on average. The 285 K's GPU utilization barely exceeded 60%, and its clock speed fluctuated, impacting its 1% lows. The 9800X3D's 1% lows had a 46% lead, with clocks locked at 5.2 GHz.

In Counter Strike 2, with all low settings, no antialiasing or upscaling, the 9800X3D led by 7% on average, while its 1% lows led by 37%. Despite the meager uplift in averages, the 285K suffered micro-stutters when encountering smoke grenades or explosions, impacting its 1% lows.

It also had lower-than-expected CPU and GPU utilization, though it's hard to tell if this is optimization since games don't use all CPU cores, even in CPU-limited scenarios. At 720p, the 9800X3D led by 28% on average, while the 285 K's 1% lows suffered due to micro-stutters, and the 9800X3D's 1% lows led by 58%. The 285 K's GPU utilization again barely exceeded 60%.

In Age of Mythology at mostly high settings with no antialiasing or upscaling, the 9800X3D led by 53% on average at 1080p—an extremely CPU-limited scenario given the extensive physics, particle effects, and characters.

The 285 K's 1% lows tanked, with the 9800X3D leading by 42%. The 285 K's utilization never exceeded 18%. At 720p, the 9800X3D led by 52% on average, though both CPUs showed a mere FPS increase—suggesting the game was limiting CPU utilization. Even so, the 9800X3D's 1% lows jumped to a 59% lead.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Graph Comparisons

Jumping into more GPU-limited games: in Cyberpunk 2077 at 1080p with the high preset, we saw a 5% increase on the 9800X3D as we became more GPU-bound, with lows within the margin of error.

In Forza Horizon 5 at high preset, averages were within the margin of error, but the 9800X3D's 1% lows were led by 3%, though we were heavily GPU-limited. In Rainbow Six Siege at high presets with no antialiasing, for the first time, the 285K led over the 9800X3D by 2%, and its 1% lows led by 10%. However, given that the 9800X3D outperforms its predecessor, this raises questions about the 285 K's position.

In Shadow of the Tomb Raider at high preset with no antialiasing, the 9800X3D led by 19% on average and 30% on 1% lows. Our overall average at 1080p shows the 9800X3D leading by 4% in more GPU-limited scenarios.

At 720p, Cyberpunk shows the 9800X3D leading by 4% but losing to the 285K by 11% in 1% lows, again due to a slight GPU limitation. Forza Horizon 5 shows the 9800X3D leading by 3%, with 6% higher than 1% lows. In Rainbow Six Siege, the 9800X3D led by 1% on average and 2% on 1% lows. Finally, Shadow of the Tomb Raider shows the 9800X3D leading by 51% on average and 8% on 1% lows. Our overall average at 720p shows the 9800X3D leading by 19%.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Geomean Summary (1080p and 720p)

Looking at performance-per-dollar or value, the 9800X3D had 46% more value, which surprises no one, given it's both less expensive and higher performance. Performance-per-watt, or efficiency, shows the 9800X3D leading by 33%. Overall power consumption shows that the 285K consumes 19% more, although the 9800X3D wasn't exactly frugal.

Benchmarks

In productivity benchmarks, the 9800X3D surprisingly led in Photoshop by 24%, a result that might raise some eyebrows. However, it lost to the 285K in both Premiere Pro and After Effects by 20% and 5%, respectively—expected since the 285K has many more cores that both Premiere and After Effects can leverage. It also takes advantage of Intel Quick Sync and QSV, especially in Premiere Pro.

I found Photoshop's result surprising on the 9800X3D.In Blender, the 285K led by 72%, which is no surprise given its higher core count. In Cinebench multicore, it also led by 78% over the 9800X3D.

However, the 285K also led single-core by 14%, suggesting its single-threaded performance is strong, and the lower-than-expected CPU and GPU utilization we saw could be due to optimization.

Regardless, AMD's inherent advantage with increased cache, thanks to 3D V-cache, might still give it the edge in gaming, even if Intel addresses optimization issues. In 7-Zip compression, the 9800X3D came surprisingly close: the 285K still led by 37% in decompression and 30% in compression, which was surprising.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Clocks, Power, and Thermal Analysis

In Cinebench's 10-minute throttle test, the 9800X3D stayed stable at around 5 GHz on average, rarely dropping below 4.8 GHz. The 285K was widely inconsistent, averaging 5.05 GHz but dropping as low as 2.4 GHz. The E-cores averaged 4.2 GHz, falling to 3.2 GHz. Although not stable, they were more consistent than the P-cores, but it was disappointing—even on a 420 mm radiator —to see that inconsistency.

In terms of power, the 9800X3D averaged 154 W, peaking at 159 W, while the 285K averaged 223 W, peaking at 244 W, as expected. The 9800X3D reached an average of 86 °C, peaking at 89 °C—higher than expected, given my 420 mm radiator—while the 285K averaged 90 °C and peaked at 97 °C. Considering that clocks remained stable, I can't fault the temperatures too much.

In gaming, clock speeds in Cyberpunk 2077 at high preset 1080p had the 9800X3D averaging around 5.2 GHz, staying near that margin gracefully. The 285 K's P-cores were more stable compared to Cinebench, averaging closer to 5.4 GHz but dropping as low as 5 GHz. The E-cores maintained a frequency of 4.6 GHz almost constantly.

Overall, the 9800X3D maintained a strong lead over the 285K in CPU-limited scenarios, but in GPU-limited scenarios, such as Cyberpunk, the 285K wasn't far behind. In games like Forza Horizon 5 (GPU-limited), the 285K nearly matched.

However, in Red Dead Redemption 2, the 285K demonstrated better frame pacing with higher 1% lows despite losing in averages, an outlier. Dropping to 720p revealed the 285K struggled, being absolutely demolished by the 9800X3D with micro-stutters and limiting 1% lows.

Intel , Core Ultra 9 285k, Ryzen 7 9800X3D, Gaming Performance, Benchmark, NoobFeed

Final Takeaways

Overall, the 285K dominates in productivity, which is no surprise given its core count. The 9800X3D, however, showed a surprising lead in Photoshop. The possibility of optimization issues on the 285K is slightly reinforced by the clock speed inconsistencies we observed, though lower utilization overall might be the more likely cause.

Its single-threaded performance in Cinebench illustrates strong per-core performance, but this didn't translate to gaming. If Intel ever resolves optimization problems—through microcode, software scheduling, etc.—AMD might still retain the gaming lead due to cache advantage. However, this also opens up the possibility that Intel could improve with updates to address utilization, giving hope to the gaming community.

But what do you think? Does Intel have a chance in gaming, or will AMD continue to demolish Intel? Your thoughts and opinions matter, so let me know in the comments. That's all for today.

Also, check our other hardware articles: