11th Gen Intel Rocket Lake-S gaming processors didn’t impress us



[ad_1]

i9-11900K version hero image
Enlarge / Our test bench is a bit nastier than usual, due to Asus’ decision to bury the processor socket in the surrounding high-rise heat sinks, with the system RAM closing just as tightly from the bottom. . (Note the pair of empty DIMM slots.)

Jim salter

Today marks the start of retail availability of the Intel 2021 line of game processors, named Rocket Lake-S. Rocket Lake-S is still stuck on Intel’s venerable 14nm process – we’ve long lost count of the number of benefits to be tacked on the end – with features backported from new 10nm designs.

Clock speed on Rocket Lake-S remains high, but thread count has decreased at the high end. Overall, most benchmarks show that Rocket Lake-S underperformed last year’s Comet Lake – not to mention its real competition, coming from AMD Ryzen processors.

Our hands-on test results didn’t seem to match Intel’s marketing claims of up to 19% improvement in IPC (Instructions Per Clock Cycle) from generation to generation over its 10th generation components. .

Multithreaded processor performance

It shouldn’t be a huge surprise that the Core i9-11900K underperformed last year’s Core i9-10900K in multithreaded tests – this year’s model only offers eight cores compared to that of the last year. If that part had shown a 19% improvement in CPI, it could have been washed away – 20% fewer cores, offset by a 19% higher CPI.

What we have really seen is a significant decrease in performance in all areas. The Cinebench R20 and Passmark scores are each down about 10%. Geekbench 5, which tends to minimize the impact of a higher thread count, shows an alarming 27% drop in performance from generation to generation.

This year’s Core i5 performs much better than its big brother Core i9. In Cinebench R20, the i5-11600K almost catches up with Ryzen 5 5600X and easily outshines last year’s Comet Lake i5 equivalent. It doesn’t come very close to its competitor Ryzen in Passmark or Geekbench multithreaded tests, but it completely outperforms last year’s model.

This represents the clearest and most significant generation-to-generation victory we’ve seen from Rocket Lake-S.

Single-threaded processor performance

The Core i9-11900K manages to set new single-thread performance records – if by extremely fine margins, compared to Ryzen 9. Compared to last year’s Comet Lake i9, those margins are much less slim.

As always, we advise readers not to overdo it. Few real workloads are truly single-threaded, and even the larger margins shown here are pretty slim.

The Core i5-11600K sits a hair’s breadth away from the Ryzen 5 5600X on single-thread benchmarks, and it does so while securing somewhat bigger wins compared to last year’s Core i5-10600K. For the most part, this is the same relationship we saw between Rocket Lake-S, Zen 3, and Comet Lake in previous Core i9 vs. Ryzen 9 slides.

Gaming performance

To our surprise, 3DMark gaming benchmark Time Spy frowned upon our Rocket Lake-S samples – and did so disproportionately compared to the results we saw from the more versatile Cinebench R20, Passmark, and Geekbench 5 benchmarks. Both Core i9-11900K and Core i5 -11600K came with heavy Time Spy drawbacks compared to their 2020 10th gen counterparts.

Before receiving as well alarmed, this is just the Time Spy CPU test – it focuses on modeling physics in a way that doesn’t lend itself to GPU rendering. To assume that typical games see similar decreases in overall frame rate would be a serious mistake. Time Spy Extreme, which allows for more advanced CPU optimizations and higher thread count, exposes an even bigger gap. (We only tested Time Spy Extreme on the Core i9.)

As we noted in the Gaming-focused Spring 2021 System Guide, the CPU has minimal impact on AAA games – the frame rate in most games is almost entirely determined by the GPU. If the alone what you have opened on your pc is the game, you just need to meet a minimum processor threshold – roughly, “i5 from the 2015 era or better”. However, the more additional tasks you load on your system – browser tabs and email clients open in the background or, worst of all, catching streams of your own games – the more CPU power you’ll need. to get things going.

We didn’t see any change in the overall Time Spy score between 10th and 11th gen processors, and we didn’t expect to. Yes, single-thread performance has increased dramatically; no, it usually doesn’t have a big impact on gaming performance.

Integrated GPU performance

Rocket Lake-S is getting an upgrade to its integrated graphics, but if you were hoping the UHD 750 would play in the same league as Iris Xe and Vega 11, you're out of luck.
Enlarge / Rocket Lake-S is getting an upgrade to its integrated graphics, but if you were hoping for the UHD 750 to play in the same league as Iris Xe and Vega 11, you’re out of luck.

Jim salter

Rocket Lake-S gets a slight but noticeable upgrade to its onboard graphics performance: UHD 630 graphics from the 10th Gen Core processor have upgraded to UHD 750. While this is an improvement, there is is nothing serious – if you were hoping for an equivalent to Intel’s Iris Xe graphics in Tiger Lake notebook processors (or AMD’s Vega 11 in desktop APUs), you will be sorely disappointed.

A modest GeForce GTX 1060 is good for a Time Spy Graphics score of around 4000. Intel’s flagship i7-1185G7 laptop processor manages almost half that at 1572, with AMD’s Vega 11 being noticeably at the bottom. lags behind at 1226. The Rocket Lake-S UHD 750 comes in at a yawn of 592, just under half the performance of the Vega 11 and just over a third of Iris Xe’s performance.

While UHD graphics are sorely disappointing, we have to point out that they at least do. to exist. Competitive Ryzen processors in the retail channel offer no integrated graphics – and with the current supply shortages and horrific price hikes of discrete GPUs, that can mean paying an unpleasant premium in full-system versions that are not supposed to focus on the game in the first place.

[ad_2]
Source link