Intel Core i9-14900KS – stay away!
14/3/2024
Translation: Patrik Stainbrook
The KS version of Intel’s i9-14900 is the Lamborghini of processors: fast, useless off the racetrack and thirstier than a Dodge Charger.
The 14900KS is now the third special version of Intel’s flagship processor that I’ve tested. I realise in advance that my verdict will be the same – a CPU only for overclockers. This is all the more true for the latest iteration, as it demands a lot of power particularly quickly.
The chip isn’t a new model, but a particularly good example of an i9-14900KS that can be clocked higher. It uses chip binning, a process in which CPUs are sorted by quality. The processor already has a higher clock rate than its little brother out of the box. Base power draw has also increased by 25 watts to 150 watts. Here’s a comparison of the two CPUs with others:
Test setup and method
I use the following components for my test:
The system runs on Windows 11 in version 23H2 (22631.3296). I use BIOS version 2002 with XMP enabled. Otherwise, I leave everything on default with Resizable BAR is activated. For the graphics card, I use version 551.76 of the driver.
Here’s an overview of the benchmarks used:
- Cinebench R23
- 7-zip
- Blender Benchmark
- UL Procyon Photo Editing Benchmark
- UL Procyon Video Editing Benchmark
- 3DMark: Fire Strike, Fire Strike Ultra, Time Spy and Time Spy Extreme
- Games: Anno 1800, Forza Horizon 5, Assassin’s Creed: Valhalla, Civilization VI, Cyberpunk 2077, Far Cry 6, Gears of War 5, Red Dead Redemption 2, Shadow of the Tomb Raider
I run each benchmark three times and take the best result. For the games, I use the highest possible presets. Otherwise, I leave everything at default except for the resolution. I left DLSS, XeSS and FSR deactivated.
Results in productivity applications: worse than the 14900K
On the following slides, you can see a comparison of the mean values for the six productivity benchmark results, comparing it to the i9-14900K, the quasi-predecessor i9-13900KS as well as the competitor Ryzen 9 7950X3D. If you click through the gallery, you’ll see the results of the individual benchmarks.
title=What the benchmarks test; content=Cinebench R23 tests how a CPU performs when rendering 3D models. From these calculations, the benchmark creates scores for single and multi-core performance. The 7-zip benchmark tests a program’s hash calculation methods, compression and encryption codecs. The Blender Benchmark renders three scenes in the 3D graphics suite in version 3.3 and higher and calculates three scores. I add these together to get a final score. UL Procyon’s Photo Editing Benchmark and Video Editing Benchmark simulate various workloads in Adobe Creative Suite. At the end of the test, the benchmark calculates a score.;]]
The average value of the 7950X3D is somewhat misleading. The processor is namely behind the 14900KS in four of six applications. In the 7-zip benchmark, the new Intel processor is clearly defeated by its AMD rival. It’s also noticeable that the 14900KS lags behind older Intel chips.
This difference is easy to explain. The new 14900KS draws so much power so quickly that it soon reaches its thermal limit and has to clock down. On average, the efficiency cores clock at 4.3 gigahertz (GHz) in Cinebench R23, the performance cores at 5.4 GHz. With the 14900K, the efficiency cores achieve 100 megahertz (MHz) more – for the 13900KS, the performance cores are ahead by 100 MHz.
Nevertheless, the 14900KS can achieve better results in the single core test thanks to its clock rate up to 6.2 GHz. This is probably also the reason why it performs better in the Premiere benchmark, where often only one core is used.
Results in games: better than the in-house competition, weaker than AMD
The 3DMark benchmarks provide a first look at the results in the games. The graphic shows the mathematical mean of the four 3DMark benchmarks, followed by the individual results.
Things look better for the 14900KS when gaming. It can at least distance itself from its little brother and is almost on par with its predecessor. Compared to the AMD chip, however, Intel has no chance. The 14900KS loses points in the Time Spy Extreme benchmark especially. I already had the same outlier with the 14900K. I dismiss the poor performance, the processor can’t yet unleash its full power in that benchmark.
On the following slides, you can see the arithmetic mean of frames per second (fps) in the nine benchmark games compared to the other three processors. Starting with 1080p, followed by 1440p resolution. If you click through the gallery, you’ll see the results of the individual games in the same order.
I’m aware that the differences in fps get smaller at higher resolutions because the graphics card takes on more computing power. In my opinion, however, testing lower resolutions than 1080p makes little sense, as you’ll hardly ever buy a flagship CPU to play in 720p resolution. Testing higher resolutions than 1440p also makes little sense, since at 2160p resolution, the GPU becomes the bottleneck, not the CPU.
As expected, differences between the processors vanish at higher resolutions. While the i9-14900KS is still 1.5 per cent behind the front-runner 7950X3D at 1080p, it’s only one per cent behind at 1440p resolution. Compared to its little brother 14900K, it offers 1.5 per cent more fps at 1080p resolution. At 1440p, they’re on par. Overall, however, the differences are minuscule.
Same applies to the percentiles. The percentile values are usually frame times measured in milliseconds. These are the time intervals from image to image, or frame to frame. 99th percentile means that 99% of all data is faster than the value provided.
The 14900KS can still keep up with the other processors in the 99th percentile. In the 99.9th percentile, however, it falls back to last place.
Temperature and power consumption: horrible
Despite sporting efficiency cores, the i9 processors aren’t frugal. This applies double to the 14900KS. It draws over 300 watts in the Cinebench R23 benchmark, starting with 312 watts, heating up to 100 degrees Celsius and having to throttle down. During particularly intensive computational processes, the CPU reaches its thermal limit and is unable to reach its full potential.
However, Cinebench is an extreme case. Nevertheless, the processor also draws an astonishing amount of power when gaming and gets hotter than the other CPUs. Across the entire benchmark suite, the average is 183 watts at 1080p resolution. That’s over 100 watts more than the 14900K achieves. I can’t explain how this comes about, because the CPU achieves similar values to the 14900K when watching Netflix, browsing, doing office work and idling. Intel clearly still needs to optimise gaming. Here are the values in a table, as well as a comparison to the other three processors.
High performance requirements in gaming are reflected in the watts required per fps. The 14900KS performs miserably here. It requires over twice as much power as the 14900K and over four times as much as the 7950X per fps.
In a nutshell
A chip with few benefits
For this verdict, I’ll return to my opener – the Intel Core i9-14900KS is only for overclockers. However, you’ll need a very powerful cooling system to do this. My all-in-one water cooling system with a 320-millimetre radiator and three fans can’t cope with the CPU. I’d therefore go even further and say that the 14900KS is only for extreme overclockers. That’s users who resort to drastic cooling methods such as liquid nitrogen and like chasing records. Everyone else is better off keeping their hands off the CPU.
In my productivity benchmarks, the i9-14900KS performs even worse than its little brother, the i9-14900K. Still, at least the special edition beats the in-house competition when gaming in 1080p resolution. However, the price to do so is enormous: on average, the 14900KS requires 100 watts more than the 14900K. In addition, AMD is still ahead with the 7950X3D. If you only play games, I don’t recommend the 14900KS. You’re better off with the 7800X3D, which is significantly cheaper and achieves similar or even better values than the 7950X3D when gaming.
The Core i9-14900KS is the last of its kind – in two ways. Intel is using a new naming scheme for the next generation of CPUs. The «i» is disappearing, and this processor also marks the end of the LGA 1700 socket. The next generation – Meteor Lake, releasing for desktop sometime this year – will use the LGA 1851 socket. If you want to upgrade from the 14900KS in future, you’ll also have to replace the mainboard. For this reason alone, I don’t recommend the CPU.
Pro
- suitable for overclocking
- massive raw power
Contra
- for enthusiasts only
- expensive
- power hungry
- no upgrade options
- needs incredible cooling
Header image: Kevin Hofer
Kevin Hofer
Senior Editor
kevin.hofer@digitecgalaxus.chFrom big data to big brother, Cyborgs to Sci-Fi. All aspects of technology and society fascinate me.
