Intel Core i5-10600(K) vs. Ryzen 5 3600: Best value?
The same old question every PC builder & gamer has to ask themselves: Which CPU offers better performance and which CPU offers the better price-performance ratio?
This article is about the two K-models Core i9-10900K and Core i5-10600K from Intel vs the Ryzen 5 3600.
The CPU is a core part of your PC and can have a huge impact on performance, good and bad.
If you have a high-performance graphics card (GPU) in your PC a weak CPU can bottleneck your GPU and therefore your gaming performance.
If that is the case with eiter the Intel i5 10600K or the Ryzen 5 3600 CPU will be topic of this article.
Test Results: Intel Core i5-10600(K) vs. Ryzen 5 3600: Best value?
Ranking First: Intel Core i5-10600(K)
- Top-notch gaming frame rates for a midrange processor
- Good thermal performance
- 6 cores with HyperThreading
- Expensive for core/thread count relative to Ryzen 5s
Best performing CPU
With Comet Lake-S, Santa Clara is now launching ten cores and up to 5.3 GHz with the top model. Intel Core i-10000, that means 32 different models, up to DDR4-2933 support (only for Core i9-10900 [K/F] and Core i7-10700 [K/F]), core variations of two to ten computing hearts including HT (SMT), a new maximum limit of 125 watts of Thermal Design Power (TDP) and a new Boost Technology, which is reserved for the top model, the Core i9-10900 (K/F).
However, the CPUs are still manufactured in 14 nm and only feature PCI-Express 3.0. As (almost) usual with Intel, the new CPUs require a new CPU socket and thus dewy motherboards.
Techtestreport will inform you here about all the innovations regarding the platform for socket 1200.
Since Comet Lake-S is a completely new platform, we have extended our test system and adapted it for the Intel Core i9-10900K and Core i5-10600K:
Newly added is the mainboard, we use an Asrock Z490 Phantom Gaming Velocita with BIOS version 1.30. The other components remain the same:
- CPU cooler: NZXT Kraken X52 + “case fan” to cool the voltage converters
- Graphics card: MSI Nvidia Geforce RTX 2080 Ti Lightning Z (clock rate per MSI Afterburner to +1000 (MEM) to bypass GPU limit).
- Power supply: Seasonic Prime Platinum (750 Watt, 80-Plus Platinum)
- 16 GiByte Dual-Ranked-RAM per available memory channel. Speed depending on CPU manufacturer specification.
Driver and operating system
- Current Geforce driver in the standard configuration
- Windows 10, 64 bit version in the latest version
Before we refer to the two test candidates, we are gathering important information about Comet Lake-S. The table hides many interesting details about the new Intel CPUs, which we will now summarize briefly:
- Only the models Core i9-10900 (K/F), Core i7-10700 (K/F) and Core i5-10600 (K/F) have a TDP of 125 watts.
- Only Core i9 and Core i7 offer memory support of up to DDR4-2933, all other models have to make do with the well-known DDR4-2666.
- Only Core i9, Core i7, Core i5-10600 (K/F) and Core i5-10400 (F) models are soldered and are based on the 10-core die of the 10900K (Stepping: Q0), all other CPUs are based on a 6-core die (Stepping: G1) and use thermal paste under the heatspreader.
- Only Core i9 CPUs, except the Core i9-10900T, offer the Thermal Velocity Boost.
- Only Core i9 and Core i7 offer Turbo Boost 3.0.
- All Pentium and Celeron processors offer no boost at all. (Explanations of the boosts below)
- All CPUs, except the smallest Celerons, offer SMT.
- No new iGPU is used, even the top models still use the old UHD Graphics 630, which is already overstrained with most titles in FullHD, but is perfectly adequate for office work.
- In terms of price, Intel has corrected the CPUs downwards, for example: The Core i7-10700K corresponds pretty much to a Core i9-9900K, but costs around 100 euro less. AMD should be the decisive reason for this with Zen 2.
Overall, the K-CPUs get the lion’s share of the new features, the advantages of a higher energy budget and better memory connectivity, which makes the cheaper representatives of the hexa and the new quad cores look relatively bad.
The upgrader with today’s tuning ambitions must therefore limit himself to the K-CPUs from Comet Lake-S and pays a hefty surcharge compared to the practically equally fast CPUs without K-suffix, especially because the new motherboard must be able to cope with the higher power consumption of those processors and Z490 boards are to be offered consistently more expensive than their predecessors.
At least the cooling of the CPUs should be a bit easier, because the heatspreader is now a bit thicker and the die a bit thinner.
Visually, this difference is hardly noticeable, but under load the two CPUs remain cool throughout in our opinion.
Intel Core i9-10900K and Core i5-10600K: Introduction
The first one is the Intel Core i9-10900K, which is the top model of Comet Lake-S.
The CPU has ten native cores, which can provide up to 20 threads per HT.
The basic clock rate is (up to) 3.7 GHz and with a special thermal velocity boost, up to 5.3 GHz single core and up to 4.9 GHz all core are possible.
This new type of boost is used for the first time in desktop processors and the Core i9-10900 (K/F) is the only CPU in the complete Comet Lake S portfolio that has it.
Intel makes it clear that if the CPU is used for a longer period of time, the clock rates will be lower. The TDP of the Core i9-10900K is 125 watts.
Coffee Lake-R has shown that this should not be taken at face value.
Intel’s Core i5-10600K on the other hand is the long awaited successor of the very popular Core i7-8700K.
Yes, there is also a Core i7-9700K, but this CPU has eight native cores without SMT, while a Core i5-9600K only has six, but also without SMT.
Therefore, a Core i7-8700K is often still the better choice.
Back to the Core i5-10600K: Intel gives the new Hexacore a TDP of 125 watts and up to 4.8 GHz. The corresponding models in comparison:
|Model||Cores / threads||Base clock||All-core boost||Turbo Boost 2.0 (SC)||Turbo Boost 3.0 (SC)||Thermal Velocity Boost (SC / MC)||TDP (watts)||Memory support||Price (US dollars)|
|Core i9-10900K||10c / 20t||3.7 GHz||4.8 GHz||5.1 GHz||5.2 GHz||5.3 / 4.9 GHz||125||DDR4-2933||488|
|Core i9-9900K||8c / 16t||3.6 GHz||4.7 GHz||5.0 GHz||–||–||95||DDR4-2666||499|
|Core i7-10700K||8c / 16t||3.8 GHz||4.7 GHz||5.0 GHz||5.1 GHz||–||125||DDR4-2933||374|
|Core i7-9700K||8c / 8t||3.6 GHz||4.6 GHz||4.9 GHz||–||–||95||DDR4-2666||399|
|Core i5-10600K||6c / 12t||4.1 GHz||4.5 GHz||4.8 GHz||–||–||125||DDR4-2666||262|
|Core i7-8700K||6c / 12t||3.7 GHz||4.3 GHz||4.7 GHz||–||–||95||DDR4-2666||370|
|Core i5-9600K||6c / 6t||3.7 GHz||4.3 GHz||4.6 GHz||–||–||95||DDR4-2666||263|
The Core i5-10600K isn’t only cheaper than its role model, but the clock rates are all higher.
Obviously, Intel has continually optimized the manufacturing process over the years, but of course higher clock rates are also possible with a purified energy corset.
The key question is whether the CPUs need such a high TDP to keep up, or whether they operate as efficiently as their predecessors, at least in terms of games.
Completely new is the ability to turn off HT for individual cores.
If you like, you can also choose to run the 10-core with 12 or 16 maximum threads instead of 20, which should provide more performance in some games.
Whether all processors of the Comet Lake generation have this feature, we couldn’t find out until the editorial deadline.
However, an inquiry has been received by Intel, we will inform you at the appropriate place as soon as we have more information.
This and other new features are included in an updated version of the well-known Intel Extreme Tuning Utility.
This tool works similar to AMD’s Ryzen Master and allows overclocking the CPU without having to go into the BIOS.
Intel Core i9-10900K and Core i5-10600K: The different boost modes explained
Everything starts at the base clock: The TDP of the respective CPU guarantees the operation with this clock frequency.
If the CPU now has a turbo mode, individual cores (or all of them) adapt to the respective workload.
Intel’s Turbo Boost Max Technology 2.0, the mode that all new core models from Comet Lake-S feature, raises individual cores to a preset maximum, which is often the case with games, as they are not yet programmed to use many CPU cores.
The Intel Turbo Boost Max Technology 3.0 follows on from this, but has a special feature:
Similar to the Zen 2 processors from AMD, Comet Lake processors also have individual “higher quality” CPU cores, which can handle a higher clock frequency with the same power supply.
This boost mode automatically detects these cores and then clocks them 100 MHz higher again.
Like Boost 2.0, Boost 3.0 is also limited to single-core.
The highest of the sensations is the Intel Thermal Velocity Boost: This feature was already introduced in April 2018, but only in the mobile sector.
A Core i9-9900K, for example, also has this boost, at least theoretically, because it is disabled by default on this model.
With the help of the Thermal Velocity Boost, the CPU boosts in the single-core range by another 100 MHz higher, but only if the core temperatures are below 70°C, which should be possible with a decent cooling.
In Comet Lake-S, only the Core i9-10900 models have this feature, with the exception of the T-model.
Finally, the all-core boost is the clock speed that is applied when all CPU cores are equally loaded, which is often the case in applications like Cinebench or Handbrake.
Note that support for the boost modes alone is not decisive for their correct functioning.
Mainboard and especially CPU cooling also play a significant role. There are certain rules for the individual Boost Modes.
Intel’s recommendation is divided into the parameters “PL1”, “PL2” and “Tau”.
PL1 describes the TDP of the model. PL2 describes the maximum TDP which may be applied for the duration of “Tau”. For Coffee Lake this time value was 28 seconds.
For Comet Lake-S Intel doubled the time to 56 seconds. This means that the CPU is allowed to use an energy budget of 250 watts for a period of 56 seconds.
Many benchmarks do not take longer than one minute. A prankster who thinks bad thoughts about this.
We have carefully analyzed the boost behavior of the Core i9-10900K and Core i5-10600K and found that the CPUs meticulously stick to their energy budget, or rather don’t even reach it.
If one limits the CPUs to the exact specifications of Intel, individual CPU cores gradually clock down in order to stay within the maximum range of the given TDP.
However, the cores do this also if you leave all settings on “Auto”.
In games, this is hardly noticeable anyway, since the limits are outside the achieved values; in applications where the CPUs are naturally better utilized, however, a limit causes a weak performance drop compared to an “open” setting, although the absolute consumption under load has not changed or only barely changed.
The temperature under load was also interesting in our case: Both in applications, as well as in games, the Intel Core i9-10900K in particular made sure not to get warmer than 70°C.
The sweet spot between waste heat, power consumption and performance is exactly there.
Overall, the Comet Lake processors thus behave very similar to AMD’s Zen 2 CPUs.
They boost as high as possible, especially in the single-core range, which accelerates games considerably, and stick to the all-core boost in applications, which brings us to the benchmarks.
No one has doubted the Core i9-10900K’s dominant gaming performance in advance.
With ten cores that clock up to 5.3 GHz, this CPU logically has to deliver the best performance.
But there are exceptions like Total War Three Kingdoms or The Witcher 3, which can’t do anything with more than eight cores.
Then even a higher clock speed is useless.
Absolutely impressive is the performance of the Core i9-10900K in Anno 1800.
Ubisoft’s construction game shows the best case when it comes to single core optimization and CPU-internal low latencies.
Even the small Core i5-10600K is faster here than a Core i9-9900KS.
Assassin’s Creed Odyssey on the other hand shows a more realistic everyday picture of the two test candidates.
Overall, the performance in games can be described as successful on both models. The small i5 is a bit faster on average than AMD’s 3700X and the i9 raises the bar up to 31 percent.
The Intel Core i9-10900K’s good frametimes once again confirm the high gaming performance, as they turn out very low and smooth throughout.
The Core i5-10600K can also shine here thanks to the high clock rate and the plus of L3 cache.
More benchmarks and frametimes can be found in the upcoming PCGH issue.
Intel Core i9-10900K and Core i5-10600K in review: Application Benchmarks
While Coffee Lake processors were relatively weak in applications despite many cores in the desktop, Intel has successfully addressed this criticism with Comet Lake-S.
Of course, the two additional cores of the Core i9-10900K help, but the higher clock speed and internal improvements to the L3 cache also bear fruit.
The ten-core comes dangerously close to the Ryzen 9 3900X and the Core i5-10600K takes on the Zen-2 hexacores.
The eight- and ten-cores of the Comet Lake generation are thus also interesting for Content Creator, even though Intel advertises the CPUs as game processors.
Intel Core i9-10900K and Core i5-10600K: Power consumption
At the latest now the bomb will explode. The first two Comet Lake-S models do not have a very high power consumption under load, as expected by most people.
Of course we evaluate the power consumption in games with more weight.
And that is pretty much on a par with the Core i9-9900K in the Core i9-10900K.
Sure, a 10900K can boost up to 250 watts, but in most games this isn’t even necessary in order to clock up to the 5.3 GHz velocity boost.
We could measure the highest single value in Shadow of the Tomb Raider with 93 watts average power consumption.
This is even below the TDP of an Intel Core i9-9900K and far away from Intel’s TDP recommendations for the Core i9-10900K.
Since we basically test CPUs with TDP lock, thus limiting them to Intel’s specifications, we expected a consistently higher power consumption with open limits.
We have increased PL1 and PL2 to 4,000 watts and the tau value to over 128 seconds.
The test with the “Unlocked”-10900K then only revealed an average power consumption in games increased by one (in number 1) watt.
The value even remained the same in the applications we use.
This naturally begs the question why Intel gives the top model such high TDP rates.
One explanation would be that Intel wants to create clear conditions in terms of TDP, as many have criticized the high power consumption of the predecessor, Core i9-9900K.
In any case, we can give the all-clear in terms of power consumption for ordinary everyday tasks:
The Comet Lake-S processors in the form of the Core i9-10900K and Core i5-10600K work more efficiently than their predecessors and don’t let the consumption explode.
The highest value was recorded with the Core i9-10900K in Handbrake: 187 watts.
On average, however, the CPU works just as efficiently in applications as an AMD Ryzen 9 3900X – and in terms of gaming efficiency, both opponents are closer to each other than AMD would like.
You’ll find out more in the upcoming issue of PCGH, as well as in the upcoming update of the efficiency index.
Intel Core i5-10600K Verdict: Best performing Gaming CPU
Apart from the lack of support for PCI-E 4.0, it is actually difficult to see anything negative about the CPUs.
The main focus of this test was gaming performance, a discipline that both processors passed with flying colors.
They are consistently more efficient than their predecessors, faster than AMD’s Zen 2 and easily outperform Coffee Lake.
The latter are quite equal, as long as games with older engines are called; but on average Comet Lake-S tops the 30 percent predicted by Intel.
Intel has also reconsidered the prices and adapted them to the competition.
Comet Lake-S is faster, cheaper and more efficient than Coffee Lake, and that with 14nm production still continuing.
Ranking Second: Ryzen 5 3600X
- Beats Intel Core i5-9600K in applications, almost matches 8700K and 10600K
- Gaming performance significantly improved, 10% over the previous generation
- Best price-performance ratio
- No integrated graphics
Best price-performance ratio: Ryzen 5 3600x
After the Ryzen 5 3600, the Ryzen 5 3600X under test is the most affordable CPU with Zen2 architecture from the current Ryzen 3000 generation.
The six-core processors only differ from each other in terms of clock speed and thermal power dissipation, but the 3600X currently costs about 30 Dollar more than the 3600 without “X”.
The obvious question, whether it’s worth spending the extra money, is investigated in the test of the Ryzen 5 3600X with benchmarks.
You can find details about the most important innovations of the Ryzen 3000 generation, as well as the X570 mainboards that were also introduced with it, in our review of the Ryzen 9 3900X and Ryzen 7 3700X.
Ryzen 5 3600X: gaming benchmarks
When two processors like the Ryzen 5 3600X and Ryzen 5 3600 rely on an identical architecture, have the same number of cores and threads and there are no differences in cache sizes, the only thing that practically remains is the clock speed under typical gaming load as a potentially important factor for performance.
In our experience, Ryzen CPUs do not benefit much from higher clock rates anyway.
On the other hand, there is only a small difference between our test samples of the two CPUs.
The Ryzen 5 3600X clocks at an average of about 4.2 GHz in our benchmark titles, whereas the Ryzen 5 3600X clocks at 4.1 GHz in the case of the Ryzen 5 3600.
The 100 MHz difference is hardly reflected in the benchmarks at all, both processors deliver practically identical performance.
In applications, the picture is the same as in the gaming tests, which is reflected in our performance rating in an identical result for the Ryzen 5 3600X and Ryzen 5 3600.
Thanks to six cores and virtual core doubling, both processors simultaneously achieve a remarkable result compared to the usually much more expensive processors with eight or more cores.
However, the Ryzen 5 3600X cannot provide any arguments that would justify its additional cost compared to the Ryzen 5 3600.
Not surprisingly, the performance of the Ryzen 5 3600X continues in a duel with the Ryzen 5 3600 in the streaming benchmarks, where the CPUs also deliver largely identical results.
For streaming in 1080p with high image quality at 60 fps, both processors are only partially suitable.
The change to the lower 720p resolution, on the other hand, ensures an absolutely smooth stream with a pleasingly low fps loss in the game itself.
Also in terms of power consumption, the Ryzen 5 3600X achieves very similar results to the Ryzen 5 3600 in our measurements, with both processors proving to be efficient models thanks to 7 nm manufacturing, which does not have to fear comparison with Intel’s competition.
The CPU test system
We use Nvidia’s extremely fast Geforce GTX 1080 Ti as the graphics card for our test system so that the GPU becomes a bottleneck as late as possible.
The change to the even faster Geforce RTX 2080 Ti is already planned for the next test system.
The tested games cover different genres.
Among them are Assassin’s Creed: Origins (action role-playing game, third-person), Civilization 6 (lap strategy), Kingdom Come: Deliverance (role-playing game, first-person), Project Cars 2 (racing game), Total War: Warhammer 2 (real-time strategy) and Wolfenstein 2: The New Colossus (first-person shooter).
DirectX 12 is deliberately left out.
The interface still hasn’t caught on and leads to lower fps on our test system in Civilization 6 as well as in Total War: Warhammer than with DirectX 11.
In the form of Wolfenstein 2, we have at least one title with the mostly technically very convincing, but not yet widely used Vulkan interface in our program.
We decided on Full HD (1920×1080) and high details with regard to resolution and detail level.
In higher detail levels and resolutions like WQHD (2560×1440) or 4K (3840×2160) the graphic card becomes more and more a bottleneck.
Lower resolutions, like 1280×70, are too far away from practice in our opinion.
So far we have measured the performance in games with Fraps, now OCAT (Open Capture and Analytics Tool) is used instead.
OCAT outputs the so-called “99th percentile” instead of the minimum achieved fps. As the name suggests, 99 percent of all measured values are above this limit.
Compared to the specification of a pure minimum fps value, the 99th percentile has the great advantage that it filters out individual slips (or the one percent of the lowest values).
When playing the game itself, these slip-ups usually don’t make themselves noticeable, so the 99th percentile is more practical, since it better describes the mostly achieved performance.
All processors are tested with 16.0 GByte DDR4 RAM in dual channel operation and with a clock rate of 2,933 MHz.
In our experience, the use of memory with even higher clock rates can help Ryzen processors in particular to achieve higher performance.
However, we decided to measure with the officially maximum guaranteed clock of the Ryzen 2000 CPUs and to create the same conditions for the other processors in the test field (even though some of them only officially support slightly lower clock rates such as 2,666 MHz or 2,400 MHz as in the case of the Core i3 8100 – or slightly higher as in the case of Ryzen 3000).
The operating system used is Windows 10, which is installed on an SSD just like the games.
We test the application performance of the processors in various practical scenarios, such as encoding a video via handbrake or measuring the load times in Civilization 6.
We also determine the performance of the CPUs when streaming via OBS to Twitch.tv.
The power consumption is also determined, on the one hand when playing Assassin’s Creed: Origins, on the other hand in Cinebench R15, which uses all cores and threads to full capacity.
Already relatively short time after its release, the price of the Ryzen 5 3600X from this test approaches more and more to the Ryzen 5 3600 – no wonder, after all both models deliver almost the same performance in our benchmarks.
Verdict: Best price-performance ratio: Ryzen 5 3600x
While it is quite possible that the Ryzen 5 3600X achieves slightly better results in overclocking, in my experience there is not much room for improvement with the Ryzen 3000 CPUs anyway.
Last but not least, there is the slightly better included cooler (Wraith Spire instead of Wraith Stealth) as a possible argument in favor of the Ryzen 5 3600X.
For us personally, this is relevant as we do not rely on separately purchased CPU coolers. AMD’s models deliver solid performance, but in terms of cooling performance and volume they still can’t keep up with many models from other manufacturers.
Personally, I would always go for the Ryzen 5 3600X instead of the Ryzen 5 3600 as the performance in OC varies. Also the better cooler is worth the money alone.
All in all, the Ryzen 5 3600X offers great performance compared to the i5 10600K at a lower price.
This makes the Ryzen 5 3600X our price-performance ratio winner.
Conclusion: Intel Core i5-10600(K) vs. Ryzen 5 3600
In the end, it was a really close battle between the Intel i5 10600(K) and the Ryzne 5 3600 from AMD.
Performance wise both CPUs are very strong.
All in all, if you want the best gaming performance without compromises, you should get the Intel i5 10600(K).
If you use your PC mostly as a workstation and not primarily as a gaming station you will benefit more from the Ryzen 5 3600.
Also: The Ryzen 5 3600 offers the best price-performance ratio between the two tested CPUs.
Considering you only really get 10% more in performance with the i5 10600K compared to the Ryzen 5 3600, but pay over 30% less for the Ryzen 5 3600, makes for an easy decision for budget conscious users.
All in all we have to say, that we prefer the Ryzen 5 3600. It just offers amazing performance for a great price.
There are also rumors that at the end of 2020 AMD will announce the new Ryzen 4000 series.
If this series convinces, it could be a deathblow for Intel, especially when AMD keeps its pricing strategy.
We will keep you updated!