Overclocking AMD’s Ryzen 5 1600X – has performance improved since launch?
We’re six months on from the release of AMD’s Ryzen CPU range. This is when the platform should be at its most stable and had all the kinks ironed out through BIOS and microcode updates, so we should be able to get the most out of it. How far can we push our favourite Ryzen chip – the Ryzen 5 1600X – with our limited overclocking skills, then?
Overclocking isn’t some dark art, learn the basics with our simple guide on how to overclock.
The short, sad answer is: not much.
The 1600X is a six-core, 12-thread CPU that costs around the same as Intel’s current range of K-series Core i5 processors. The Intel chips were the go-to gaming CPUs because of their mix of great gaming performance, decent price, and overclocking performance. When the Ryzen 5 arrived, however, it became impossible to recommend a resolutely quad-core processor when you could pick one up for effectively the same price, but with three times the thread count.
The straight gaming performance of Ryzen isn’t quite there, but in real terms the differential between the Intel and AMD architectures, when most of the load is going onto the graphics card, is pretty minimal. But, at launch, there was no way to get anywhere near the overclocked performance of the Core i5, so we were hoping that six months later we’d have a little more headroom available to us.
The rated single-core Turbo clockspeed for the Ryzen 5 1600X is 4GHz, so that ought to be an absolute minimum frequency we should be looking at, especially with the competing Core i5 7600K able to hit 5.1GHz without too much effort.
To see what the 1600X is now capable of we’ve used a set of 3,200MHz G.Skill memory and MSI’s X370 XPower Gaming Titanium motherboard. The MSI board’s funky overclocking knob (it’s a physical thing, we’re not denigrating their pro-overclockers) claims in the manual it’s capable of running six-core Ryzens at 4.3GHz when you push it all the way up to 11, but we only got it booting automatically into 4.
1GHz… and even then it crashes as soon as you load it up.
Even at lower, stable automatic settings the board pushes the CPU voltage up to 1.480v, which seems rather high to me. AMD themselves have said that 1.45v is sustainable on their Zen-based CPUs, but running at that level will affect the lifespan of your shiny new processor. So 1.480v is pushing it.
At launch, we were only able to get an all-core overclocked frequency of a flat 4GHz out of the Ryzen 5 1600X. That was with overclocking and this time around we’ve managed to hit that using AMD’s Wraith MAX air-cooler. The Wraith MAX is rated to deal with 140W TDP, while the Ryzen chips top out at 95W, so it’s got a fair amount of headroom in the chiller.
That said, we were able to push it a touch further using Corsair’s h210 liquid cooler. But, in all honesty, the performance difference between 4GHz and 4.05GHz is pretty minimal. Still, the benefits of water-cooling means we get a much quieter machine (48dB vs.
around 60dB) and the peak-to-idle temperature performance is far superior too.
The intervening time has allowed AMD to enable better memory support on the Ryzen platform, however. At launch, we struggled to get a consistent 2,933MHz from all the chips and boards we tested, but now we’re able to get to 3,200MHz with little trouble, and that can make a difference on the AMD architecture.
In all, with the mildly-extended overclocking frequency, and improved memory support, we’ve been able to get another 13%-odd extra performance out of the Ryzen 5 1600X with the more stable, mature platform and a CPU voltage that isn’t going to melt our poor processor. And considering that’s all for free, that ain’t bad.
We’re still hoping that the Zen refresh planned for next year, which looks like it’s getting a die-shrink down from 14nm to 12nm, will provide greater efficiency from the new chips, and allow us to hit higher frequencies than Ryzen’s vanguard has afforded us.
Ryzen 5 1600X vs.
1600: Which should you buy?
The answer to that question may seem obvious to many of you, not least because our Ryzen 5 review said the non-X model looked like a better value. However, we didn’t actually have a plain 1600 on-hand at the time, we just assumed based on what we saw with the Ryzen 7 chips that the 1600 would overclock as well, or at least nearly as well as the 1600X.
Although there isn’t likely to be a dramatic performance difference between the two chips, the $220 1600 comes bundled with the 95-watt ‘Wraith Spire’ cooler, while the $250 1600X doesn’t come with a cooler at all. If you add the cost of a basic heatsink to the 1600X, the non-X version is around 20% cheaper, yet it shouldn’t be more than 10% slower out of the box.
The ‘X-rated’ 1600X also features «XFR» or Extended Frequency Range, which is basically a beefier version of Turbo that lets the CPU overclock itself a little harder when attributes such as power, current and thermals are within acceptable ranges.
XFR is enabled by default on ‘X’ model Ryzen CPUs, though it’s disabled when manually overclocking.
The 1600X has a base clock speed of 3.6GHz, a boost clock of 4.0 GHz and a maximum XFR boost speed of 4.1GHz. However those figures are a little misleading.
For example, boost frequencies will only allow the 1600X to hit 4 — 4.1GHz for single-threaded workloads, whereas the CPU won’t actually boost higher than 3.7GHz on multi-threaded workloads, or a mere 3% increase over the base clock.
The 1600 on the other hand operates at a base clock frequency of just 3.2GHz and is advertised to boost as high as 3.6GHz, though again that boost frequency will only be achieved when taxing a single core. However, despite advertising a boost speed of 3.6GHz, stressing a single core in most cases also sees the 1600 exhibit XFR-like gains, allowing it to reach 3.7GHz. Interestingly, it seems like all Ryzen CPUs feature XFR.
The 1600’s maximum multi-thread frequency is 3.4GHz, which suggests that for heavy workloads it will only be clocked 8% lower than the 1600X.
Of course, these figures will be different if you manually overclock, which is certainly worth considering given that Ryzen CPUs are unlocked and the 1600 can be pushed quite hard with the Wraith Spire box cooler.
We can already say that if you don’t plan to overclock, the 1600X is the faster CPU while the 1600 is the better value as it can’t be more than 10% slower and yet it’s at least 20% less expensive thanks to the included box cooler and lower MSRP.
However, we aren’t sure if that story remains true when overclocking, especially if the 1600X offers more headroom. Rumor is, the X-models are capable of greater frequencies and we’ve seen some evidence of this with the 1700X and 1700, though for the most part the margins were thin.
The 1600X chip we have on-hand for testing had no trouble hitting 4.1GHz at just 1.38 volts and it was even possible to benchmark the system at 4.2GHz after increasing the voltage to 1.5, but AMD recommends a maximum of 1.45 volts for sustained use.
Set at the same 1.38 volts, the 1600 reached 4GHz — not bad — and increasing the voltage a fraction more to 1.4 allowed us to achieve a stable overclock of 4050MHz, only 50MHz shy of what the 1600X managed.
Of course, this could have very easily been around the other way. I just happened to receive a good 1600X. I’ve seen plenty of reviewers and now users limited to 4GHz with the 1600X, so as always with overclocking, your mileage will vary (4GHz does seem nearly guaranteed, though).
For single-threaded workloads, the manual overclock won’t actually be any faster than the stock 4.1GHz XFR performance, but multi-threaded performance should be up to 11% faster for the 1600X and 21% faster for the 1600.
Before moving on, we should note that this isn’t an in-depth Ryzen 5 1600 review — in fact, it’s not really a review at all. That will come later along with coverage of the quad-core 1400 model.
Instead, we thought it would be interesting to take a detailed look at the performance differences between AMD’s six-core Ryzen 5 processors, assuming you’re lucky enough to be in the market for a new CPU priced between $200 and $250.
Test System Specs
Benchmarks
Something I should note right away is that even with G.Skill’s new FlareX DDR4-3200 CL14 memory, the Ryzen 5 1600 was limited to running the memory at 2933. This was the same limitation we faced with the 1500X, so high-speed memory support is still dependent on the CPU. This will hand the 1600X a slight advantage, over the advantage it already had due to those higher clock speeds.
In any case, even with DDR4-2933 memory the 1600 was still good for a memory throughput of 35GB/s.
When testing with Cinebench R15, the 1600X is 10% faster in the multi-threaded test and 11% faster in the single-threaded test, which is about what we would expect to see out of the box. Remember the 1600X is clocked 11% higher for single-threaded work and 9% higher for multi, so these figures make sense.
7-Zip provides a glimpse into real world application performance with both the compression and decompression tests using all available threads on the 1600X and 1600.
Here the 1600X was around 7 to 8% faster than the 1600, which is in line with what was previously seen in Cinebench and again given the out of the box clock speeds these are the margins we expected to see.
Testing game performance with Hitman we find that the 1600X is just 4% faster than the 1600 when comparing either the average or minimum frame rates. Obviously not a significant difference and it explains why we didn’t see many worthwhile gains in most games when overclocking Ryzen.
Ashes of the Singularity: Escalation is the last game I used to compare the 1600 and 1600X. This time we see a mere 2 to 3% performance advantage in favor of the 1600X, so again it looks like you can expect a poor return on investment when buying the 1600X for gaming.
When overclocked, performance is equalized on the Ryzen 5 CPUs. The 1600 fell just 50MHz shy of the 1600X when using reasonable voltages — obviously a negligible difference and as you can see both CPUs allowed the Titan XbigP to average 88fps.
Wrap Up
The Ryzen 5 1600 is the obvious choice as suspected. Between its $220 asking price and bundled cooler, it presents too much of a value proposition to give the cooler-less 1600X much consideration at $250.
All told, with the 1600 CPU and its cooler mounted on an affordable B350 motherboard along with memory such as G.Skill’s FlareX DDR4-2400 16GB kit, you have the core components of a killer six-core/12-thread machine for no more than $400.
How much does Intel’s Core i7-6800K cost again?
The 1600-based build has even more appeal when you realize that it should overclock to at least 4GHz while our G.Skill RAM reached 2933MHz with relaxed timings.
Looking at it that way, the Ryzen 5 1600 is an exceptional value. As I found earlier in the week, even compared to more affordable CPUs such as the Core i5-7600K, the new six-core Ryzen chips are a much better buy, regardless of what it is you plan to do with them.
However, there’s not much reason for anyone to purchase the 1600X over the 1600 as you’re only paying more for less.
You might be wondering why AMD would release a 1600X at all and I’d say it exists simply to maximize profit.
Shopping shortcuts:
- AMD Ryzen 5 1600X on Amazon
- AMD Ryzen 5 1600 on Amazon
- AMD Ryzen 5 1600X on Newegg
- AMD Ryzen 5 1600 on Newegg
AMD’s playing a similar game as Intel with these chips. The difference being that they are doing it without completely taking advantage of enthusiasts who want to save a little money. When it comes to the 1600, you’re being rewarded for doing your homework and a little tinkering.
Windows 10 CPU frequency control
Most modern processors use various power-saving technologies such as Intel SpeedStep or AMD Cool’n’Quiet. These technologies are based on the dynamic change in the frequency of the processor depending on the load in order to reduce power consumption and heat dissipation.
For understanding, let’s consider the general principles of these technologies.
The processor has performance states (P-States), which are a combination of frequency multiplier (Frequency ID, FID) and supply voltage (Voltage ID, VID). The clock frequency of the processor is obtained by multiplying the frequency of the system bus (FSB) by the FID, respectively, the larger the multiplier, the higher the frequency, and vice versa. The number of supported states depends on the characteristics of the processor (max. frequency, multiplier, etc.). nine0003
During initial boot, the BIOS creates a description of possible performance states. This description, according to the ACPI interface, is read by the operating system at startup. During operation, the operating system monitors the processor load, when the load decreases, it calls the processor driver and puts the processor in a reduced state. By lowering the frequency and voltage, the processor will consume less power and, accordingly, heat up less. Well, when the load increases, the operating system will again request a change in the state of the processor, but in a big way.
nine0003
When and in which of the states to transfer the processor, the operating system decides automatically, in accordance with the current power saving policy. But, besides this, Windows has the ability to manually set the adjustment range, limiting the minimum and maximum state.
My computer has an Intel Core I7 4790K. According to the specification, it has a base frequency of 4GHz, and using Turbo Boost technology, it can be accelerated to a maximum of 4.4GHz. In order to see the current speed of the processor, launch the «Task Manager» (Ctrl + Shift + Esc) and go to the «Performance» tab. As you can see, at the moment the load is low and the processor is working at half strength, its frequency is 1.84 GHz. nine0003
Let’s try tweaking the CPU clock settings a bit and see what happens. To launch the power management snap-in, press the keys Win + R and execute the command powercfg.cpl .
In the window that opens, select the current power plan, click on the link «Setting the power plan»
and click on the link «Change advanced power settings».
nine0003
The Minimum Processor State and Maximum Processor State parameters located in the Processor Power Management section are responsible for the frequency of the processor.
Note that laptops have two settings for each setting. The first is responsible for the frequency of the processor during autonomous operation (on battery power), the second — when operating from the network.
Let’s check how changing the settings affects the frequency of the processor. First, let’s reduce the maximum state to 20% and check the result. As you can see, the operating frequency has dropped to about 0.78 GHz, which is just about 20% of the base frequency. nine0003
Interestingly, I did not succeed in lowering the frequency below 20%. When setting the value below 20%, the frequency continues to remain at the same level, i.e. for my processor 800MHz is the minimum supported frequency.
To find out why, let’s run the CPU-Z diagnostic utility.
As you can see, in our case, the bus frequency is 100 MHz, and the multiplier varies in the range from 8 to 44. Hence, we get the possibility of changing the frequency from the minimum 800 MHz to the maximum 4.4 GHz in 100 MHz steps. nine0003
But switching is not carried out by each multiplier, but more discretely. In other words, the number of performance states does not match the number of multiplier values. You can view all available states with the RightMark Power Management utility. For example, only 15 states are available for the processor under test, and switching between them occurs with a variable step of 200-300 MHz.
Well, what does the minimum processor frequency affect? To check, let’s set the minimum frequency to 100% and make sure that the processor speed immediately rose almost to the maximum and reached 4.3GHz. And this despite the fact that the processor load was only 14%. nine0003
There is one more setting responsible for the frequency of the processor.
By default, it is hidden and in order to see it, you need to make some manipulations in the registry. Therefore, open the registry editor (Win + R -> regedit), go to the HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Power\PowerSettings\54533251-82be-4824-96c1-47b60b740d00\75b0ae3f-bce0-45a7-8c89-c9611c25e100 section, find the parameter Attributes and change its value to 2 .
After that, the «Maximum processor frequency» parameter will appear in the settings window. As the name implies, this parameter is responsible for limiting the maximum frequency of the processor, its value is set in megahertz. The default value of the parameter is 0, which means no restrictions.
Set the limit to 1500MHz and check the result. Indeed, the limitation works and the frequency does not rise above the specified value. nine0003
It must be understood that frequency control works in the same discrete way as in the case of states.
For example, when setting the maximum frequency to 1200 MHz, the real limitation will be in the region of 1000 MHz. This is not surprising, because switching is still performed between P-states, and the nearest minimum value is chosen.
You can also change additional processor power settings from the command console using utility powercfg . To begin with, let’s display all the available power settings with the command:
powercfg /query
In the results obtained, you need to find the required parameters. Each section and setting can be identified by a GUID or an alias. For example, the «Processor Power Management» setting group has GUID 54533251-82be-4824-96c1-47b60b740d00, and the alias is SUB_PROCESSOR, and the «Maximum Processor State» setting has GUID 75b0ae3f-bce0-45a7-8c89-c9611c25e100 and the alias PROCTHROTTLEMAX. nine0003
Knowing the necessary aliases or GUIDs, you can operate with the settings. So you can see the value of the “Maximum processor state” parameter with the following command (SCHEME_CURRENT means the current power scheme):
powercfg /query SCHEME_CURRENT SUB_PROCESSOR PROCTHROTTLEMAX
In the command line, the parameter values are shown in hexadecimal form, i.
e. the value 0x00000064 means 100%.
Please note that both mains and battery settings are available on the command line. Depending on the required option, the command will be different. So the key / SET 9 is responsible for setting the power supply from the network0019 AC VALUEINDEX and /SET DC VALUEINDEX is used for battery power. The difference is only one letter, so you need to be careful and try not to confuse them. For example, let’s reduce the maximum processor state for mains power to 50%:
powercfg /setacvalueindex CHEME_CURRENT SUB_PROCESSOR PROCTHROTTLEMAX 50
Let’s check the result. As you can see, the current value is 0x00000032 (50%).
So, we made sure that there are processor frequency controls in Windows 10 and even work. But in order to get a positive effect from them, a competent approach is needed. The first step is to decide what you want to get as a result — to increase performance, increase battery life, reduce temperature, or something else.
Then you need to find out the main parameters of your processor, such as bus frequency, multiplier, maximum operating frequency. And already on the basis of this, empirically select the optimal values for yourself. nine0003
Something like this 🙂
Ryzen 5 1600X processor [in 3 benchmarks]
AMD
Ryzen 5 1600X
- Interface
- Core frequency
- Video memory size
- Memory type
- Memory frequency
- Maximum resolution
Description
AMD launched AMD Ryzen 5 1600X on March 16, 2017 at a MSRP of $249. This is a desktop processor based on the Zen architecture, primarily designed for office systems. It has 6 cores and 12 threads and is manufactured in 14 nm process technology, the maximum frequency is 4000 MHz, the multiplier is unlocked.
Compatibility is a socket processor
AMD socket AM4
with TDP 95W. It supports DDR4 memory.
It provides poor benchmark performance at
10.
53%
nine0003
from the leader, which is AMD EPYC 9654.
EPYC
9654
Compare
General information
Information about the type (desktop or laptop) and architecture of the Ryzen 5 1600X, as well as when sales started and cost at that time.
| Place in the performance ranking | 607 |
| Core | + |
Compatible
Information on Ryzen 5 1600X compatibility with other computer components.
Useful, for example, when choosing the configuration of a future computer or to upgrade an existing one.
Please note that the power consumption of some processors can significantly exceed their nominal TDP even without overclocking. Some may even double their claims if the motherboard allows you to adjust the power settings of the processor. nine0003
| AES-ni | + | |
| AVX | + |
| AMD-V | + |
Memory support
2
Benchmark tests
These are the results of the Ryzen 5 1600X performance tests in non-gaming benchmarks.
The overall score is set from 0 to 100, where 100 corresponds to the fastest processor at the moment.
Overall benchmark performance
This is our overall performance rating. We regularly improve our algorithms, but if you find any inconsistencies, feel free to speak up in the comments section, we usually fix problems quickly. nine0003
Ryzen 5 1600X
10.53
- Passmark
- GeekBench 5 Single Core
- GeekBench 5 Multi-Core
Passmark
Passmark CPU Mark is a widely used benchmark that consists of 8 different tests, including integer and floating point calculations, extended instruction tests, compression, encryption, and game physics calculations. Also includes a separate single-threaded test. nine0003
Benchmark coverage: 67%
Ryzen 5 1600X
13065
GeekBench 5 Single-Core
GeekBench 5 Single-Core is a cross-platform application designed as CPU benchmarks that independently recreate certain real world tasks that can accurately measure performance.
This version uses only one processor core.
Benchmark coverage: 37%
nine0019 Ryzen 5 1600X
942
GeekBench 5 Multi-Core
GeekBench 5 Multi-Core is a cross-platform application designed as CPU benchmarks that independently recreate certain real world tasks that can accurately measure performance. This version uses all available processor cores.
Benchmark coverage: 37%
Ryzen 5 1600X
5120
nine0003
Relative capacity
Overall performance of the Ryzen 5 1600X compared to its closest competitor in desktop processors.
AMD Ryzen 5 3500X
101.23
AMD Ryzen 5 2600
101.23
Intel Core i5-11500T
100.66
AMD Ryzen 5 1600X
100
Intel Core i5-10400F
99.72
AMD Ryzen 3 5300G
99.
62
Intel Core i7-8700
99.24
Competitor from Intel
We believe that the nearest equivalent to Ryzen 5 1600X from Intel is Core i5-10400F, which is roughly equal in speed and lower by 1 position in our rating. nine0003
Core i5
10400F
Compare
Here are some of Intel’s closest competitors to the Ryzen 5 1600X:
Intel Core i5-11400T
102.18
Intel Core i7-9700
101.9
Intel Core i5-11500T
100.66
nine0003
AMD Ryzen 5 1600X
100
Intel Core i5-10400F
99.72
Intel Core i7-8700
99.24
Intel Core i3-12100TE
99.15
Other processors
Here we recommend several processors that are more or less similar in performance to the reviewed one.
nine0003
Core i7
8700
Compare
Core i7
7800X
Compare
Ryzen 5
3500X
Compare
Ryzen 5
2600
Compare
Ryzen 5
PRO 2600
Compare
Core i7
9700
Compare
Best graphics cards for Ryzen 5 1600X
We have 3998 configurations based on Ryzen 5 1600X in our database.
According to statistics, these cards are most often used with Ryzen 5 1600X:
GeForce GTX
nine0188 1050 Ti
13.
4%
GeForce GTX
1060 6GB
9%
Radeon RX
580
7.5%
Radeon RX
570
6.6%
GeForce GTX
1060 3 GB
5%
GeForce GTX
1660 Super
3.5%
GeForce GTX
1650
3.2%
GeForce GTX
1070
2.7%
GeForce GTX
1650 SUPER
2.2%
GeForce GTX
1050
2.1%
Here are the most powerful graphics cards used with the Ryzen 5 1600X according to user statistics:
GeForce RTX
4090
0.
2% (9/3998)
GeForce RTX
3090 Ti
nine0020
0.2% (6/3998)
Radeon RX
6950XT
0.03% (1/3998)
GeForce RTX
3080 Ti
0.3% (13/3998)
GeForce RTX
3090
0.2% (9/3998)
Radeon RX
nine0188 6900XT
0.1% (5/3998)
GeForce RTX
3080
0.2% (8/3998)
Radeon RX
6800XT
0.05% (2/3998)
GeForce RTX
3070 Ti
0.1% (4/3998)
GeForce RTX
3070
0.
