Vibrations of the cesium atom. Atomic clocks: history and modernity

On DDR4 SDRAM – RAM new generation. The standard describes the memory design in every detail, but very little or nothing is said about what, in fact, the differences from the memory of the previous generation are, or what advantages the transition to DDR4 promises (probably, hoping for the imagination of readers :-)) .
In this post I will try to outline the main differences between DDR3 and DDR4, and what benefits it brings to end users new standard memory.

Increased volume and productivity

A comparison of DDR3 and DDR4 shows that the largest DDR3 module that could theoretically be designed would be 128GB in size (using a QDP (quad die package) and 8Gb dies) (Figure 1). For DDR4, using 16Gb crystals and eight-layer packaging of crystals into a chip, it is theoretically possible to create a memory module with a capacity of up to 512GB. The number of pins on DDR4 modules has increased to 284 to address this amount of memory. Each DDR4 memory chip can be a stack of 2, 4 or 8 DRAM chips. The 8-layer stack is described in the addendum to the specification and will most likely require the use of TSV (through silicon via) for its practical implementation.
In general, all these changes are aimed at creating higher-capacity memory modules and increasing performance.

Figure 1. Comparison of DDR3 and DDR4 modules

Improved Energy Efficiency

The electrical implementation of the data input/output interface has also changed. New interface is called pseudo-open drain (POD, “pseudo-open drain”) and its main difference is that no current flows in the circuit when a high level voltage. The electrical interfaces of DDR3 and DDR4 are shown in Figure 2.

Figure 2. Electrical data I/O interfaces for DDR3 and DDR4.

Reducing the I/O voltage, changing the electrical interface and reducing the length of lines in memory banks lead to a significant reduction in power consumption compared to DDR3. Preliminary estimates indicate a 30% gain. Although, of course, this depends on the nature of memory accesses, the technical process and many other factors. Such a gain can be used to increase the clock frequency and, accordingly, the speed of operation, or to save a little energy for the same performance.

Reliability

Another example is that DDR4 chips have a connection test mode. This mode allows the memory controller to test electrical connections (and find broken lines) much faster than before, and without the use of initialization sequences.
The DDR4 module can also be configured to discard commands containing parity errors. In DDR3, such commands were skipped and reached the memory chips, making recovery from failures much more difficult.
And as an example of one of the optional “features” that the specification contains, one can cite checking checksums for data written to memory.

All these and other features are aimed at ensuring an increase in operating frequencies and memory volumes (associated with an increase in the number of errors in operation), while guaranteeing stable operation.

With higher performance and energy efficiency, DDR4 memory should easily take its place in multi-core server and desktop systems in 2014. And then, due to the lower price per unit of volume in addition to other advantages, DDR4 should reach other devices.

Is DDR4. This is what current Intel processors support (although they also left partial support for DDR3). New boards for DDR4 will also be released for AMD processors, which will appear in early 2017. This article will help you find out how to choose DDR4 for building or a new PC or upgrade, and how RAM for laptops differs from it.

Before choosing DDR4 memory, you need to familiarize yourself a little with the features of this type of RAM. Essentially, DDR4 memory is a surface-mount BGA (solder ball array) chip, which makes it universal for all types of electronics, from routers to servers. However, for ease of installation, as well as maintaining the possibility of upgrading, increasing the available amounts of RAM and overall unification, PCs usually use a modular design.

Format

DDR4 memory chips are soldered onto small boards, ranging from 4 to 16 pieces. Such boards are called DIMM (Double In-line Memory Module - double-sided memory module) and are equipped with 284 contacts. They have same sizes(5.25″ or a little more than 13 centimeters), but are physically incompatible with DDR3, since the latter’s DIMM module has 240 pins. In addition, the modules have different locations of a special key cutout that prevents the board from being installed on the wrong side or in an incompatible slot. DIMMs are the main type of RAM memory for desktop computers and servers.

For laptops with increased requirements for compactness, SO-DIMM modules (Small Outline DIMM - double-sided memory module with small pins) have been created. Also, similar boards are used in monoblocks, nettops and other types of compact personal computers. They are half the size of DIMMs (6.76 cm) and have only 260 pins.

Characteristics

The second criterion by which DDR4 memory can be classified is its performance characteristics. The main ones are clock speed (and the bandwidth directly related to it), latency and voltage.

Clock frequency and bandwidth characterize the performance of memory in sequential reading and writing mode. DDR4 RAM is available with support for frequencies from 1600 MHz (very rare in practice) to 3200 MHz. The most common frequencies at the moment are 1866 MHz (bandwidth - 12800 MB/s), 2133 MHz (17064 MB/s) and 2400 MHz (19200 MB/s). Most computers are designed to work with them.

CAS latency is the delay (measured in the number of work cycles) between submitting a request to read/write data and completing this operation. This parameter characterizes memory performance in random read/write mode. How less value delays - the more responsive the memory. At equal frequencies— the faster module will be the one whose delay duration (latency) is lower.

Voltage— module supply voltage. At the moment, the only common value is 1.2 V. There is also LPDDR4 (low power DDR4) memory that uses lower voltages. It is not yet popular, and is used only in compact devices (ultrabooks, tablets, smartphones) that do not support its upgrade. The disadvantage of this type of memory is its reduced performance for the sake of efficiency.

Choosing DDR4 memory when building a new PC

Choosing DDR4 RAM for a computer that is being built from scratch is the easiest way. As of late 2016-early 2017, the only platform supporting this memory is Intel SkyLake (Core i3-i7 6xxx, Celeron and Pentium of this family). The base memory frequency for this platform is 2133 MHz. Higher frequencies are not supported by all boards and are only achieved during overclocking.

When purchasing a board equipped with two memory slots, it is advisable to purchase one large-capacity stick (8 or 16 GB). This will leave the possibility in the future to add another one to it and double the amount of RAM. For boards with four DIMM slots, you can choose a set of 2 smaller strips. In this case, the possibility of an upgrade remains, and the performance is at least slightly increased due to the dual-channel mode.

Heatsinks on DDR4 memory sticks are more a decorative element than a functional one. The power consumption of this generation of RAM is tiny (about 0.5-2 W), so there is no need for additional cooling. In a case with a transparent wall and backlight, heatsinks on the memory strips will decorate the inside of the PC. However, if the choice is between planks with the same parameters, and the aesthetic component fades into the background, there is no point in overpaying for radiators. They are really useful only for overclockers who overclock RAM much higher than factory frequencies.

The range of available DDR4 memory on the market is gradually increasing. To date, this memory is only compatible with motherboards based on the Intel X99 chipset and, accordingly, processors codenamed Haswell-E (LGA2011-v3 socket). Actually, the fact that DDR4 memory is only compatible with the specified Intel platform already means that it is intended for the most powerful PCs today. All Intel X99 chipset motherboards support up to 64 GB of DDR4 memory in quad-channel mode (assuming the board has eight memory slots). Let's say right away that we're talking about about unregistered (UDIMM) non-ECC memory. The fact is that some boards with the Intel X99 chipset support server processors of the Intel Xeon E5 v.3 family (having the same LGA2011-v3 socket and the same processor architecture). In this case, ECC memory is supported, both registered (RDIMM) and non-registered (UDIMM), and the maximum memory capacity is already 128 GB. However, we will not consider server memory in this article, and in the future, by DDR4 memory we will mean non-register memory without ECC.

As for the capacity of DDR4 memory modules, modules with a capacity of 4 GB (these are the most common) and 8 GB are available on sale. DDR4 memory is sold both in the form of individual modules and in the form of kits consisting of two, four and even eight modules. But the most common are sets of four memory modules (quad-channel sets). Accordingly, the total capacity of such a kit can be either 16 or 32 GB. The most common on the market today are four-channel memory sets with a total capacity of 16 GB, that is, sets of four memory modules with a capacity of each module of 4 GB.

The minimum DDR4 memory frequency required by the standard is 1066 MHz. Accordingly, the effective frequency in this case is 2133 MHz (DDR4-2133 memory), and the bandwidth is 17056 MB/s (in single-channel mode). The maximum memory frequency provided by the standard is 2133 MHz, its effective frequency in this case is 4266 MHz (DDR4-4266 memory), and the bandwidth is 34128 MB/s (in single-channel mode). True, the frequency of 2133/4266 MHz is a reserve for the future, while such memory is not on sale. In fact, today on the market there is memory with an effective frequency from 2133 MHz to 3000 MHz, and it seems that only DDR4-2133 memory is standardized, and faster memory is implemented through XMP profiles.

As a rule, more expensive and faster DDR4 memory modules are equipped with heatsinks that serve no purpose other than attracting the attention of users. Heatsinks on memory modules are purely decorative and, by and large, a pointless thing because memory chips simply don't get hot enough to require cooling using heatsinks. Let's not be unfounded and confirm what has been said with facts. In order to demonstrate the pointlessness of heatsinks on memory modules, we used a pyrometer, which allows us to remotely determine temperature changes. For the test, a DDR4-2133 (15-15-15) memory module was used without a heatsink, the supply voltage was 1.2 V. In idle mode, the temperature of the memory chips was 31.2 °C, and when loading the memory using the Stress System stress test Memory in the AIDA64 utility, the temperature of the memory chips increased to 35.5 °C. When overclocking the same memory to a frequency of 2400 MHz and a supply voltage of 1.35 V, in idle mode the temperature of the memory chips was 32.7 °C, and when the memory was loaded it increased to 38.1 °C. It is clear that at such temperatures there is simply no point in radiators. Additionally, all 4GB DDR4 memory modules are single-sided, meaning the memory chips are located on one side of the module. It would seem that if you glue the radiator, then only on one side. However, heatsinks on such memory modules are always on both sides - it’s just prettier that way.

Now about the cost. As a first approximation, DDR4 memory costs approximately 1 thousand rubles per 1 GB. That is, a memory module with a capacity of 4 GB costs approximately 4 thousand rubles, and a memory module with a capacity of 8 GB costs 8 thousand rubles. However, you need to keep in mind that decorative heatsinks and higher declared operating frequencies lead to an increase in the cost of memory. That is, a DDR4-3000 memory module will be more expensive than a DDR4-2133 memory module (with equal capacity).

AMD Radeon R7 Performance Series (R744G2133U1S)

As strange as it may seem, AMD produces DDR4 memory sets that are currently compatible only with Intel processors. However, this is modestly kept silent, and therefore to find any technical information about DDR4 memory is not possible there. Apparently, pride does not allow this fact to be made public, but the company does not want to give up making money.

According to the information we have, AMD currently offers two four-channel DDR4 memory kits, which differ only in capacity: these are sets of four modules with a total capacity of 32 GB (R748G2133U2S) and sets of four modules with a total capacity of 16 GB (R744G2133U1S). For both sets, the memory frequency is 2133 MHz, and the timings are 15-15-15-36.

Next, we will look at a memory kit of four modules with a total capacity of 16 GB (R744G2133U1S), which belongs to the AMD Radeon R7 Performance series. As already noted, AMD R744G2133U1S memory modules have a frequency of 2133 MHz and timings of 15-15-15-36, and the supply voltage is 1.2 V (this is the standard value).

The declared memory frequency is low (this is the minimum value for DDR4), but there is a high probability that this memory can be made to work at higher speeds. high frequency.

Memory modules are equipped with dark gray cooling radiators, which are two metal plates glued to each side of the module. Moreover, the modules themselves are single-sided, that is, the memory chips are located on only one side.

On our test bench with the default settings in the UEFI BIOS, the AMD Radeon R7 Performance Series (R744G2133U1S) memory started up at a frequency of 2133 MHz with timings of 15-15-15-36, that is, exactly as it should be.

In addition, it turned out that the memory can operate at a frequency of 2400 MHz. When the memory is launched at this frequency, the timings are automatically set to 18-18-18-40, but at a frequency of 2400 MHz this memory can also work with timings 18-11-11-36.

The following are the test results in the AIDA64 program for a set of AMD Radeon R7 Performance Series memory modules (R744G2133U1S) with default settings (DDR4-2133; 15-15-15-36) and overclocked (DDR4-2400; 18-11-11- 36).

Geil Evo Potenza GPR416GB3000C16QC

The Geil GPR416GB3000C16QC quad-channel memory kit belongs to the series. These are four DDR4-3000 memory modules with a total capacity of 16 GB (4 × 4 GB). The memory modules are equipped with burgundy-colored cooling radiators. The memory modules themselves are single-sided, that is, all memory chips are located on one side. In general, it should be noted that the radiators in the memory, let’s say, do not look impressive. The thickness of the plates from which the radiator is made is less than 1 mm. The height of the memory module with radiator is 47 mm.

According to information on the manufacturer’s website, at a frequency of 3000 MHz, Geil Evo Potenza GPR416GB3000C16QC memory modules can operate with timings of 16-16-16-36 with a supply voltage of 1.35 V. Moreover this mode The operation of memory modules is ensured when the XMP profile is activated.

Note that the Geil Evo Potenza series of Quad Channel memory also includes DDR4-2133/2400/2666/2800 memory kits, as well as faster DDR4-3200 memory. Geil Evo Potenza DDR4-3000 quad-channel memory kits can also be different: so, in addition to 16 GB kits, there are also kits with a total capacity of 32 GB. Memory timings may also differ: 15-15-15-35 or 16-16-16-36. Taking into account two possible capacities and two sets of timings, the Geil Evo Potenza DDR4-3000 series includes four sets of memory:

• GPR416GB3000C15QC: timings 15-15-15-35, total volume 16 GB;
• GPR416GB3000C16QC: timings 16-16-16-36, total volume 16 GB
• GPR432GB3000C15QC: timings 15-15-15-35, total volume 32 GB;
• GPR432GB3000C16QC: timings 16-16-16-36, total volume 32 GB.

Now let's talk about the difficulties we encountered when testing the Geil Evo Potenza GPR416GB3000C16QC memory.

First of all, we note that the declared frequency of 3000 MHz with timings 16-16-16-36 and a supply voltage of 1.35 V are the characteristics of the XMP profile. And, of course, it is not a fact that this profile will work on any board and that the memory will even “start up” at this frequency. As practice shows, there are boards on the Intel X99 chipset that, with the default UEFI BIOS settings, try to immediately activate the XMP profile and make the memory work when specified characteristics. With such boards, this memory kit will have big problems and, most likely, it simply will not work. In particular, we tested this memory kit on three boards (Gigabyte GA X99-Gaming G1 WIFI, Asus Rampage V Extreme and ASRock Fatal1ty X99X Killer) and it turned out that the ASRock Fatal1ty X99X Killer board is not compatible with this memory at all.

But on the Gigabyte GA X99-Gaming G1 WIFI and Asus Rampage V Extreme boards with default UEFI BIOS settings, the Geil Evo Potenza GPR416GB3000C16QC memory was detected differently.

So, in the case of the Asus Rampage V Extreme board, the Geil Evo Potenza GPR416GB3000C16QC memory kit is defined as DDR4-2400 with timings 17-15-15-35 (supply voltage 1.2 V).

In the case of the Gigabyte GA X99-Gaming G1 WIFI board, the same memory kit was defined as DDR4-2400, but with timings of 16-16-16-35.

Now about the most important thing. On none of our test boards was the Geil Evo Potenza GPR416GB3000C16QC memory able to work with the settings defined in the XMP profile, that is, at an effective frequency of 3000 MHz with timings of 16-16-16-36 and a supply voltage of 1.35 V. If If you manually set the UEFI BIOS to a frequency of 3000 MHz, timings of 16-16-16-36 and a supply voltage of 1.35 V, the system will not boot. We also tried to roughen up the timings for 3000 MHz, but it was all in vain. At this frequency the memory refused to work.

Through trial and error, it was found that our Geil Evo Potenza GPR416GB3000C16QC memory kit can operate at a maximum frequency of 2666 MHz, no higher. In fact, the declared frequency of 3000 MHz turned out to be simply a hoax. However, we will not make such loud statements at all and will clarify that our specific Geil Evo Potenza GPR416GB3000C16QC memory kit with our specific Intel Core i7-5960X processor and our Gigabyte GA X99-Gaming G1 WIFI board does not meet the declared characteristics.

For 2666 MHz the best timings we could find were: 13-14-14-30. With such timings at a frequency of 2667 MHz, everything works stably, without freezing.

The following are the test results in the AIDA64 program for a set of Geil Evo Potenza GPR416GB3000C16QC memory modules with default settings (DDR4-2400; 16-16-16-35) and overclocked (DDR4-2667; 13-14-14-30).

Kingston HyperX Predator HX424C12PBK4/16

Kingston HyperX Predator HX424C12PBK4/16 memory belongs to the Kingston HyperX Predator overclocker memory series.

As follows from the information, the company produces a very wide range of DDR4 memory kits. The capacity of the kits can be 16, 32 and 64 GB, the number of modules in one kit can be four or eight, and the capacity of one module can be 4 or 8 GB. At the same time, the company produces DDR4 memory kits with effective frequencies of 2133, 2400, 2666, 2800 and 3000 MHz.

On the Kingston website there is a way to decipher the name of the memory module. Using this information, you can understand that the name of the HX424C12PBK4/16 module encrypts the following information: this is a UDIMM DDR4-2400 memory module with CAS 12 latency. The memory belongs to the HyperX Predator series, is equipped with a black heatsink, and the total capacity of the set of four modules is 16 GB.

On our test bench with default UEFI BIOS settings, the Kingston HyperX Predator HX424C12PBK4/16 memory started up at 2133 MHz with timings of 15-15-15-36 and a supply voltage of 1.2 V.

The promised frequency of 2400 MHz with timings 12-13-13-35 is implemented through the XMP profile. Moreover, for the Kingston HyperX Predator HX424C12PBK4/16 memory there are two XMP profiles: one for a frequency of 2400 MHz with timings 12-13-13-35 at a supply voltage of 1.4 V, and the second? for a frequency of 2133 MHz, but with timings 13-13-13-36 and with a supply voltage of 1.2 V.

When the first XMP profile is activated in the UEFI BIOS (for a frequency of 2400 MHz), the memory, as it should, starts at a frequency of 2400 MHz with timings of 12-13-13-35 at a supply voltage of 1.4 V. However, manually for a frequency of 2400 MHz You can also select shorter timings. In particular, on our test bench the memory worked with timings of 12-12-12-35 (at a frequency of 2400 MHz).

But we were unable to run the Kingston HyperX Predator HX424C12PBK4/16 memory at a higher frequency (2600 MHz), even with coarser timings.

AData XPG AX4U2400W4G16-QRZ

AData company in two series: Consumer (user) and Gaming (game). There is also server memory, but we are not considering it now. The memory kit belongs to the Gaming series.

Not worth it in this case take the word Gaming seriously. This is just a marketing positioning of memory, which is aimed at attracting attention. The Gaming series memory differs from the regular Consumer series in the presence of decorative radiators (the radiators have no other meaning) and in the fact that the Gaming series memory is faster.

The AData Gaming series features a very large number various memory kits. Moreover, any AData Gaming series memory module can be purchased separately (one module), in a set of two modules, or in a set of four modules. In addition, both 4GB and 8GB modules are available. This is precisely why the range of possible AData Gaming DDR4 memory kits is very wide.

However, it is not difficult to understand this assortment. There is DDR4-2133 memory with timings of 13-13-13 and 15-15-15. Taking into account the possible capacity of the modules (4 and 8 GB), as well as the different configurations of the sets (one, two and four modules), we find that there are twelve options for DDR4-2133 memory alone.

Next, there is DDR4-2400 memory with timings 16-16-16, DDR4-2666 memory with timings 16-16-16, DDR4-2800 memory with timings 17-17-17 and DDR4-3000 memory with timings 16-16-16 . Again, any memory can be represented in sets of one, two or four modules, and the module capacity can be 4 or 8 GB.

There is also faster DDR4-3200/3300/3333 memory. But for this memory the timings are only 16-16-16, and the modules have a capacity of 4 GB.

Next, we will look at a set of four AData XPG AX4U2400W4G16-QRZ memory modules. As you can easily guess from the name, we are talking about DDR4-2400 memory modules with 16-16-16 timings. The supply voltage of these memory modules is 1.2 V.

On our test bench with default UEFI BIOS settings, the AData XPG AX4U2400W4G16-QRZ memory started up at 2133 MHz with timings of 15-15-15-36 and a supply voltage of 1.2 V.

The promised frequency of 2400 MHz with timings of 16-16-16 is implemented through the XMP profile.

When the XMP profile is activated in the UEFI BIOS, the memory, as it should, starts at 2400 MHz with timings of 16-16-16-39.

We were unable to run the AData XPG AX4U2400W4G16-QRZ memory at a higher frequency. However, at a frequency of 2400 MHz, better timings can be selected. The best timings that we were able to find for this memory at a frequency of 2400 MHz were 13-12-12-36.

AData AD4U2133W4G15-B

If the previous AData kit belonged to the gaming series, then the memory kit belongs to the Consumer series, that is, to the simplest series of DDR4 memory.

The Consumer series includes two types of DDR4-2133 memory modules: 4 GB and 8 GB. In the first case, the modules are called AData AD4U2133W4G15-B, and in the second - AData AD4U2133W8G15-B. All other characteristics of the modules are absolutely the same. The effective memory frequency is 2133 MHz, timings are 15-15-15-36, and the supply voltage is 1.2 V. Memory modules with a capacity of 4 GB are single-sided and are based on SKhynix H5AN4G8NMFR memory chips (8 chips of 512 MB each).

Note that there are no heatsinks on the AData AD4U2133W8G15-B memory modules.

On our test bench with default UEFI BIOS settings, the AData AD4U2133W8G15-B memory started up without problems in full accordance with the specification, that is, at a frequency of 2133 MHz with timings of 15-15-15-36 and with a supply voltage of 1.2 V.

Moreover, it turned out that this memory can operate at a frequency of 2400 MHz. When setting this frequency, the timings in automatic mode are set to 16-17-17-40. The best timings that we were able to select for this memory without loss of stability were 14-14-14-36.

Testing

So, in total, five sets of four-channel DDR4 memory took part in our testing, each of which was tested in two operating modes: with default settings and with settings corresponding to maximum overclocking.

First of all, we note that all memory kits, with the exception of Geil Evo Potenza GPR416GB3000C16QC, were defined by default as DDR4-2133 memory with timings of 15-15-15-36. In all our tests, all kits in DDR4-2133 mode with timings of 15-15-15-36 produced almost identical results. And in order not to clutter the article with unnecessary data, in the future we will simply talk about DDR4-2133 memory with timings 15-15-15-36, meaning by it any kit with default settings - with the exception of Geil Evo Potenza GPR416GB3000C16QC memory.

For testing we used a stand with the following configuration:

• Intel Core i7-5960X processor;
• Gigabyte X99-Gaming G1 WIFI motherboard;
• Intel X99 chipset;
• Intel SSD 520 Series (240 GB):
• operating system Windows 8.1 (64-bit).

Performance measurements were carried out using real applications from our iXBT Application Benchmark 2015 test script. We consider the use of synthetic tests, which memory manufacturers love so much, in this case to be simply pointless, since the “parrots” they produce have nothing to do with reality.

From the iXBT Application Benchmark 2015 package, we intentionally excluded tests whose execution speed depends on the data storage subsystem (copy speed, application installation and uninstallation speed, etc.). In addition, the Adobe After Effects CC 2014.1.1 test (Test #2) was excluded. The fact is that for this test, in the case of using an 8-core (16 logical cores) Intel Core i7-5960X processor, it is advisable to use not 16, but 32 GB of memory. Otherwise, the test will be performed without multiprocessing technology, or you need to forcefully reduce the number of processor cores used. In a word, it’s easier to exclude this test, especially since the method includes another test using the Adobe After Effects CC 2014.1.1 application. In addition, we excluded tests that have a large measurement error and require large number repetitions When testing memory, when changing frequencies and timings leads to only a negligible increase in performance, it is very important to use tests in which the result has very good repeatability (with a small measurement error).

As a result, we left the following tests:

• MediaCoder x64 0.8.33.5680,
• Adobe Premiere Pro CC 2014.1,
• Adobe After Effects CC 2014.1.1,
• Photodex ProShow Producer 6.0.3410,
• Adobe Photoshop CC 2014.2.1,
• ACDSee Pro 8
• Adobe Illustrator CC 2014.1.1,
• Adobe Audition CC 2014.2,
• WinRAR 5.11, archiving,
• WinRAR 5.11, unzipping.

So, let's start with a video transcoding test using the MediaCoder x64 0.8.33.5680 application. As we see, this task is not very sensitive to memory speed: the worst result differs from the best by only 6%. It is interesting to note that the Geil Evo Potenza memory at 2667 MHz with timings 13-14-14-30 shows the same result as the Kingston HyperX Predator memory at 2400 MHz with timings 12-12-12-35. And at 2400 MHz (with timings of 16-16-16-35), Geil Evo Potenza memory performs approximately the same as DDR4-2133 memory.

In Adobe Premiere Pro CC 2014.1 we get a similar result. The difference in test execution time between DDR4-2133 and DDR4-2400 memory is approximately 5%. And in this test Geil Evo Potenza memory at 2667 MHz with timings 13-14-14-30 shows the same result as any other memory in DDR4-2400 mode. And at 2400 MHz (with timings of 16-16-16-35), Geil Evo Potenza memory performs approximately the same as DDR4-2133 memory.

In the test based on the Adobe After Effects CC 2014.1.1 application, the difference between the worst and best results is no more than 5%. Once again, Geil Evo Potenza memory at 2667 MHz with timings of 13-14-14-30 shows the same result as any other memory in DDR4-2400 mode. And at 2400 MHz (with timings of 16-16-16-35), Geil Evo Potenza memory performs approximately the same as DDR4-2133 memory.

Photodex ProShow Producer 6.0.3410 is slightly more sensitive to memory speed, and in our test the difference between worst and best results is about 6%. But again, the fastest Geil Evo Potenza memory at 2667 MHz performs the same as any other DDR4-2400 memory, and at 2400 MHz the results of Geil Evo Potenza memory are comparable to the results of DDR4-2133.

Adobe Photoshop CC 2014.2.1 turned out to be insensitive to memory speed. In our test, the difference between the worst and best results was about 3.5%. And again, the “weird” Geil Evo Potenza memory at 2667 MHz performs about the same as any other DDR4-2400 memory, and at 2400 MHz the results of the Geil Evo Potenza memory are comparable to the results of DDR4-2133.

In the test using the ACDSee Pro 8 application, the dependence on memory speed is very insignificant: the difference between the worst and best results was about 1.5%. The Geil Evo Potenza memory did not surprise us with anything pleasant: at a frequency of 2667 MHz it performs approximately the same as any other DDR4-2400 memory, and at a frequency of 2400 MHz the results of the Geil Evo Potenza memory are even slightly worse than the results of DDR4-2133.

In the test using the Adobe Illustrator CC 2014.1.1 application, nothing at all depends on memory speed. Here, for all memory sets in different operating modes, the same results are obtained.

But in the test using the Adobe Audition CC 2014.2 application, the dependence on memory speed, although insignificant, is there: the difference between the worst and best results was 4.8%. For Geil Evo Potenza memory, as in other cases, we get the following: at 2667 MHz it performs slightly worse than any other DDR4-2400 memory, and at 2400 MHz the results of Geil Evo Potenza memory are approximately the same as the results of DDR4- 2133.

In the archiving test using the WinRAR 5.11 application, the difference between the worst and best results was 5.6%. Geil Evo Potenza memory at 2667 MHz performs slightly worse than any other DDR4-2400 memory, and at 2400 MHz Geil Evo Potenza results are about the same as DDR4-2133.

In the unzipping test using the WinRAR 5.11 application, the difference between the worst and best results was 4%. And as always, Geil Evo Potenza memory at 2667 MHz shows results typical of DDR4-2400 memory, and at 2400 MHz results typical of DDR4-2133.

Conclusions

Actually, the conclusions that can be drawn from our testing are quite predictable. There is no particular point in high-speed DDR4 memory today, and the DDR4-2133 option is quite sufficient for most user applications. The maximum performance increase that can be obtained by using high-speed DDR4-2400 memory instead of standard DDR4-2133 is about 5%. Moreover, we did not find any significant difference between modules/kits from different manufacturers.

Moreover, as it turned out, high-speed memory, which is sold under the guise of DDR4-2400, is actually an overclocked version of DDR4-2133 memory, that is, the DDR4-2400 operating mode is implemented only through the XMP profile. And most likely, if you buy the most ordinary DDR4-2133 memory, you can turn it into DDR4-2400. So does it make sense to overpay?

The DDR4-3000 memory (Geil Evo Potenza GPR416GB3000C16QC) turned out to be DDR4-2400 memory, and it simply refused to work at the promised speed of 3000 MHz. In general, the Geil Evo Potenza GPR416GB3000C16QC memory is very strange. In DDR4-2667 mode (the maximum frequency at which it was able to run) it works like DDR4-2400 memory, and in DDR4-2400 mode it works like DDR4-2133 memory. Actually, this is an example for those who think that high-speed memory is cool.

As for the various oddly shaped heatsinks on high-speed memory modules, as we have already said, this is nothing more than a decorative element. Modern memory DDR4, even with the supply voltage increased to 1.4 V, does not require radiators at all.

MOSCOW, October 27 – RIA Novosti, Olga Kolentsova. What is time? Directors of science fiction films believe that this is a kind of dimension through which you can move. IN real world time is determined by the position of objects in space. Theoretically, if we can return every particle in the Universe to the state and position in which it was at a certain moment, we will travel back in time.

So, for now, our knowledge makes it possible to determine time depending on the mechanical changes occurring in the world. For example, one full turn The Earth around its axis is determined by the day, and around the Sun - the year. But people have a need to break the day into smaller and clearly defined segments - hours, minutes, seconds.

To count these units, people came up with special devices- watch. Their history lasts centuries, and along with technology, the requirements for the accuracy of time measurement are growing. If in everyday life we ​​get along well with mechanical and electronic watch, then science requires much more accurate instruments.

The basis for calculating time is a certain repeatable event when an object returns to its initial state after a strictly defined period of time. For example, in a mechanical watch gears rotate (or a pendulum swings), and in hourglass there comes a moment when all the grains of sand fall to the bottom of the vessel.

Of course, modern electronic and mechanical watch much more accurate than their predecessors - water, sand and solar. But some areas required even more precise mechanisms. And people created a clock that worked based on the processes occurring inside the atom.

As you know, an atom consists of a nucleus and an electron cloud. Electrons are located at different energy levels. The further an electron is from the nucleus, the more energy it has. Imagine a dog tied to a steel beam with a strong but stretchable leash. The further she wants to move away, the tighter she needs to tighten the leash. Of course, a strong, large dog will be able to move further than a small and weak one.

© AP Photo/Focke Strangmann

© AP Photo/Focke Strangmann

When moving to a lower level, an electron emits energy, and when moving to a higher level, it absorbs energy. "Jumping" electrons can be controlled using electromagnetic radiation, which is a source of energy. The radiation has a certain frequency. This value is the inverse of the period of oscillation, that is, the time required for an object performing “closed” movements to return to its original state.

Atomic clocks use calcium, hydrogen, thulium, strontium, rubidium, thorium, iodine and methane, and most often cesium. Electrons in a cesium-133-based atomic clock upon transition from one energy level the other emits electromagnetic radiation with a frequency of 9,192,631,770 Hz. It is this number of intervals that a second in this natural clock is divided into. According to the definition officially adopted back in 1967 at the General Conference on Weights and Measures, the cesium-133 atom is recognized as the standard for measuring time. The accuracy of the second determines the authenticity of other basic units physical quantities, such as volts or watts, which are defined over time.

An ultra-precise clock works like this: Cesium-133 is heated, and some atoms leave the main substance, and then pass through a magnetic field, which eliminates atoms with the desired energy states. Selected atoms pass through a magnetic field with a frequency close to the frequency of electromagnetic radiation when an electron passes from one level to another in cesium-133. Under the influence of the field, atoms change energy states and fall on a detector, which records the moment when necessary energy state will have greatest number atoms. Then the frequency value of the electromagnetic field is fed into the frequency divider, which determines its unit by dividing the second. The result is a “new second”, taken as the standard of the minimum unit of time.

© Illustration by RIA Novosti. Alina Polyanina

A sensation has spread around the scientific world - time is evaporating from our Universe! So far this is only a hypothesis of Spanish astrophysicists. But the fact that the flow of time on Earth and in space is different has already been proven by scientists. Time flows slower under the influence of gravity, accelerating as it moves away from the planet. The task of synchronizing the earthly and cosmic time carry out hydrogen frequency standards, which are also called “atomic clocks”.

The first atomic time appeared along with the emergence of astronautics; atomic clocks appeared in the mid-20s. Now atomic clocks have become an everyday thing, each of us uses them every day: with their help, digital communications, GLONASS, navigation, and transport work.

Owners mobile phones hardly think about what difficult work in space it is carried out for strict time synchronization, but we are talking about only millionths of a second.

The exact time standard is stored in the Moscow region, in Scientific Institute physical-technical and radio-technical measurements. There are 450 such watches in the world.

Russia and the USA have monopolies on atomic clocks, but in the USA clocks operate on the basis of cesium - radioactive metal, very harmful to the environment, and in Russia - based on hydrogen - a safer, durable material.

This watch does not have a dial or hands: it looks like a large barrel of rare and valuable metals, filled with the most advanced technologies - high-precision measuring instruments and equipment with atomic standards. The process of their creation is very long, complex and takes place in conditions of absolute sterility.

For 4 years now, the clocks installed on the Russian satellite have been studied dark energy. By human standards, they lose accuracy by 1 second over many millions of years.

Very soon an atomic clock will be installed on Spektr-M - space observatory, which will see how stars and exoplanets are formed, will look beyond the edge black hole in the center of our Galaxy. According to scientists, due to the monstrous gravity, time flows so slowly here that it almost stops.

tvroscosmos