Intel x299 overclocking guide: for intel skylake-x and intel kaby lake processors

Intel X299 Overclocking Guide | The "Silicon Lottery"

A first point that we must take into account when overclocking any processor is that no two processors are exactly the same , even if they are the same model. Processors are made from thin silicon wafers, and with manufacturing processes like Intel's current 14nm, transistors are about 70 atoms wide. Therefore, any minimal impurity in the material can dramatically worsen the behavior of the chip .

Manufacturers have long taken advantage of these failed models, using them at lower frequencies, or disabling some of the worst-performing cores to sell it as an inferior processor. For example, AMD manufactures all its Ryzen from the same DIE, and Intel in the high-end socket (HEDT) usually does the same.

But it is that even in the same model there are variations, for this same reason. A processor that has come out almost perfect from the process will reach 5 Ghz with very little extra voltage, while one of the "bad guys" will barely rise 200mhz from its base frequency without the temperatures soaring. For this reason it is useless to search for an overclock and what voltage is necessary on the internet, since your processor is not the same (not even the same "batch", or BATCH) as that of the user who is publishing their results.

The most optimal overclocking for each chip is obtained by increasing the frequency little by little, and looking for the lowest possible voltage in each step.

What do we need before we start?

You have to follow these four essential points before entering the world of overclocking:

Lose the fear of crashes and blue screenshots. Let's see a few. And nothing happens. Update the motherboard BIOS to the latest available version. Clean our refrigeration, fans and radiators, changing the thermal paste if necessary. Download Prime95, to test stability, and HWInfo64, to monitor temperatures.

Terminology

In this guide we will limit ourselves to modifying simple parameters, and we will try to simplify the steps as much as possible. However, we will briefly explain some concepts, which will help us understand what we are doing.

Multiplier / Multiplier / CPU Ratio: It is the ratio between the clock frequency of the processor and that of an external clock (usually the bus or BCLK). This means that for each cycle of the bus to which the processor is connected, the processor has performed as many cycles as the value of the multiplier. As its name suggests, multiplying the speed of the BCLK (100Mhz series on this platform, and on all recent ones from Intel) by the multiplier gives us the processor's working frequency.

That is, if we put a multiplier of 40 for all the cores, our processor will work at 100 x 40 = 4, 000 Mhz = 4Ghz. If we put a multiplier of 41 in the same processor it will work at 100 x 41 = 4, 100 Mhz = 4.1Ghz, with which we have increased the performance (if it is stable) by 2.5% compared to the previous step (4100/4000 * 100). BCLK or Base clock: It is the clock at which all the chipset buses, the processor cores, the memory controller, the SATA and PCIE buses work… unlike the main bus of previous generations, it is not possible to increase it beyond a few few MHz without having problems, so the usual thing is to keep it at 100Mhz that is used as standard and to overclock using only the multiplier. CPU Voltage or Core Voltage: Refers to the voltage that the processor core receives as power. It is probably the value that has the most impact on the stability of the equipment, and it is a necessary evil. The more voltage, the more consumption and heat we will have in the processor, and with an exponential increase (against the frequency, which is a linear increase that does not worsen the efficiency by itself). However, when we force the components above the frequencies specified by the manufacturer, many times we will have no choice but to slightly increase the voltage to eliminate the failures that we would have if we only increased the frequency . The more we can lower our voltage, both stock and overclocked, the better. Offset Voltage: Traditionally, a fixed voltage value was set for the processor, but this has the great disadvantage that, even without doing anything, the processor is consuming more than necessary (far from its TDP, but wasting a lot of energy anyway).. The offset is a value that is added (or subtracted, if we seek to reduce consumption) to the serial voltage of the processor (VID) at all times, such that the voltage continues to drop when the processor is idle, and at full load we have the voltage we need. By the way, the VID of each unit of the same processor is different. Adaptive Voltage: Same as the previous one, but in this case instead of adding the same value at all times, there are two offset values, one for when the processor is idle, and the other when the turbo boost is active. It allows a very slight improvement in the idle consumption of an overclocked equipment, but it is also more complicated to adjust, since it requires many trial and error tests, and the idle values are more difficult to test than those of turbo, since with low load even an unstable system has little chance of failure.

First steps of overclocking

These processors feature a slightly improved version of Turbo Boost Technology 3.0 that debuted in Haswell-E. This means that when two or fewer cores are in use, tasks are assigned to the cores that the board identifies as best (since not all silicon is equally perfect, and some could support higher frequencies) and the turbo frequency. boost is raised to a much higher value than usual. In the case of the Intel Core i9-7900X, this Boost for two cores is 4.5Ghz.

Before we start, let's discuss the equipment we have used:

Corsair Obsidian 900D.Intel Core i9-7900X.Asus Strix X299-E ROG. 16 GB DDR4 memory. Hanging prime95 (most common) or some other program that is running in the background, but the operating system is still working.

The entire pc hangs, either freezing, with a blue screen, or with a sudden restart / shutdown.

In any of these cases, what we will do is raise the offset slightly, with small steps, around 0.01V more each time, and try again. We will stop rising when temperatures rise too high (more than 90º in extreme tests) or when the voltage approaches dangerous levels. With air cooling, we should not go from 1.3V for all cores, 1.35 maximum with liquid. We can see the total voltage value with HWInfo, since the offset is only what is added and not the final value.

What to do if the equipment is stable

In case our system is more or less stable , we will stop it after approximately 10 minutes with the option that we have seen above. We say "more or less" since in 10 minutes we will not be able to know for sure. After stopping the tests, we will see a screen like the following one, with all the workers (the work blocks that run in each core) finishing correctly. We look at the boxed part, all tests must have ended with 0 errors / 0 warnings. The number of tests that have finished may vary, because the processor is doing other things while running prime95, and some cores may have had more free time than others.

This is the ideal case, as it means that we have multiplier and offset settings that we can test with a longer stability test, and that improve the standard performance of the processor. At the moment, if our temperatures are not high, we write them down and keep increasing the frequency, in the next section, to return to the last stable value when we reach a point where we cannot go up.

We keep going up

In case a quick test like the previous ones has been stable and our temperatures are at acceptable values, the logical thing is to keep increasing the frequencies. To do this, we will increase the multiplier by another point, to 46 in our 7900X:

As the previous stability test has been passed without raising the voltage (we remember that each processor is different, and it might not be the case in your specific processor), we keep the same offset. At this point we pass the stability tests again. If it is not stable, we raise the offset slightly, from 0.01V to 0.01V (other steps can be used, but the smaller, the better we will adjust). When it's stable, we keep going up:

We pass the stability tests again. In our case we have needed an offset of + 0.010V for this test, being as follows:

After leaving it stable, we raise the multiplier again, to 48:

This time we have needed an offset of + 0.025V to pass the stability test successfully.

This configuration has been the highest that we have been able to maintain with our processor. In the next step, we raised the multiplier to 49, but as much as we increased the offset, it was not stable. In our case we have stopped at + 0.050V offset, since we were dangerously close to 1.4V and almost 100ºC in the vaguer cores, too much for it to make sense to continue rising, and more in an overclock thinking of 24/7.

We take advantage of that we have touched the ceiling of our microprocessor to test with lower offset values for AVX instructions, down from 5 to 3. The final frequency for all cores are 4.8Ghz and 4.5Ghz on AVX, which is an increase of approximately 20% compared to stock frequencies . The necessary offset, again in our unit, has been + 0.025V.

Advanced overclocking

In this section we are going to test the possibilities of per core overclocking, keeping the Turbo Boost 3.0 technology active and trying to scratch additional 100-200mhz in the two best cores without increasing the voltage. We say advanced overclock because we multiply the possible tests, and there is much more time for trial and error. These steps are not essential, and at best they will only bring us improvements in applications that use few cores.

We are not going to discuss the voltage increase in other parameters related to the memory controller or the BCLK, since usually the limitation will be the temperatures before reaching frequencies that make it necessary to play nothing else, and the competition overclock with extreme cooling is left out the scope of this guide. Furthermore, as the professional overclocker der8auer mentioned, the phases of a mid / high-end motherboard of this socket may be insufficient for the consumption of an i9 7900x (or even its younger siblings) raised well above its stock frequency.

First of all, it is interesting to comment on one of the advantages of this boost 3.0 technology, and that is that the board detects the best cores automatically, that is, those that need less voltage and apparently will be able to increase their frequency. We note that this detection may or may not be correct, and that on our board we can force the use of other cores, and choose the voltage for each one. In our processor the board tells us, as we anticipated when seeing the information from HWInfo, that the best cores are # 2, # 6, # 7 and # 9.

We can corroborate this choice in the Intel Turbo Boost Max Technology 3.0 Application program, which will have been installed automatically through windows update, and is minimized in the taskbar, since these cores will be the first, and will be the ones that are They will send the tasks that are not parallelized when possible.

In our case it seems logical to try to raise the two best cores to 4.9Ghz first, 100mhz more than what all the cores hold. To do this, we changed the CPU Core Ratio option from XMP to By Core Usage . Next, the Turbo Ratio Limit # values will appear, which allow us to choose the multiplier for the fastest core (0 for the fastest, 1 for the second fastest, etc.), as well as the Turbo Ratio Cores # option, which will allows you to choose which will be the nucleus that we want to upload, or leave it in Auto, in such a way that the board will use the detection that we have seen in the previous step to determine which are the fastest nuclei

To do this we set the values of the Turbo Ratio Limit 0/1 to 49, which will put the two fastest cores at 4.9Ghz. The rest of the Turbo Ratio values we leave at 48, since we know that all the other cores work well at 4.8Ghz.

The way to test stability is the same, although now we must be careful to launch only 1 or 2 test threads, since if we put more the processor will work at the usual turbo frequency. For this we choose only one thread on the screen that we already know from Prime95:

It is convenient to check in the task manager that the work is being assigned to the correct cores (we count 2 graphics per core, since with hyperthreading each 2 threads is a physical core, and in Windows they are ordered together), as well as the frequency is what we expect at HWInfo64. Below we can see the core # 6 at full load, and how the frequency is at 5Ghz.

I personally have not had much success using the above method, even with a little extra voltage , although each processor is different and may be different for someone else. The result seen in the previous screenshot has been achieved using the manual option, with which we have been able to upload a couple of cores up to 5Ghz. With this mode we can choose the voltage and the multiplier for each nucleus, so we can give a high voltage, around 1.35V, to the highest nuclei, without exacerbating the TDP excessively or uncontrolling our temperatures. Go for it:

First we choose the By Specific Core option

A new screen opens for us to open. On this new screen, setting all Core-N Max Ratio values to 48 with the rest in Auto would leave us the same as in the previous steps, at 4.8Ghz all cores. We will do that, except in two of the best cores (7 and 9, marked with * on the plate, and two of the four that we had identified as best), which we will test with 50 (in the screenshot we can see 51, but this value did not work correctly)

As a suggestion, although the voltage in Manual Mode is faster to adjust to the value we want, it would be more correct to do the same with Offset, testing until obtaining the desired VID.

The gain on tasks that only use one core is noticeable. As a quick example, we have passed the popular Super Pi 2M benchmark, obtaining a 4% improvement in test time (less is better), which is expected with this frequency increase (5 / 4.8 * 100 = 4.16%).

4.8Ghz

5Ghz

Final steps

Once we have found a configuration that convinces us, it is time to test it thoroughly, since it should not only appear stable for 10 minutes, it should be stable for several hours . In general, this configuration will be the one immediately prior to the one we were at when we hit the ceiling, but in some processors it will have to lower 100mhz more if we do not get it to be stable. Our candidate is 4.8Ghz at + 0.025V Offset.

The process to follow is the same as in the stability tests we have done, only now we must leave it for several hours. From here we recommend about 8 hours of Prime95 to consider a stable overclock. Although I personally have not observed temperature problems in the phases of the Asus X299-E Gaming board, it is advisable to make short breaks of 5 minutes approximately every hour so that the components can cool down.

If we have the possibility to measure the temperatures of the phases, we can skip this step. In our case we see that, after 1 hour of prime, the heatsink is around 51ºC. If we don't have an infrared thermometer, we can carefully touch the top heatsink on the motherboard. The maximum temperature that can be held without removing the hand by the hair, is about 55-60ºC for a normal person. So if the heatsink burns but can hold, we are at correct margins.

The screen we want to see is the same as before, all workers stopping, with 0 warnings and 0 errors. In our case we had an error after 1 hour of testing, so we raised the offset slightly, up to + 0.03V, which is the minimum that allowed us to finish the test correctly.

What do you think of our overclocking guide for LGA 2066 socket and X299 motherboards? What has been your stable overclocking with this platform? We want to know your opinion!