beat the competition in single-wire performance

beat the competition in single-wire performance

The battle that the processors offered by Intel and AMD are about to wage for the 2022-2023 school year will be fierce. And it is that this new generation will put on the table two very different strategies which have already reached full maturity. On the one hand AMD brandishes Ryzen 7000 which promise us to be competitive if we stick to their brute force, but which also aspire to stand out for their performance per watt.

Intel, however, took a different path over the last generation with its Alder Lake microarchitecture. Its new 13th Gen Core processors with Raptor Lake microarchitecture reinforce its commitment to a hybrid architecture which combines high-performance (AR) cores and high-efficiency (AE) cores, and openly aspires to dominate their competition based on raw performance.

Soon we will have the opportunity to analyze these two families of microprocessors in depth, and, of course, when we do, we will share with you our conclusions and we will check together if both brands keep their promises. Still, AMD’s Ryzen 7000s and Intel’s 13th-gen cores have one important thing in common: both families have clearly raised the bar for the maximum clock rate they’re capable of running at. And they did it with an ambitious goal.

The 6 GHz barrier (without overclocking) is just around the corner

The performance of a microprocessor is conditioned by a wide range of parameters. The clock frequency at which you work has a profound impact on your productivity, but there are other factors which are also very relevant. The work that the CPU is able to do in each cycle of the clock signal is also decisive and, interestingly, is described by a large number of microarchitecture characteristics.

CPI (Instructions per cycle) reflects the average number of instructions that the processor executes with each cycle of the clock signal. And this is very important. The cache memory also has a direct impact on performance, but not only its size and the number of levels are relevant; The strategies it uses both to update the contents of the sub-cache levels and to maintain consistency between it and main memory matter a lot. But it doesn’t all end there.

The performance of the CPU is also conditioned by its number of cores, the implementation of SMT technology and other essential characteristics of the microarchitecture, such as the number of execution units, the strategy of branch prediction algorithms or even processor CPU design. execution channel (pipeline). It is not necessary for us to know in detail what these components consist of, but all this reminds us that the performance of a processor is conditioned by many parametersand not just because of the clock frequency at which it operates.

AMD and Intel of this generation have both set foot on the clock frequency accelerator

Even so, there is a context in which the maximum clock rate at which one of the processor’s cores is capable of operating has a monumental impact on its performance: the single-threaded execution scenario. Games take better advantage of having multiple cores in the CPU, but most games still rely heavily on the performance that a single core can deliver when powered on.

In any case, there’s no doubt that this generation’s AMD and Intel have fully set their foot on the clock rate accelerator, and they’ve done so with every intention of beating their competitor in a scenario single-threaded execution. Moreover, the increase in frequency that they offer us is far from being timid. Completely the opposite. Ryzen 7000 and 13th Gen Core can run at maximum clock speed significantly higher than its predecessors.

In the following table, we’ve gathered the 13th Gen Intel Core specs we’re interested in for this article. The base clock rate of AR and AE cores is similar in Alder Lake and Raptor Lake, however, the maximum clock rate is noticeably higher in newer Intel processors. Indeed, as we can see, the Core i9-13900K is capable of reach 5.8 GHz. This is not but nothing bad.

Intel Core i9-13900k

Intel core i7-13700k

Intel core i5-13600k

series

Raptor Lake-S

Raptor Lake-S

Raptor Lake-S

photolithography

Intel 7

Intel 7

Intel 7

total nuclei

24

16

14

high performance (ar) cores

8

8

6

high yield cores (ae)

16

8

8

son

32

24

twenty

maximum clock frequency

5.8GHz

5.4GHz

5.1GHz

turbo boost max frequency 3.0

5.7GHz

5.4GHz

Not available

ar base frequency cores

3GHz

3.4GHz

3.5GHz

maximum frequency of ar nuclei

5.4GHz

5.3GHz

5.1GHz

ae base frequency cores

2.2GHz

2.5GHz

2.6GHz

nuclei of maximum frequency ae

4.3GHz

4.2GHz

3.9GHz

AMD has also taken this route. And he did it with the same conviction as Intel. In fact, their new Ryzen 7000 is much faster if we stick to its clock frequency than the Ryzen 5000. The base clock frequency of the latter in no way exceeds 4 GHz, while this parameter in several Ryzen 7000s reaches 4.7 GHz. And if we look at the maximum clock frequency, the 4.9 GHz of the Ryzen 9 5950X pales in comparison. compared to 5.7 GHz of the Ryzen 9 7950X.

amd ryzen 9 7950x

AMD RYZEN 9 7900X

AMD RYZEN 7 7700X

AMD RYZEN 5 7600X

series

AMD Ryzen 7000 (Zen 4)

AMD Ryzen 7000 (Zen 4)

AMD Ryzen 7000 (Zen 4)

AMD Ryzen 7000 (Zen 4)

photolithography

FinFET 5nm TSMC

FinFET 5nm TSMC

FinFET 5nm TSMC

FinFET 5nm TSMC

total nuclei

16

12

8

6

son

32

24

16

12

maximum clock frequency

5.7GHz

5.6GHz

5.4GHz

5.3GHz

base clock frequency

4.5GHz

4.7GHz

4.5GHz

4.7GHz

The cards are on the table. and upside down On paper, Intel and AMD both offer us competitive processors, but our verdict will come when we test them thoroughly and in both a single-threaded execution scenario and in a demanding multi-threaded context. . We’ll see what happens when the time comes. The swords are high and the users are happy that these two brands offer us truly competitive solutions. Hopefully both live up to the expectations they raise.

Leave a Comment

Your email address will not be published. Required fields are marked *