Skip to content

Best gaming mice under 30

In this article, we will compare the performance of Nvidia's standard 2070 GPU and 2070 Super, the latter aims to further improve from the standard model.

We will compare various performance indicators to help you determine whether one is definitely better than the other, and if so, which one is. Elements such as its architecture, VRAM and resolution will determine which of the two is more worth buying.

clock speed

1620 MHz

VRAM

8GB GDDR6

memory bus width

256 bit

clock speed

1815 MHz

VRAM

GDDR6 8GB

4 444 444 bit memory bus speed is better, hope

4Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit

Bit solves. If not, you need to answer questions about how to justify its name and price.

In this comparison, it is worth knowing how the Super version performs on many indicators. This allows players to judge for themselves whether upgrading to the super version is worthwhile for them.

The greater the performance gap between the standard 2070 and Super 2070, the more likely the Super is to attract gamers to buy. If the gap is relatively small, gamers can choose to keep their GPU funding until the Super price drops, or they can look for high-end graphics cards that can improve the playability they need. Chapter

The game begins!

2070 and 2070 Super are based on Nvidia's relatively new Turing architecture.

This means that in terms of basic architecture, in terms of thought and design protocols, there is very little that can separate them.

What is the problem with the Turing architecture itself?

is quite large, yes, for two reasons.

First, it separates the overall workload from the GPU, allowing AI functions to be handled by a dedicated tensor core and ray tracing functions to be handled by a dedicated RT core. Chapter

What Good Is This To You? Dedication, focus, flow and speed. If a dedicated single core is responsible for individual functions, there will be fewer linear errors in image processing and gameplay. Increase fluidity and smoothness.

Second, Turing architecture brings Turing NVENC to your party.

This is a new and faster data encoding method, so the transmitter does not need to buy a dedicated transmitter. Speed, processing power, fluency and the potential to turn heads - this is the promise of the Turing architecture.

The flaw with this plan is that the games industry has yet to fully embrace the Turing architecture and its opportunities, which means that relatively few games actually get the most benefit from the architecture. This is equivalent to owning a sports car and driving at a speed of 50 years. Taking full advantage of the Turing architecture, there will be games that will amaze you. They just haven't turned up yet; This means that buying Turing GPUs now is actively looking to the future or spending money on technology that you can't use yet.

In short, 2070 Super seems to have brought a significant improvement in both basic clock and boost, but the memory clock speed is the same. In addition, if you do the math, the significant gain feels less significant—the base clock speed is increased by approximately 14%, and the accelerated clock speed is increased by approximately 9%.

As we compare the elements of the infrastructure, let's do a quick core comparison to see what your money brings to you in terms of Turing architecture.

2070 Super

Like the basic speed and boost results, the central super number has also increased significantly. But doing math calculations, it will show that the tensor and RT core have only increased by 11%.

Since they are based on the same architecture and are from the same manufacturer, the fans and cooling options on these cards are also very similar. When the

is sold, the 2070 should be able to run between 7080 degrees (the official maximum number, 79 degrees) under heavy load, while the 2070 Super rarely exceeds 71 degrees, even under dense loads (the official record is the most large number, 74 degrees). Spend). This will have different factors depending on your configuration build, internal airflow, and whether you installed an additional fan or water cooling system when you installed the card.

If you want to overclock your system, the lower you can keep the GPU temperature, the more space you will need to increase it. Aftermarket corner fans and water cooling systems can help you lower the maximum temperature of equipment and systems. Although the maximum fan noise sold by the 2070 is 47.7dB, the 2070 Super is reduced by 2.1dB to only 45.6dB. The

2070 has a TDP of 175W, and the 2070 Super has a TDP of 215W.

Be sure to determine the size of your card before paying, sending, and trying to place it. Doing something wrong is fundamentally a problem, not to mention a bit embarrassing.

As mentioned above, Super is designed as an upgrade to 2070, so it is not surprising that they are the same size.

Having said that, in terms of power connectors, although they all use eight-pin connectors, Super also requires an additional six-pin connector.

Let's talk about graphics turkey. Everyone wants to know where the power goes in terms of resolution. let's see.

As expected, the 2070 Super is always ahead of 2070. We know or expect this to happen. What might surprise you is how small this advantage is by certain standards. NBody's calculation speed under the basic clock is only 6 fps per second; when overclocked it is only 3fps with the same gauge.

In terms of ray tracing, 2070

Two-card Turing architecture.

The irony is that compared to ray tracing, you get more differences at the core of the AI, which may be more directly useful to you. Now look at the price of each card and ask yourself if you are willing to pay the difference for the other four RT cores that are currently useless.

is similar to 2070 and 2070 Super in VRAM. Both provide you with 8GB GDDR6 memory and 448GB / s memory bandwidth.

This means that both graphics cards allow you to play 1080p games at exactly the same level and speed; This is the lowest smooth and effective 1080p gaming speed, and there is hardly any room to adapt to the future.

Think about it. There is no difference between the 2070 and 2070 Super standard in VRAM. Chapter

Shouldn't There Be? Considering that the 2070 Super is an upgrade card, the performance of its actual upgrade from the standard card has been surprisingly low: here it is 11%, there is 9%. In areas that may eventually be separated, the 2070 sits comfortably, congested.

clock speed

1620 MHz

VRAM

8GB GDDR6

memory bus width

256-bit

clock speed

1815 MHz

VRAM

GDDR6 8GB 4

two GPU

4 memory will depend on these

4 memory where you are now And how much money you have in your pocket. If you're upgrading from a significantly lower or slower graphics card, the question is whether Super offers enough incentive to spend money.

Overall, this seems unlikely - Super's current areas of expertise haven't translated into very practical advantages. Of course, when the industry creates games that use the Turing architecture to their full potential, the additional RT cores will at least play a role. But before that, card makers can launch new variants that will bring the price of 2070 Super down to a more suitable level for gamers.

Of course, if you want to upgrade now, nothing can and should stop you from getting more powerful cards. But if it's our money, we might feel like the upgrade is a little less than Super and the price is a little more than Duper.

Previous article Flight simulator intro games hardware

Leave a comment

Comments must be approved before appearing

* Required fields