The GeForce RTX 4090 was announced by NVIDIA recently, and has been drawing a lot of attention for its power, price, and size. In a benchmark test performed in the 3DMark application , it was revealed that the RTX 4090 can be up to 82% faster than its predecessor RTX 3090 , an impressive result given that the 3090 is still one of the most powerful GPUs on the market.
The RTX 4090 is carrying an AD102 GPU with 16384 CUDA cores, 24 Gbps of GDDR6X memory and operating at a clock speed of 2520 MHz. But after all, how was the comparison between the two boards made?
Test comparison between RTX 4090 and 3090 GeForce
Using 3DMark, the video card entered a test session in three different applications of the application, namely FireStrike Ultra, TimeSpy Extreme and Port Royal . In this case, each application serves to perform stress tests on the video card, working as follows:
- FireStrike Ultra creates a scene in 4K resolution and running DirectX 11
- TimeSpy Extreme creates a scene in the same resolution as before, but this time running in DirectX 12
- Port Royal is based on a custom engine with DirectX Ray Tracing support.
Each application monitors the performance of the video card in different stress situations and counts points that serve to compare how well it did. In the test performed, the 4090 reached 18,892 points on TimeSpy, 25,256 on FireStrike and 24,886 points on Port Royal, breaking records that were previously held by NVIDIA’s 3090.
When comparing the two video cards, the result looks like this:
- FireStrike Ultra: 2.00 to 2.04x faster (100% to 104%)
- TimeSpy Extreme: 1.84 to 1.89x faster (84% to 89%)
- Port Royal: 1.82 to 1.86 x faster (82 to 89%)
It’s important to note that the high values happened because the data was compared to the average of the 3090 taken in 3DMark. As the RTX 4090 reaches the market and the homes of enthusiasts, tests performed on different games can end up narrowing this difference.
Comparison in tables GeForce
To facilitate the visualization of the comparisons, several tables were also generated that compare the two high-end NVIDIA cards through the results of their tests in 3DMark, check it out:
We can see that all the models of the new RTX 40 line have excelled compared to the previous models from NVIDIA and also from AMD. The new line, which will be the first with the Ada Lovelace architecture, also brings new Ray Tracing and DLSS 3.0 cores to several games.
Source: Videocardz
More:
- Xiaomi 12T, 12T Pro, Redmi Pad and more devices on sale!
- Top 15 of the best Resident Evil games to know and play this year
- BGS 2022 Intel talks about 13th generation for the year 2023
What is the difference between GeForce GTX and RTX?
Technical differences
The first is in the number of CUDA cores. While the RTX 2060 has 1,920 cores, running at a clock between 1,365 and 1,680MHz, the GTX 1660 Ti has slightly fewer cores, with 1,536 units running at a clock between 1,500 and 1,770MHz
Which GTX has Ray Tracing?
Nvidia has just announced that it is bringing ray tracing support to more cards, requiring at least a GTX 1060 with 6GB of RAM.
What is the best video card for PC gaming?
- Radeon RX 6900 XT.
- Nvidia RTX 3070.
- Nvidia RTX3060 Ti.
- GeForceRTX 3050.
- AMD Randeon RX 6600.
- GeForce RTX 2060.
- NVIDIA Geforce GTX 1660 Super.
- Randeon RX 550.
What is the best RTX value for money?
“Evolution” of the previous GPU, the RTX 3060 Ti is one of the best value for money models today . Unlike its smaller sister, this RTX graphics card is already indicated for the public that wants to aim at Quad HD resolution with very high quality graphics or Full HD with Ray Tracing and higher frame rate.
