3DMark benchmarks of the upcoming AMD Radeon RX 7600 & NVIDIA GeForce RTX 4060 Ti 8 GB graphics cards have been leaked by Videocardz.
AMD Radeon RX 7600 Up 39% Faster Than RX 6600, NVIDIA GeForce RTX 4060 Ti Up To 14% Faster Than RTX 3060 Ti In Leaked Benchmarks
The AMD Radeon RX 7600 and NVIDIA GeForce RTX 4060 Ti graphics cards are both configured with 8 GB memory and will be offered as brand-new mainstream solutions to gamers. Although the GPU used by both chips are relatively mid to low-end, they cannot be compared directly since one of the cards costs $399 US (4060 Ti) while the other is expected to retail close to $299 US (RX 7600). These benchmarks only show what to expect from the graphics cards in 3DMark synthetic benchmarks.
Once again, we should inform our readers that synthetic benchmarks don't reflect actual graphics performance in gaming. Before talking benchmarks, let's take a quick recap of the specs on both of these cards.
NVIDIA GeForce RTX 4060 Specifications
For the GeForce RTX 4060, NVIDIA is expected to utilize the AD107-400-A1 GPU, the fully enabled AD107 Ada die. Do note that the same configuration is used by the GeForce RTX 4060 Laptop GPU. The chip is expected to feature 3072 CUDA cores and 8 GB of GDDR6 memory running at 17 Gbps memory speeds across a 128-bit bus interface. That gives the card 272 GB/s bandwidth which is effectively rated at 453 GB/s or 26% more than the RTX 3060. That's all thanks to the 24 MB of L2 cache.
The NVIDIA GeForce RTX 4060 should end up with a 35% lower TGP than its predecessor, the RTX 3060 12 GB. The card is said to feature a 115W TGP but actual gaming power consumption should be around 110W with idle power rated at 7W and video playback power rated at 11W. The card will be available in various custom models at launch but there will be no Founders Edition variant for it.
AMD Radeon RX 7600 Specifications "Rumored":
The AMD Radeon RX 7600 will feature the Navi 33 XL GPU and is configured with 8 GB of GDDR6 memory across a 128-bit bus interface. This will be the third and only chip of the RDNA 3 lineup that utilizes a monolithic design. The memory specs are identical to the NVIDIA GeForce RTX 4060 Ti 8 GB, with a max bandwidth of 288 GB/s (20 Gbps). The GPU will offer 32 Compute Units (CU) with 2048 Stream Processors (shown in the screenshot as Unified Shaders), the exact core count of the Navi 23 GPU.
- AMD Navi 33: 2048 Cores, 128-bit Bus, 32 MB Infinity Cache, 204mm2 GPU Die @6nm
- AMD Navi 23: 2048 Cores, 128-bit Bus, 32 MB Infinity Cache, 237mm2 GPU Die @7nm
Image Credits: Videocardz
The new AMD Radeon RX 7600 graphics card has a PCIe 4.0 x8 interface, identical to the NVIDIA GeForce RTX 4060. The GPU clock and memory clocks display 2250 MHz (GPU), 2250 MHz (base memory), and 2655 MHz (boosted memory), but the card has an actual peak of 2.85 GHz
AMD Radeon RX 7600 & NVIDIA GeForce RTX 4060 Ti 3DMark Benchmarks
Starting with these leaked benchmarks, we first have 3DMark Time Spy (DX12) which was tested at 4K and 1440P resolutions. None of these graphics cards is designed to run 4K gaming and even 2K (1440P) is only recommended with upscaling techniques or by lowering the visual quality due to the limited memory and bus interface these cards support. The NVIDIA GeForce RTX 4060 Ti offers a 14 percent boost over the RTX 3060 Ti and the RX 7600 offers a 31% boost in Time Spy. At 4K, the 4060 Ti delivers a 9% boost over the 3060 Ti, and the RX 7600 delivers a 36% boost.
The NVIDIA GeForce RTX 4060 Ti takes a bigger hit at 4K than its predecessor due to its limited 128-bit bus interface whereas its predecessor had a 256-bit bus. Although NVIDIA has stated that large L2 cache accommodated by the 4060 series should overcome the bandwidth bottlenecks, synthetic benchmarks don't seem to make use of that well so we will see if gaming performance is better. In the next benchmark, we have 3DMark Speed Way, a pure ray tracing benchmark where NVIDIA has a big lead.
Performance Gains Gen Over Gen (RTX 4060 Ti vs RTX 3060 Ti):
- 3DMark Time Spy (1440P): +14%
- 3DMark Time Spy Extreme (2160P): +10%
- 3DMark Speed Way (1440P): +8%
- 3DMark Fire Strike (1080P): +14%
- 3DMark Fire Strike Extreme (1440P): +10%
- 3DMark Fire Strike Ultra (2160P): +1%
- Average Improvement: +9.5%
The NVIDIA RTX 4060 Ti sat at 8% faster than the RTX 3060 Ti while the RX 7600 delivered a 39% boost over the RX 7600.
Next up, we have 3DMark Fire Strike which runs well with AMD's graphics hardware. Once again, the NVIDIA GeForce RTX 4060 Ti delivers a 14% performance boost at 1080p over the 3060 Ti but that difference starts to decrease as the resolution preset is increased and the cards offer virtually the same performance at 4K. The same is the case with the RX 7600 but it does end up close to the 4060 Ti at 4K.
Performance Gains Gen Over Gen (RX 7600 vs RX 6600i):
- 3DMark Time Spy (1440P): +31%
- 3DMark Time Spy Extreme (2160P): +37%
- 3DMark Speed Way (1440P): 39%
- 3DMark Fire Strike (1080P): +31%
- 3DMark Fire Strike Extreme (1440P): +33%
- 3DMark Fire Strike Ultra (2160P): +33%
- Average Improvement: +34%
Note: There's no mention if the cards used here were reference models or overclocked variants. An overclocked 3060 Ti versus a reference 4060 Ti can skew the benchmarks by a big margin.
Based on these results, it looks like the AMD Radeon RX 7600 8 GB graphics cards will be highly competitive against the RTX 4060 Non-TI 8 GB in older DX11 titles but the 4060 Non-Ti should have an advantage in newer DX12 and Ray Tracing titles. The RTX 4060 Ti is expected to be 15% faster than the RTX 3060 Ti and the RTX 4060 Non-Ti is expected to be 20% faster than the RTX 3060 without the use of DLSS.
source https://wccftech.com/amd-radeon-rx-7600-nvidia-geforce-rtx-4060-ti-8-gb-3dmark-benchmarks-leak-out/