Tag: Geekbench

August 6, 2025
Samsung Galaxy A57 Leaked: New Chipset and Xclipse 550 GPU
Key Takeaways
1. The Galaxy A57 is rumored to be in development, following the Galaxy A56’s launch in early March 2023.
2. The Galaxy A57 will feature the Exynos 1680 processor, which has been spotted in a prototype on Geekbench.
3. The Exynos 1680 has 8 CPU cores arranged in three clusters, similar to its predecessor, the Exynos 1580.
4. The Exynos 1680 includes an Xclipse 550 GPU, which has 2 Compute Units, compared to the 4 in the previous model.
5. Performance of the Exynos 1680’s GPU is nearly comparable to the Xclipse 540, but there are no indications of an early release for the Galaxy A57.
The Galaxy A56 has been on the market for less than half a year, having launched in early March along with the more affordable Galaxy A36, which is currently priced at $374.99 on Amazon. However, rumors about a possible successor, the Galaxy A57, began to circulate this past spring.
New Information on the Galaxy A57
As reported by GalaxyClub in May, Samsung is planning to unveil the Galaxy A57, equipped with its own Exynos 1680 processor. Recently, a prototype featuring this chipset was spotted on Geekbench. For now, only a single OpenCL score has been made public, and we’ve included the evidence below.
Details on the Exynos 1680
The listing ‘Samsung Electronics Co., Ltd. Full Android on S5E8865 ERD’ directly connects to ‘S5E8865’, which is the codename for the Exynos 1680. From what we can derive from the listing, the Exynos 1680 consists of 8 CPU cores organized into three clusters, similar to the previous model. However, it’s best not to pay too much attention to the CPU clock speeds, as most of them are significantly lower compared to the Exynos 1580.
GPU Specifications and Performance
In addition, the same listing shows that the Exynos 1680 comes with an Xclipse 550 GPU, as reported by GalaxyClub. Currently, Geekbench indicates that this GPU has 2 Compute Units (CUs) compared to the four found in the earlier version. Still, the prototype’s performance is nearly on par with the Xclipse 540 when evaluating GPU capabilities in OpenCL. At this moment, Samsung’s usual release patterns imply that this could either be a prototype for the Galaxy A57 or a development board. There’s no sign yet that an earlier release is in the works.
Source:
Link

Tags: Exynos 1680, Galaxy A57, Geekbench
July 26, 2025
Nvidia N1X GPU Specs Leaked in Geekbench Listing
Key Takeaways
1. The Nvidia N1X laptop processor shows impressive performance but does not fully compete with AMD and Apple, as it is an engineering sample.
2. The N1X GPU scored 46,361 points on Geekbench OpenCL, similar to an RTX 2050 laptop GPU, operating at a maximum frequency of 1,048 MHz.
3. The GPU features 48 Streaming Multiprocessors (SMs), potentially providing 6,144 CUDA cores, placing it between the RTX 5070 Ti and RTX 5080.
4. The laptop includes 128 GB of RAM, with 64 GB allocated to the GPU, and runs on Windows 11 Enterprise.
5. The N1X’s release may be delayed until 2026, with further updates expected through leaks.
An earlier Geekbench entry gave us insights into how Nvidia’s N1X laptop processor compares with strong competitors like AMD and Apple. While the performance is impressive, it doesn’t quite match up, which is not surprising considering it’s an engineering sample. Now, we have a look at its GPU performance through a Geekbench OpenCL entry.
GPU Performance Insights
The Nvidia N1X sample recorded a score of 46,361 points, which is roughly similar to what an RTX 2050 laptop GPU achieves. It’s important to note that the GPU isn’t running at its top speed, with a maximum frequency of 1,048 MHz confirmed. Additionally, we have details on its CPU, which includes two clusters, each with 10 cores, summing up to 20 cores in total.
Specifications Breakdown
According to Geekbench, the GPU has 48 Streaming Multiprocessors (SMs). If Nvidia maintains its standard setup of 128 CUDA cores per SM, the N1X’s GPU would feature 6,144 CUDA cores. This positions it between the RTX 5070 Ti with 5,888 cores and the RTX 5080 which has 8,192 cores.
The laptop is equipped with 128 GB of RAM, with 64 GB dedicated to the GPU. It operates on Windows 11 Enterprise. Sadly, this is all the information we have about the N1X for the time being. With the release possibly pushed to 2026, updates will likely come through leaks.
Source:
Link

Tags: Geekbench, Nvidia N1x, RTX 2050
July 25, 2025
Exynos 2600: Galaxy S26 SoC’s Weak Geekbench Performance
Key Takeaways
1. The Exynos 2600, an upcoming SoC, scored 2,155 in single-core and 7,788 in multi-core tests on Geekbench 6.4, which is lower than its predecessor, the Exynos 2500.
2. This chipset features a 10-core architecture with one Cortex-X930 core at 3.55 GHz, three Cortex-A730 cores at 2.96 GHz, and six Cortex-A730 cores at 2.46 GHz.
3. The Exynos 2600 includes an Xclipse 960 GPU, which may be developed in-house by Samsung rather than based on AMD’s RDNA architecture.
4. It introduces Scalable Matrix Extension (SME) instructions, aimed at improving performance in AI tasks.
5. The early appearance of the Exynos 2600 on Geekbench suggests advancements in Samsung’s production capabilities, but its use in the Galaxy S26 series remains uncertain.
Although the future of Samsung’s upcoming SoC, the Exynos 2600, is still unclear, it has made its way onto Geekbench. While the performance isn’t particularly impressive, it’s important to remember that this is just an engineering sample that’s still some time from being officially released. We also have confirmation of its specifications.
Performance Metrics
The Exynos 2600, which has the model number S5E9965, achieved scores of 2,155 in single-core and 7,788 in multi-core tests on Geekbench 6.4. These results are a bit lower than those of the Exynos 2500 found in the Galaxy Z Flip 7 and are significantly behind its projected scores of 2,950 and 10,200 points. Since this is a pre-production model, there is room for enhancement, but it is doubtful that it will match the performance of the Snapdragon 8 Elite 2 or Dimensity 9500.
Specifications Overview
In terms of specifications, the Exynos 2600 follows Samsung’s established 10-core architecture, featuring one core clocked at 3.55 GHz, three cores at 2.96 GHz, and six cores running at 2.46 GHz. According to a leak from the South Korean forum DCinside, Samsung appears to have adopted MediaTek’s approach with an all-P-core configuration. The main CPU core is identified as a Cortex-X930, while the other nine cores are Cortex-A730.
GPU and Features
The back-end info from Geekbench indicates that the Exynos 2600 is equipped with an Xclipse 960 GPU. Although the name suggests it draws from AMD’s RDNA architecture, a recent leak hinted that it might actually be developed in-house by Samsung. A significant update for the Exynos 2600 includes the introduction of SME (Scalable Matrix Extension) instructions, which should enhance performance in AI-related tasks.
The early appearance of the Exynos 2600 on Geekbench doesn’t guarantee that it will be the chipset for the Galaxy S26 series next year. We’ve seen similar situations unfold previously with the Exynos 2500 and a Galaxy S25 prototype. However, the fact that the Exynos 2600 has surfaced much earlier indicates that Samsung has managed to stabilize SF2 production enough to begin prototype manufacturing.
Source:
Link

Tags: Exynos 2600, Geekbench, Samsung
July 18, 2025
AMD Ryzen Threadripper 9980X Benchmark Shows Performance Boost
Key Takeaways
1. AMD is set to launch its new Ryzen Threadripper 9980X CPUs later this month (July).
2. The Threadripper 9980X has shown significant performance improvements, with a single-core score of 3,259 and a multi-core score of 28,666.
3. Compared to its predecessor, the Threadripper 7980X, the 9980X offers about 10-12% better single-core performance and 13-15% better multi-core performance.
4. The Threadripper 9980X features 64 cores, 128 threads, boost speeds up to 5.4 GHz, and a 256 MB L3 cache with a TDP of 350W.
5. The official release date for the Threadripper 9000WX series is set for July 23, available through regional system integration partners and global OEMs.
AMD is getting ready to launch its new Ryzen Threadripper CPUs soon. This expected release is happening later this month (July), and the new lineup is already showing up on several benchmarking sites. Notably, the Threadripper 9980X recently set a record on PassMark and has also appeared on Geekbench, revealing similar performance improvements compared to its predecessor.
Performance Highlights
The 64-core AMD Ryzen Threadripper 9980X was identified by Benchleaks on X, where they provided details about its performance. The CPU achieved a single-core score of 3,259 and a multi-core score of 28,666. Testing was done using an Asus Pro WS TRX50-SAGE WIFI motherboard paired with 128 GB of RAM. In contrast, the Threadripper 7980X, tested on a similar setup, managed a single-core score of 2,882 and a multi-core score of 24,790.
Performance Comparison
These results indicate an improvement of about 10% to 12% in single-core performance and roughly 13% to 15% in multi-core performance. Though Geekbench may not perfectly reflect real-world CPU performance, it provides a general sense of how the CPU measures up against other processors.
The findings are consistent with the recent scores from PassMark, which has placed the Threadripper 9980X at the forefront, showing a 7.7% increase in performance compared to the Threadripper 7980X.
Specifications Overview
AMD has released complete specifications for the Threadripper 9980X. This CPU features 64 cores, 128 threads, and is built on the Zen 5 architecture. It can reach boost speeds of up to 5.4 GHz and comes with a 256 MB L3 cache. The standard TDP is 350W, and it also supports the latest AMD Ryzen Technologies, including EXPO.
While AMD has yet to announce an official release date for the new Threadripper 9000X series, their recent blog post mentions that the Threadripper 9000WX lineup will be available starting July 23 through various regional system integration partners and global OEMs.
Source:
Link
Tags: amd, Geekbench, Ryzen Threadripper 9980X
July 16, 2025
AMD Ryzen Z2 Extreme Leaks on Geekbench with MSI Claw Handheld
Key Takeaways
1. AMD’s Ryzen Z2 Extreme APU is built on the Strix Point architecture and features two CPU clusters: three Zen 5 cores and five Zen 5c cores.
2. In Geekbench 6.4, the Ryzen Z2 Extreme achieved single-core and multi-core scores of 2,748 and 12,182, respectively, outperforming the previous Ryzen Z1 Extreme.
3. The APU’s boost clock is capped at 4.3 GHz, lower than the promised 5.0 GHz.
4. It has 16 GB of DDR5 memory running at a relatively slow 1,944 MT/s.
5. The integrated Radeon 890M GPU shows significant performance improvements, scoring 45,064 in Vulkan and 37,970 in OpenCL tests, making it up to 52% faster than the Radeon 780M.
AMD has been, and continues to be, quite secretive about the details of its upcoming Ryzen Z2 Extreme APU designed for handheld gaming devices. Earlier speculations suggested that this APU would be built on Team Red’s latest Strix Point architecture. A recent listing on Geekbench supports this theory by showing two separate CPU clusters: one featuring three cores (likely Zen 5) and another with five Zen 5c cores. This chip is set to power MSI’s new AMD-based Claw 8, which has already been tested with the same processor.
Performance Scores
In Geekbench 6.4, it achieved scores of 2,748 for single-core and 12,182 for multi-core performance. In comparison, the previous Ryzen Z1 Extreme managed scores of 2,534 and 11,358 in the same test. A quick look at its Geekbench listing suggests that the Ryzen Z2 Extreme is not operating at its promised boost clock of around 5.0 GHz, as this sample is capped at 4.3 GHz.
Memory Specifications
The Geekbench listing also indicates that the 16 GB of memory runs at 1,944 MT/s, which is considered quite slow for DDR5 memory. The Ryzen Z2 Extreme includes AMD’s RDNA 3.5-based Radeon 890M integrated GPU. It scored 45,064 and 37,970 points in Geekbench’s Vulkan and OpenCL tests, respectively, making it potentially up to 52% faster than the previous generation Radeon 780M. Overall, this chip appears to be a strong contender, and it will be fascinating to see how it performs at lower power levels.
Source:
link
Tags: amd, Geekbench, Ryzen Z2 Extreme
July 10, 2025
Tensor G5 and Pixel 10 Pro Fold Show Average Geekbench Performance
Key Takeaways
1. The Google Tensor G5 achieved scores of 2,276 (single-core) and 6,173 (multi-core) in Geekbench tests, showing improvement over earlier scores but still falling short of competitors.
2. The chip features a CPU layout of one Cortex-X4 core, five Cortex-A725 cores, and two Cortex-A520 cores, with a new PowerVR DXT-48-1536 GPU.
3. The Tensor G5 demonstrates consistent performance without throttling, indicating stable clock speeds during testing.
4. Despite its improvements, the Tensor G5 is unlikely to compete with leading chips like the Snapdragon 8 Elite and Dimensity 9400 due to its use of older Arm cores.
5. Future potential may be seen if the chip transitions to TSMC’s 3 nm process, although the Exynos modem could negatively affect overall performance.
The first performance test of the Google Tensor G5 on Geekbench wasn’t very impressive, which makes sense since it was likely an engineering model. Now, we can see it tested again, this time paired with Google’s upcoming foldable device, the Pixel 10 Pro Fold.
Performance Scores
In the latest Geekbench results, the Tensor G5 achieved scores of 2,276 for single-core and 6,173 for multi-core tests. The back-end data from Geekbench indicates that the chip isn’t throttling and can sustain a consistent boost clock. Although these numbers show improvement over the initial scores of 1,323 and 4,004, they still fall significantly short compared to leading competitors like the Snapdragon 8 Elite (3,070/9,251) and the Dimensity 9400 (2,597/8,109). It does come closer to the older Dimensity 9300, which scored 2,207 in single-core and 7,408 in multi-core tests.
CPU Layout and GPU
We also get an insight into the CPU configuration of the Tensor G5, which features one Cortex-X4 prime core clocked at 3.78 GHz, five Cortex-A725 performance cores running at 3.05 GHz, and two Cortex-A520 efficiency cores at 2.25 GHz. Additionally, we see its new GPU from Imagination Technologies, the PowerVR DXT-48-1536. The Pixel 10 Pro Fold being tested has 16 GB of RAM and operates on Android 16.
Future Potential
Realistically, the Tensor G5 is unlikely to match up against the current generation giants, particularly because of its use of older Arm cores. However, switching to TSMC’s 3 nm (or 5 nm, depending on whom you ask) process might improve its power efficiency. On the flip side, the Exynos modem could negatively impact performance.
Source:
Link

Tags: Geekbench, Google Tensor G5, Pixel 10 Pro Fold
July 9, 2025
AMD Ryzen Threadripper 9995WX 96-Core Shows Strong Geekbench Boost
Key Takeaways
1. The Ryzen Threadripper PRO 9995WX is based on the Zen 5 architecture and uses TSMC’s 4nm technology.
2. It features 96 cores and 192 threads, with a TDP of 350 watts, offering a 22% boost in multi-core performance compared to the previous model.
3. The CPU achieved a multi-core score of 30,170 and a single-core score of 2,800 on Geekbench, showing a 26% increase in multi-core and 9% increase in single-core performance from the Ryzen Threadripper PRO 7995WX.
4. The price for the Ryzen Threadripper PRO 9995WX is approximately $12,700 in the US and €12,300 in the EU.
5. The CPU is positioned as AMD’s flagship model in the Shimada Peak series, reflecting a significant price increase.
Thanks to a recent Geekbench listing that was found by VideoCardz, we now have some solid information about the performance gains offered by AMD’s newly announced Ryzen Threadripper PRO 9995WX high-end desktop CPU, which is the top dog in the Shimada Peak series.
Overview of the Ryzen Threadripper PRO 9995WX
Before diving into the Geekbench 6 results, let’s recap what we know about the Ryzen Threadripper PRO 9995WX CPU. This processor is based on the Zen 5 architecture and is manufactured using TSMC’s 4nm technology. It features an impressive 96 cores and 192 threads, along with a TDP of 350 watts. AMD claims that there is a 22% boost in multi-core performance compared to the previous model, the Zen 4 Ryzen Threadripper PRO 7995WX.
Performance Scores
According to the Geekbench listing, the Threadripper PRO 9995WX achieved an impressive score of 30,170 points in the multi-core performance. However, its single-core score of 2,800, while quite good, doesn’t really blow anyone away. When you compare it to its older counterpart, which scored 23,913 in the same benchmark, the 9995WX shows a notable 26% increase. Nevertheless, the single-core score improvement is a more modest 9%.
Price Tag
VideoCardz also discovered several online listings for the Ryzen Threadripper PRO 9995WX, indicating that this workstation CPU comes with a steep price of around $12,700 in the US and about €12,300 in the EU. Clearly, AMD’s flagship model in the Shimada Peak lineup is arriving with a significant price hike.
Source:
Link

Tags: amd, Geekbench, Ryzen Threadripper PRO 9995WX
July 8, 2025
Samsung Galaxy Z Flip FE with Exynos 2400 Surfaces on Geekbench
Key Takeaways
1. The Galaxy Z Flip FE is set to launch soon and has been spotted on Geekbench, indicating its performance levels.
2. It will feature the Exynos 2400 chip, differing from the Exynos 2500 in the Galaxy Z Flip7.
3. Performance scores on Geekbench show the Exynos 2400 has single-core scores of 1,940 and multi-core scores of 6,136, placing it between Snapdragon 8 Gen 2 and Snapdragon 8 Gen 3.
4. The device will come with 8 GB of RAM and has the model number “SM-F761B.”
5. It achieves a GPU score of 12,890 in OpenCL tests, comparable to flagship devices like the Galaxy Z Fold6 and Oppo Find X7.
Samsung is set to introduce the Galaxy Z Flip FE soon, and the device has appeared on Geekbench before its official release, showcasing its performance capabilities.
Performance Details
According to Geekbench, the Galaxy Z Flip FE will be equipped with the Exynos 2400 chip. This is different from the Exynos 2500 found in the upcoming Galaxy Z Flip7. The Exynos 2400 is a familiar component, as it was also used in last year’s Galaxy S24 and S24 Plus models in Europe. Its performance is quite similar to those devices, achieving scores of 1,940 for single-core and 6,136 for multi-core tests on Geekbench 6. These figures suggest that its CPU performance falls between Qualcomm’s Snapdragon 8 Gen 2 and Snapdragon 8 Gen 3.
Specifications and Features
The Galaxy Z Flip FE will come with 8 GB of RAM and has the model number “SM-F761B.” Regarding GPU performance, it also shows comparable results. The Exynos 2400 achieves a score of 12,890 in Geekbench’s OpenCL test, which aligns with last-generation flagship devices like the Galaxy Z Fold6 and Oppo Find X7, which feature the Snapdragon 8 Gen 3 and MediaTek Dimensity 9300, respectively.
Source:
Link
Tags: Exynos 2400, Galaxy Z Flip FE, Geekbench
July 6, 2025
Exynos 2500 vs Snapdragon 8 Elite: GPU Benchmarks Compared
Key Takeaways
1. Samsung introduced its flagship chipset, the Exynos 2500, for the Galaxy Z Flip7.
2. The Exynos 2500 achieved an impressive GPU score of 18,601 on the OpenCL benchmark.
3. Its GPU, based on RDNA 3 with Xclipse 950, shows potential for strong performance.
4. The chipset’s CPU performance is weaker, scoring 2,313 (single-core) and 7,965 (multi-core) on Geekbench.
5. CPU scores indicate performance similar to last year’s Snapdragon 8 Gen 3.
A few weeks back, Samsung quietly introduced its flagship chipset for the year, the Exynos 2500. While it’s not really expected to match its competitors in terms of CPU performance, it looks like the GPU capabilities might tell a different story.
Impressive GPU Scores
Abhishek Yadav has revealed that the Exynos 2500, found in the upcoming Galaxy Z Flip7, achieved a score of 18,601 on the OpenCL benchmark from Geekbench. This score is quite impressive, especially since the Snapdragon 8 Elite for the Galaxy S25 Ultra scored 18,365 on the same test, according to Geekbench’s official rankings.
GPU Performance Expectations
Though this is just a single benchmark and shouldn’t be taken as final proof of the chipset’s abilities, it suggests that the Exynos 2500’s RDNA 3-based Xclipse 950 GPU might perform quite well. This isn’t too surprising, considering the Exynos 2400 has already outperformed the A17 Pro and Snapdragon 8 Gen 3 in several GPU tests.
CPU Performance Lags Behind
On the downside, the CPU performance isn’t as strong. The Galaxy Z Flip7, model “SM-F766B”, has shown up on Geekbench several times, but the results have generally been disappointing. In its latest test, the foldable device managed scores of 2,313 and 7,965 in single-core and multi-core tests, respectively, indicating that its CPU performance is pretty much on par with last year’s Snapdragon 8 Gen 3.
Source:
Link
Tags: Exynos 2500, Galaxy Z Flip7, Geekbench
July 3, 2025
Ryzen AI 5 330: New Krackan Point CPU Debuts on Geekbench
Key Takeaways
1. The Ryzen AI 5 330 is a new AMD laptop processor, part of the Ryzen AI 300 Krackan Point series.
2. It features a less powerful configuration compared to the Ryzen AI 5 340 and has a reduced Radeon 820M iGPU.
3. Performance scores on Geekbench 6.3 are 1,949 for single-core and 7,047 for multi-core, lower than the Ryzen AI 5 340.
4. The processor has four CPU cores: one Zen 5 core and three Zen 5c cores, marking a new configuration.
5. The Ryzen AI 5 330 was tested with an Acer laptop that includes 32 GB of RAM.
A new AMD laptop processor has appeared on Geekbench. The Ryzen AI 5 330 is part of the Ryzen AI 300 Krackan Point series, and it serves as a less powerful version of the Ryzen AI 5 340. This new chip also comes with a reduced Radeon 820M iGPU, though its specifications are not currently available.
Performance Scores
The Ryzen AI 5 330 achieved scores of 1,949 in single-core and 7,047 in multi-core tests using Geekbench 6.3. These scores are significantly lower than those of the Ryzen AI 5 340, which scored 2,776 in single-core and 10,668 in multi-core. The Geekbench backend indicates that the processor is not experiencing throttling, suggesting that the lower single-core score is due to its modest single-core boost clock of 3.721 GHz.
Core Configuration
Conversely, the multi-core score reflects the limited number of CPU cores. The Geekbench listing shows that the Ryzen AI 5 330 has a total of four CPU cores: one Zen 5 core and three Zen 5c cores. This makes it the first processor to feature such a configuration. It was also observed alongside an Acer laptop that has 32 GB of RAM.
Source:
Link
Tags: Acer, AMD Ryzen AI 5 330, Geekbench