- Exynos 2400 has a bonkers GPU with 4x more compute units than Xclipse 920 SamMobile – Samsung news
- Exynos 2400 GPU To Have 4x The Compute Units Of The Exynos 2200, New Rumor Claims Cortex-X4 Is Part Of 10-Core CPU Cluster Wccftech
- Latest Galaxy S24 rumors: powerful deca-core Exynos 2400 SoC, 16GB RAM for S24 Ultra PhoneArena
- Samsung MX greenlit the Exynos 2400 SoC for the Galaxy S24 SamMobile – Samsung news
- Exynos 2400 GPU Core Count Revealed, New Report Claims Samsung Will Use Its ‘Low Power’ Node To Mass Produce It Wccftech
- View Full Coverage on Google News
Tag Archives: Compute
Exynos 2400 GPU To Have 4x The Compute Units Of The Exynos 2200, New Rumor Claims Cortex-X4 Is Part Of 10-Core CPU Cluster – Wccftech
- Exynos 2400 GPU To Have 4x The Compute Units Of The Exynos 2200, New Rumor Claims Cortex-X4 Is Part Of 10-Core CPU Cluster Wccftech
- Exynos 2400 has a bonkers GPU with 4x more compute units than Xclipse 920 SamMobile – Samsung news
- Latest Galaxy S24 rumors: powerful deca-core Exynos 2400 SoC, 16GB RAM for S24 Ultra PhoneArena
- Exynos 2400 GPU Core Count Revealed, New Report Claims Samsung Will Use Its ‘Low Power’ Node To Mass Produce It Wccftech
- Samsung MX greenlit the Exynos 2400 SoC for the Galaxy S24 SamMobile – Samsung news
- View Full Coverage on Google News
NVIDIA GeForce RTX 4080 16 GB Graphics Card Benchmarks Leak Out, Up To 29% Faster in 3DMark Tests & 53 TFLOPs Compute
The first benchmarks of NVIDIA’s GeForce RTX 4080 16 GB graphics card have leaked online and show over 20% performance gain in 3DMark tests.
NVIDIA GeForce RTX 4080 16 GB Graphics Card Beats The RTX 3090 Ti 24 GB In Leaked 3DMark & Gaming Benchmarks
The benchmarks of the NVIDIA GeForce RTX 4080 16 GB graphics card have leaked over at Chiphell Forums. It is unknown if the graphics card was a Founders Edition variant or an AIB model but the chip was running on an AMD Ryzen 7 5800X3D CPU. The graphics card was tested across 3DMark & a few titles. But first, we got to lay our eyes on the AIDA64 GPGPU Benchmark which shows that the card offers up to 53.6 TFLOPs of single-precision performance which is 8% higher than the officially reported figure of 49 TFLOPs. For comparison, the GeForce RTX 3090 Ti produces 40 TFLOPs so this is a 32.5% improvement in single-precision compute.
In terms of synthetic benchmarks, the NVIDIA GeForce RTX 4080 16 GB scores 13977 points in the 3DMark Time Spy Extreme (Graphics), 17465 points in the 3DMark Fire Strike Ultra (Graphics), and 17607 Points in the 3DMark Port Royal benchmark.
Just for comparison, my heavily overclocked & custom RTX 3090 Ti SUPRIM X scores 11643 points in Time Spy Extreme, 13554 points in Fire Strike Ultra, and 15124 points in Port Royal. If we compare these figures to the RTX 4080 16 GB, we get the following performance:
- RTX 4080 16 GB vs RTX 3090 Ti 24 GB in Time Spy Extreme – 20% Faster
- RTX 4080 16 GB vs RTX 3090 Ti 24 GB in Fire Strike Ultra – 29% Faster
- RTX 4080 16 GB vs RTX 3090 Ti 24 GB in Port Royal (DXR) – 16% Faster
0
4000
8000
12000
16000
20000
24000
The user also shared performance benchmarks of two games, Red Dead Redemption 2 (with and without DLSS) and Shadow of The Tomb Raider (Full High DLSS Quality). The results can be seen in the screenshots below:
NVIDIA GeForce RTX 4080 16 GB ‘Official’ Specifications
The NVIDIA GeForce RTX 4080 16 GB graphics card is expected to utilize a cut-down AD103-300 GPU configuration with 9,728 cores or 76 SMs enabled of the total 84 units whereas the previous configuration offered 80 SMs or 10,240 cores. While the full GPU comes packed with 64 MB of L2 cache and up to 224 ROPs, the RTX 4080 might end up with 48 MB of L2 cache and lower ROPs too due to its cut-down design. The card is expected to be based on the PG136/139-SKU360 PCB. The graphics card is said to offer a peak clock rate of 2505 MHz.
As for memory specs, the GeForce RTX 4080 is expected to rock 16 GB GDDR6X capacities that are said to be adjusted at 22.5 Gbps speeds across a 256-bit bus interface. This will provide up to 720 GB/s of bandwidth. This is still a tad bit slower than the 760 GB/s bandwidth offered by the RTX 3080 since it comes with a 320-bit interface but a lowly 10 GB capacity. To compensate for the lower bandwidth, NVIDIA could be integrating a next-gen memory compression suite to make up for the 256-bit interface.
- NVIDIA GeForce RTX 4080 16 GB “Official” TBP – 320W
- NVIDIA GeForce RTX 3080 12 GB “Official” TBP – 350W
The card will have a TBP of 320W which is 30W lower than the TBP of the 12 GB RTX 3080 and much lower than the TBP of the RTX 3090 Ti while offering a big performance jump. The NVIDIA GeForce RTX 4080 16 GB and RTX 4080 12 GB graphics cards will be launching in November and be priced at $1199 US and $899 US, respectively. Considering if the card does end up around 20-30% faster in games than the RTX 3090 Ti, then it could
NVIDIA GeForce RTX 40 Series Official Specs:
Graphics Card Name | NVIDIA GeForce RTX 4090 | NVIDIA GeForce RTX 4080 16G | NVIDIA GeForce RTX 4080 12G | NVIDIA GeForce RTX 3090 Ti |
---|---|---|---|---|
GPU Name | Ada Lovelace AD102-300 | Ada Lovelace AD103-300 | Ada Lovelace AD104-400 | Ampere GA102-225 |
Process Node | TSMC 4N | TSMC 4N | TSMC 4N | Samsung 8nm |
Die Size | 608mm2 | 378.6mm2 | 294.5mm2 | 628.4mm2 |
Transistors | 76 Billion | 45.9 Billion | 35.8 Billion | 28 Billion |
CUDA Cores | 16384 | 9728 | 7680 | 10240 |
TMUs / ROPs | 512 / 176 | 320 / 112 | 240 / 80 | 320 / 112 |
Tensor / RT Cores | 512 / 128 | 304 / 76 | 240 / 60 | 320 / 80 |
Base Clock | 2230 MHz | 2210 MHz | 2310 MHz | 1365 MHz |
Boost Clock | 2520 MHz | 2510 MHz | 2610 MHz | 1665 MHz |
FP32 Compute | 83 TFLOPs | 49 TFLOPs | 40 TFLOPs | 40 TFLOPs |
RT TFLOPs | 191 TFLOPs | 113 TFLOPs | 82 TFLOPs | 78 TFLOPs |
Tensor-TOPs | 1321 TOPs | 780 TOPs | 641 TOPs | 320 TOPs |
Memory Capacity | 24 GB GDDR6X | 16 GB GDDR6X | 12 GB GDDR6X | 12 GB GDDR6X |
Memory Bus | 384-bit | 256-bit | 192-bit | 384-bit |
Memory Speed | 21.0 Gbps | 23.0 Gbps | 21.0 Gbps | 19 Gbps |
Bandwidth | 1008 GB/s | 736 GB/s | 504 GB/s | 912 Gbps |
TBP | 450W | 320W | 285W | 350W |
Price (MSRP / FE) | $1599 US | $1199 US | $899 US | $1199 |
Launch (Availability) | October 2022 | November 2022 | November 2022 | 3rd June 2021 |
News Source: Olrak
Products mentioned in this post
BTC mining firm Compute North files for bankruptcy
Bitcoin (BTC) mining hosting firm Compute North has filed for chapter 11 bankruptcy, amid growing pressure on the firm due to the effects of crypto winter and rising energy costs. The firm’s CEO Dave Perrill has also stepped down but will remain on the board.
The company submitted a Chapter 11 bankruptcy filing in the U.S. Bankruptcy Court for the Southern District of Texas on Sept. 22, which is now pending before Judge David Jones.
Under a Chapter 11 filing, the firm is still able to keep its operations going as it works out a plan to repay creditors. The filing reportedly outlines that Compute North owes around $500 million to 200 creditors, while its assets are said to be worth between $100 million and $500 million.
Compute North offers large scale crypto mining hosting services and facilities, hardware and a BTC mining pool. It is one of the largest data center providers in the U.S. has big name partners in the BTC mining sector such as Compass Mining and Marathon Digital.
Both companies have come out with statements via Twitter, noting that with the information they have at this stage, their business operations will continue as normal.
“Compute North’s staff informed us today that the bankruptcy filing should not disrupt business operations. We are continuing to monitor the situation and will provide further updates as they become available,” noted Compass Mining.
Today, a filing related to one of our hosting providers was published. Based on the information available at this time, it is our understanding that this filing will not impact our current mining operations.
— Marathon Digital Holdings (NASDAQ: MARA) (@MarathonDH) September 22, 2022
The bearish performance of BTC in 2022 has had a significant impact on the mining sector this year, and in the context of Texas, rising energy costs and multiple power outages during intense heat waves haven’t helped either.
Related: Maple Finance launches $300M lending pool for Bitcoin mining firms
Bloomberg Business reporter David Pan highlighted on Twitter that Compute North may have been impacted by a costly delay to a large mining facility in Texas that it wasn’t able to monetize for months.
“Compute North’s massive 280MW mining facility in TX was supposed to run rigs in April but it couldn’t due to pending approvals. From then to later this year when it finally was able to energize the machines, Bitcoin prices had gone through multiple downward cycles, fundraising opportunities dried up and major lenders scaled back,” he wrote.
Compute North adds to a long list of crypto firms that have either fallen victim to crypto winter — or in some cases helped create it — including Voyager Digital, Three Arrows Capital, Celsius Network and BlockFi to name a few.
AMD RDNA 3 “Navi 3X” GPUs Feature Double The Cache Per Compute Unit & Shader Array
AMD has listed down the latest information regarding the Cache sizes of their upcoming RDNA 3 “Navi 3X” GPUs within Linux patches.
AMD’s Next-Gen RDNA 3 GPUs For Navi 3X Lineup To Double The Cache Size For Compute Units & Shader Array
Published over at the FreeDesktop Linux repository by AMD’s Aaron Liu and discovered by Coelacanth-Dream and Kepler_L2, we get to see the first details regarding the cache sizes in upcoming RDNA 3 GPUs such as the Navi 31, Navi 32, and Navi 33 chips which were recently leaked.
Coming to the details, the AMD RDNA 3 (GFX11) GPU lineup will feature double the L0 vector cache for each Compute Unit or CU and also double the GL1 data cache (RDNA L1 cache for each Shader Array or SA. As per the new information, the Vector Register File per SIMD will increase to 192KB vs 128KB on RDNA 2, the L0 Vector/Texture cache will increase from 16 KB to 32 KB per CU, the GPU L1 Data Cache per Shader Array will increase from 128 KB to 256 KB while the L2 Data Cache will remain the same as RDNA 2.
There are also cache sizes listed for AMD’s Navi 33 and Phoenix APUs which will also feature an RDNA 3 graphics core but in a monolithic package. The L0 Vector/Texture size is increased from 16 KB to 32 KB while the L1 Data Cache (Graphics) is increased from 128 KB to 256 KB. The Register file size remains untouched on Navi 33 GPUs & Phoenix APUs.
Cache Info | Yellow Carp (Rembrandt) | RDNA 3(GFX11 Navi 31/32) | Phoenix (GC 11.0.1, GFX1103) |
---|---|---|---|
L0 Vector Register File per SIMD | 128KiB | 192KiB | 128KiB |
L0 Vector Data (per CU) | 16KiB | 32KiB | 32KiB |
L1 Scalar Inst. (per WGP) | 32KiB | 32KiB | 32KiB |
L1 Scalar Data (per WGP) | 16KiB | 16KiB | 16KiB |
GL1 Date (per SA) | 128KiB | 256KiB | 256KiB |
L2 Data | 2048KiB (2MiB) | 2048KiB (2MiB) | 2048KiB (2MiB) |
L3 (MALL) | N/A | Yes | N/A |
Coelacanth-Dream also states that all RDNA 3 “Navi 3X” GPUs come with VODP (Dual-Issue Wave32) instructions, WMMA (Wave Matrix Multiply-Accumulate) support and the performance per WGP has been improved a lot. The increased GL1 cache is said to improve pixel processing performance and is amongst the many changes that AMD is bringing within its RDNA 3 Navi 3x GPU family.
AMD confirmed that its RDNA 3 GPUs will be coming later this year with a huge performance uplift. The company’s Senior Vice President of Engineering, Radeon Technologies Group, David Wang, said that the next-gen GPUs for Radeon RX 7000 series will offer over 50% performance per watt uplift vs the existing RDNA 2 GPUs. Some of the key features of the RDNA 3 GPUs highlighted by AMD will include:
- 5nm Process Node
- Advanced Chiplet Packaging
- Rearchitected Compute Unit
- Optimized Graphics Pipeline
- Next-Gen AMD Infinity Cache
- >50% Perf/Watt vs RDNA 2
AMD will be rearchitecting the compute units within RDNA 3 to deliver enhanced raytracing capabilities. Although there’s no mention of what these capabilities are if we were to guess, we would say it’s definitely talking about performance and a set of advanced features on the RDNA 3 GPU core for Radeon RX 7000 graphics cards. The AMD Radeon RX 7000 graphics cards are going to launch later this year and offer a big leap in gaming performance so stay tuned for more info in the coming weeks.
AMD RDNA 3 Navi 3X GPU Configurations (Preliminary)
GPU Name | Navi 21 | Navi 33 | Navi 32 | Navi 31 | Navi 3X |
---|---|---|---|---|---|
Codename | Sienna Cichlid | Hotpink Bonefish | Wheat Nas | Plum Bonito | TBD |
GPU Process | 7nm | 6nm | 5nm/6nm | 5nm/6nm | 5nm/6nm |
GPU Package | Monolithic | Monolithic | MCM (1 GCD + 4 MCD) | MCM (1 GCD + 6 MCD) | MCM (TBD) |
GPU Die Size | 520mm2 | 203mm2 (Only GCD) | 200mm2 (Only GCD) 425mm2 (with MCDs) |
308mm2 (Only GCD) 533mm2 (with MCDs) |
TBD |
Shader Engines | 4 | 2 | 4 | 6 | 8 |
GPU WGPs | 40 | 16 | 30 | 48 | 64 |
SPs Per WGP | 128 | 256 | 256 | 256 | 256 |
Compute Units (Per Die) | 80 | 32 | 60 | 96 | 128 (per GPU) 256 (Total) |
Cores (Per Die) | 5120 | 4096 | 7680 | 12288 | 8192 |
Cores (Total) | 5120 | 4096 | 7680 | 12288 | 16,384 |
Memory Bus | 256-bit | 128-bit | 256-bit | 384-bit | 384-bit x2? |
Memory Type | GDDR6 | GDDR6 | GDDR6 | GDDR6 | GDDR6 |
Memory Capacity | Up To 16 GB | Up To 8 GB | Up To 16 GB | Up To 24 GB | Up To 32 GB |
Memory Speed | 16-18 Gbps | TBD | TBD | 20 Gbps | TBD |
Memory Bandwidth | 512-576 GB/s | TBD | TBD | 960 GB/s | TBD |
Infinity Cache | 128 MB | 32 MB | 64 MB | 96/192 MB | TBD |
Flagship SKU | Radeon RX 6900 XTX | Radeon RX 7600 XT? | Radeon RX 7800 XT? Radeon RX 7700 XT? |
Radeon RX 7900 XT? | Radeon Pro |
TBP | 330W | ~150W | ~250W | ~350W | TBD |
Launch | Q4 2020 | Q4 2022? | Q4 2022? | Q4 2022? | 2023? |
Intel Meteor Lake CPUs With TGPU Reportedly Features Ray Tracing Support, Sports FP64 Compute But Lacks XMX Units
In a new commit published on GitHub, Intel might have confirmed ray tracing support for their Meteor Lake CPUs with Xe-HPG TGPU. The Intel 14th Gen Meteor Lake CPUs are scheduled for release in 2023 and will come with a range of new features, including brand new hybrid cores, integrated graphics using tiled- designs, and a VPU accelerator which we talked about in our last coverage.
Intel Meteor Lake CPUs With TGPU Design Reportedly Comes With Ray Tracing & FP64 Support But Lacks XMX Units
According to a detailed report published by Coelacanth-Dream, it looks like the GPU featured on the Intel Meteor Lake CPUs is detected as a Xe-HPG product within IGC (Intel Graphics Compiler). Considering that, we can assume that it will be using the same architecture as the existing DG2 or Alchemist GPUs but that’s not entirely the case.
The Meteor Lake GPU which is also known as tGPU (Tiled-GPU) does not support DPAS (Dot Product Accumulate Systolic) instructions executed through the XMX units because it reportedly is not equipped with those. In a previous report, we covered how Intel’s Meteor Lake CPU’s graphics unit would lack XMX, resulting in lower feature-level support for technologies such as XeSS but it will instead feature partial FP64 support, something that’s been missing in Intel iGPUs for the past few generations.
The more interesting detail is that the TGPU (Xe-MTL) on the Intel Meteor Lake CPUs is reportedly going to come with hardware-level raytracing support. The TGPU is said to offer the same level of ray-tracing support as Alchemist and Ponte Vecchio GPUs since there’s no change in the design, at least as far as ray tracing is concerned.
Regarding support for HW ray tracing on Meteor Lake GPUs
IGC/AdaptorCommon/RayTracing/PrologueShaders.cpp
, itIGFX_METEORLAKE
is believed that they support it because of the addition of .Other than that, in the code part that determines his HW ray tracing support, it seems that Meteor Lake GPU supports it as well as Alchemist / DG2 and Ponte Vechhio because there is no change.
via Coelacanth-Dream
Intel’s Meteor Lake CPU tiled-GPU won’t be the first integrated graphics chip to offer support for ray-tracing. AMD has been offering it for almost a year now with their Ryzen 6000 “Rembrandt” APUs that feature RDNA 2 iGPUs. This will be Intel’s first ray-tracing-enabled integrated graphics but we hope that some partial XeSS or AI-level upsampling support is added. As we have seen with AMD’s APUs, technologies such as FSR 2.0 can really help deliver playable FPS in gaming titles and that would be really awesome for mobile gamers.
Intel Meteor Lake Mobility CPU Lineup Expected Features:
- Triple-Hybrid CPU Architecture (P/E/LP-E Cores)
- Brand New Redwood Cove (P-Cores)
- Brand New Crestmont (E-Cores)
- Up To 14 Cores (6+8) For H/P Series & Up To 12 Cores (4+8) For U Series CPUs
- Intel 4 Process Node For CPU, TSMC For tGPU
- Intel ‘Xe-MTL’ GPU With Up To 192/128 EUs
- Up To LPDDR5X-7467 & DDR5-5200 Support
- Up To 96 GB DDR5 & 64 GB LPDDR5X Capacities
- Intel VPU For AI Inferencing With Atom Cores
- x8 Gen 5 Lanes For Discrete GPU (Only H-Series)
- Triple x4 M.2 Gen 4 SSD Support
- Four Thunderbolt 4 Ports
The Intel 14th Gen Meteor Lake CPUs are expected to launch by the second half of 2023 and will be utilizing the “Intel 4” process node along with a host of 3rd party IPs/process nodes for the rest of the tiles. Expect more information on Meteor Lake CPUs at Hot Chips 34.
Intel Mobility CPU Lineup:
CPU Family | Arrow Lake | Meteor Lake | Raptor Lake | Alder Lake |
---|---|---|---|---|
Process Node (CPU Tile) | Intel 20A ‘5nm EUV” | Intel 4 ‘7nm EUV’ | Intel 7 ’10nm ESF’ | Intel 7 ’10nm ESF’ |
CPU Architecture | Hybrid (Four-Core) | Hybrid (Triple-Core) | Hybrid (Dual-Core) | Hybrid (Dual-Core) |
P-Core Architecture | Lion Cove | Redwood Cove | Raptor Cove | Golden Cove |
E-Core Architecture | Skymont | Crestmont | Gracemont | Gracemont |
Top Configuration | TBD | 6+8 (H-Series) | 6+8 (H-Series) | 6+8 (H-Series) |
Max Cores / Threads | TBD | 14/20 | 14/20 | 14/20 |
Planned Lineup | H/P/U Series | H/P/U Series | H/P/U Series | H/P/U Series |
GPU Architecture | Xe2 Battlemage ‘Xe-LPG’ or Xe3 Celestial “Xe-LPG” |
Xe-LPG ‘Xe-MTL’ | Iris Xe (Gen 12) | Iris Xe (Gen 12) |
GPU Execution Units | 192 EUs (1024 Cores)? | 128 EUs (1024 Cores) 192 EUs (1536 Cores) |
96 EUs (768 Cores) | 96 EUs (768 Cores) |
Memory Support | TBD | DDR5-5600 LPDDR5-7400 LPDDR5X – 7400+ |
DDR5-5200 LPDDR5-5200 LPDDR5-6400 |
DDR5-4800 LPDDR5-5200 LPDDR5X-4267 |
Memory Capacity (Max) | TBD | 96 GB | 64 GB | 64 GB |
Thunderbolt 4 Ports | TBD | 4 | 2 | 2 |
WiFi Capability | TBD | WiFi 6E | WiFi 6E | WiFi 6E |
TDP | TBD | 15-45W | 15-45W | 15-45W |
Launch | 2H 2024? | 2H 2023 | 1H 2023 | 1H 2022 |
News Source: HardwareTimes
An Entry-Level Design With 4 Cores, 5 Vega Compute Units, 35W & 3.7 GHz Max Boost
AMD seems to have released an entry-level Athlon Gold PRO 4150GE in the Asian Pacific market which will feature the Zen 2 core architecture.
AMD’s Athlon Gold PRO 4150E APU Comes With 4 Zen 2 CPU & 5 Vega GPU Cores, Up To 3.7 GHz at 35W
It’s been a while since AMD has touched the entry-level desktop CPU segment. Their Zen 3 Ryzen 5000X and 5000G lineups have been all but Ryzen 5 and beyond with a few Ryzen 3 parts going to the OEM segment. It looks like AMD is finally readying a new entry-level solution in the form of a new Athlon part. The Athlon Gold 4000 series was also rumored around mid of 2021 so it is looking like we are close to their launch.
Starting with the specs, the AMD Athlon Gold PRO 4150GE is based on the Zen 2 core architecture and features 4 cores / 4 threads. It has a base frequency of 3.3 GHz and boosts up to 3.7 GHz. The chip packs 4 MB of L2 cache and has an operating TDP of 35W since it’s a ‘GE’ part. A standard ‘G’ SKU may also launch with a TDP of 50-65W.
The standout upgrade, aside from the Zen 2 core architecture, is the addition of 3rd Gen Vega GPU cores. The AMD Athlon Gold PRO 4150GE not only packs the third iteration of the Vega integrated graphics but also comes with more compute units. There are 5 Compute Units or 320 cores on the APU which operate at 1500 MHz. For comparison, the Athlon Gold PRO 3150GE only housed 3 CUs and ran at 1100 MHz. Aside from that, the chip will feature support on the AM4 socket and retains DDR4-3200 support.
Since this is a PRO part, the AMD Athlon Gold 4150GE is likely to be a stock OEM part but as we have seen with earlier APU releases, the Asian Pacific market is likely to see a DIY launch too. The chip is currently being sold at Ali Express for a price of $118.29 US which is quite a hefty price to be paying for a chip of this tier but it is also expected since Ali Express isn’t an official AMD seller and 3rd party sellers can charge whatever they want for this chip. We are likely to see this chip in OEM PCs sooner or later at much better prices.
News Sources: Komachi , Momomo_US