First images of alleged NVIDIA GeForce RTX 4080 “Founders Edition” graphics card have been leaked Kitten Yuko.
NVIDIA GeForce RTX 4080 Founder Edition graphics card pictured, dual-slot but thicker design
The so-called NVIDIA GeForce RTX 4080 Founders Edition graphics card is wrapped in plastic, so it’s hard to fill in the details, but if we compare the cooler to the GeForce RTX 3080 FE, then we’re sure to see a big change. While the overall shroud is very similar to the existing RTX 30-series Founders Edition, the RTX 4080 itself appears to be getting a larger heatsink, on par with the RTX 3090.
The Founders Edition or FE cooler appears to have a dual-slot design, but it comes with a thicker package with a wider cooler and shroud and a larger fan, including 7 blades, while the RTX 3080 and RTX 3090 coolers are There are 9 blades on the fan. The card also appears to have a slightly different “RTX 4080” logo than the RTX 3080. It does appear that NVIDIA is planning to reuse its existing cooler and modify it, a bit like the RTX 4090 Ti shroud GPU we saw earlier for Ada Lovelace.
NVIDIA GeForce RTX 4080 “Expected” Specs
The NVIDIA GeForce RTX 4080 is expected to use a scaled-down AD103-300 GPU configuration, enabling 9,728 cores or 76 SMs in a total of 84 units, compared to the previous configuration offering 80 SMs or 10,240 cores. While the full GPU comes with 64 MB of L2 cache and up to 224 ROPs, the RTX 4080 may end up with 48 MB of L2 cache and a lower ROP due to its stripped-down design. The card is expected to be based on the PG136/139-SKU360 PCB.
As for memory specs, the GeForce RTX 4080 is expected to have 16 GB GDDR6X capacity, which is said to be adjustable at 23 Gbps over a 256-bit bus interface. This will provide up to 736 GB/s of bandwidth. That’s still a bit slower than the 760 GB/s bandwidth offered by the RTX 3080, which has a 320-bit interface, but only 10 GB in capacity. To compensate for the lower bandwidth, NVIDIA may integrate a next-generation memory compression suite to compensate for the 256-bit interface.
Power-wise, the TBP is now rated at 340W, a 20W increase over the 320W spec we previously got. This puts the TBP on par with existing RTX 3080 graphics cards (up to 350W). It’s unclear if other RTX 40-series graphics cards will also get faster GDDR6X memory processing, but we do know that Micron has already started full-scale mass production of GDDR6X memory modules up to 24 Gbps, so they have to go somewhere.
- NVIDIA GeForce RTX 4080 “expected” TBP – 340W
- NVIDIA GeForce RTX 3080 “official” pending – 350W
NVIDIA GeForce RTX 4080 Series Preliminary Specifications:
|Graphics card name||NVIDIA GeForce RTX 4080Ti||Nvidia GeForce RTX 4080||Nvidia GeForce RTX 3090 Ti||Nvidia GeForce RTX 3080|
|GPU name||Ada Lovelace AD102-250?||Ada Lovelace AD103-300?||Ampere GA102-225||Ampere GA102-200|
|process node||TSMC 4N||TSMC 4N||Samsung 8nm||Samsung 8nm|
|transistor||pending||pending||28 billion||28 billion|
|TMU/ROP||Pending / 232?||Pending / 214?||320 / 112||272/96|
|Tensor/RT Core||pending / pending||pending / pending||320 / 80||272 / 68|
|base clock||pending||pending||1365 MHz||1440 MHz|
|boost clock||~2600 MHz||~2500 MHz||1665 MHz||1710 MHz|
|FP32 computing||~55TFLOPs||~50 TFLOPs||34 TFLOPs||30 TFLOPs|
|RT TFLOPs||pending||pending||67 TFLOPs||58 TFLOPs|
|Tensor-TOP||pending||pending||273 likes||238 likes|
|Memory Capacity||20GB GDDR6X||16GB GDDR6X||12GB GDDR6X||10GB GDDR6X|
|memory bus||320 bits||256 bits||384 bit||320 bits|
|memory speed||21.0 Gbps?||21.0 Gbps?||19Gbps||19Gbps|
|bandwidth||840 GB/sec||736 2GB/sec||912Gbps||760Gbps|
|start (availability)||2023?||July 2022?||June 3, 2021||September 17, 2020|
The NVIDIA GeForce RTX 4080 and RTX 4070 graphics cards will be among the first cards to roll out to gamers other than the RTX 4090, but we can’t say for sure whether the green team plans to launch them soon after the RTX 4090 or wait until they launch earlier next year. As of now, the RTX 4090 is expected to launch around October 2022.
News source: Videocardz
Products mentioned in this article