RTX 4080 has been on the market for about three weeks now, and today we're looking at another partner card. Palit's RTX 4080 GameRock OC carries over the blinged-out design from the RTX 4090 GameRock OC we reviewed in October, while also featuring an adjustable power limit that can reach up to 400W, alongside dual-BIOS support. It certainly has a lot going for it on paper, but is it worth buying?
Palit's RTX 4080 GameRock OC is certainly an eye-catching card, with its full-on RGB implementation that will probably end up a bit like marmite – you'll either love it or you'll hate it. Currently priced at £1299.95, it's only £30 more expensive than Nvidia's baseline MSRP, and considering the features on offer, it certainly looks tempting.
But with the arrival of AMD's RDNA 3-based GPUs now just two weeks away, is this RTX 4080 actually worth buying? Let's find out…
RTX 4090 | RTX 4080 | RTX 3090 Ti | RTX 3090 | RTX 3080 Ti | RTX 3080 | |
Process | TSMC N4 | TSMC N4 | Samsung 8N | Samsung 8N | Samsung 8N | Samsung 8N |
SMs | 128 | 76 | 84 | 82 | 80 | 68 |
CUDA Cores | 16384 | 9728 | 10752 | 10496 | 10240 | 8704 |
Tensor Cores | 512 | 304 | 336 | 328 | 320 | 272 |
RT Cores | 128 | 76 | 84 | 82 | 80 | 68 |
Texture Units | 512 | 304 | 336 | 328 | 320 | 272 |
ROPs | 176 | 112 | 112 | 112 | 112 | 96 |
GPU Boost Clock | 2520 MHz | 2505 MHz | 1860 MHz | 1695 MHz | 1665 MHz | 1710 MHz |
Memory Data Rate | 21 Gbps | 22.4 Gbps | 21 Gbps | 19.5 Gbps | 19 Gbps | 19 Gbps |
L2 Cache | 73729 KB | 65536 KB | 6144 KB | 6144 KB | 6144 KB | 5120 KB |
Total Video Memory | 24GB GDDR6X | 16GB GDDR6X | 24GB GDDR6X | 24GB GDDR6X | 12GB GDDR6X | 10GB GDDR6X |
Memory Interface | 384-bit | 256-bit | 384-bit | 384-bit | 384-bit | 320-bit |
Memory Bandwidth | 1008 GB/Sec | 716.8 GB/Sec | 1008 GB/Sec | 936 GB/Sec | 912 GB/Sec | 760 GB/Sec |
TGP | 450W | 320W | 450W | 350W | 350W | 320W |
Above: Performance BIOS, left, Silent BIOS, right.
First, for a quick spec recap. RTX 4080 uses the AD103 die, and this is notably smaller than AD102, measuring 378.6mm2. Accordingly, transistor count is reduced from 76.3 billion, down to 45.9 billion. The fundamental building blocks are still the same of course, with the RTX 4080 offering a total of 84 Streaming Multiprocessors (SMs), each housing 256 CUDA Cores, for a total of 9728. We also find 84 RT cores, 336 Tensor cores, 336 Texture Units, and 112 ROPs.
TSMC's N4 node has Nvidia cranking up the clock speed significantly this generation, with the RTX 4080 sporting a 2505MHz rated boost clock. Palit has increased this further with a 35MHz factory overclock when using the Performance BIOS, though the Silent BIOS runs at reference speed.
The memory configuration is another area where AD103 has been cut-back significantly. The memory interface has been reduced to 256-bit, and even with 16GB GDDR6X running at 22.4Gbps, that cuts memory bandwidth down to 716.6GB/s. That said, there has been a substantial upgrade to the L2 cache with the Ada architecture, with the RTX 4080 now offering 65.5MB, compared to just 6MB for GA102.
Considering RTX 4080 is notably cut-down versus the RTX 4090, rated power draw is lower, with a 320W TGP. Palit has increased this to 340W for both the Performance BIOS and the Silent BIOS.