{"id":593884,"date":"2023-01-02T18:00:00","date_gmt":"2023-01-03T00:00:00","guid":{"rendered":"https:\/\/news.sellorbuyhomefast.com\/index.php\/2023\/01\/02\/kfa2-geforce-rtx-4080-sg-desktop-gpu-review-great-performance-meets-questionable-pricing-notebookcheck-net-reviews\/"},"modified":"2023-01-02T18:00:00","modified_gmt":"2023-01-03T00:00:00","slug":"kfa2-geforce-rtx-4080-sg-desktop-gpu-review-great-performance-meets-questionable-pricing-notebookcheck-net-reviews","status":"publish","type":"post","link":"https:\/\/newsycanuse.com\/index.php\/2023\/01\/02\/kfa2-geforce-rtx-4080-sg-desktop-gpu-review-great-performance-meets-questionable-pricing-notebookcheck-net-reviews\/","title":{"rendered":"KFA2 GeForce RTX 4080 SG desktop GPU review: Great performance meets questionable pricing – NotebookCheck.net Reviews"},"content":{"rendered":"
\n
\n

FPS and RGB lighting galore!<\/b> Alongside the GeForce RTX 4090, Nvidia has released the GeForce RTX 4080, a slightly less expensive model based on the company\u2019s latest Ada Lovelace architecture. Nonetheless, manufacturers are asking more than \u20ac1,000 (US$1,070) for the second fastest graphics card from team green. We have now taken a closer look at an RTX 4080 model made by KFA2. In this review, you will find out where this card stands in terms of performance.<\/p>\n<\/div>\n

\n
\"KFA2
KFA2 GeForce RTX 4080 Serious Gaming<\/figcaption><\/figure>\n<\/div>\n
\n
\n
\"External<\/a>
External connectors on the KFA2 GeForce RTX 4080 SG<\/figcaption><\/figure>\n<\/div>\n
\n

After we recently reviewed the KFA2 GeForce RTX 4090 SG<\/a>, it is now time for the smaller KFA2 GeForce RTX 4080 SG to showcase what it has got under the hood. Speaking of under the hood, unlike the Ampere lineup, the second fastest model of the current generation is based on a different graphics chip from the one on the top-end model. It uses the AD103 GPU, which is substantially smaller and thus cheaper to produce than the AD102. However, the lower production cost isn’t fully reflected in the current pricing. This is a widely discussed issue, and we too believe that prices must be reduced significantly so that the latest graphics cards don’t end up becoming shelf warmers. The RRP of the GeForce RTX 4080 has more than doubled compared with its predecessor (GeForce RTX 3080). We are going to leave our criticism of the pricing here for the time being. From this point on, we want to instead focus on the performance and features of the KFA2 GeForce RTX 4080 SG and discuss how well this model stacks up particularly against AMD’s RDNA3 GPUs<\/a>.<\/p>\n

Although KFA2 is often one of the manufacturers with the lowest prices, the company considers itself a competitor of other manufacturers offering more expensive cards, such as Asus with its Strix series and MSI with its Suprim-X models. Unfortunately, we weren’t able to compare the different cards head-to-head because we have yet to get our hands on any Asus or MSI models. Nevertheless, we can say that the KFA2 GeForce RTX 4080 SG has the same level of workmanship as the KFA2 GeForce RTX 4090 SG. Despite having a much lower rated power usage than the top model, the KFA2 GeForce RTX 4080 SG is fitted with an enormous quad-slot cooler. There is also a ton of RGB lighting that draws attention to the card. As with all the current RTX 40 Series models, the RTX 4080 is powered via the new 12VHPWR connector. KFA2 hasn’t done anything differently here and has included an appropriate 3×6+2 pin PCIe to 12VHPWR adapter.<\/p>\n

On the outside, the KFA2 GeForce RTX 4080 SG features 1x HDMI 2.1a port and 3x DisplayPort 1.4a.<\/p>\n<\/div>\n<\/div>\n

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
Specification<\/td>\nRTX 4090 (FE)<\/td>\nRTX 4080<\/td>\nKFA2 RTX 4080<\/b><\/td>\nRTX 3090 Ti<\/td>\nRTX 3090 (FE)<\/td>\nRTX 3080 Ti (FE)<\/td>\nRTX 3080 (FE)<\/td>\nTitan RTX<\/td>\n<\/tr>\n
Chip<\/td>\nAD102<\/td>\nAD103<\/td>\nAD103<\/b><\/td>\nGA102<\/td>\nGA102<\/td>\nGA102<\/td>\nGA102<\/td>\nTU102<\/td>\n<\/tr>\n
FinFET process<\/td>\n4 nm<\/td>\n4 nm<\/td>\n4 nm<\/b><\/td>\n8 nm<\/td>\n8 nm<\/td>\n8 nm<\/td>\n8 nm<\/td>\n12 nm<\/td>\n<\/tr>\n
CUDA cores<\/td>\n16.384<\/td>\n9.728<\/td>\n9.728<\/b><\/td>\n10.752<\/td>\n10.496<\/td>\n10.240<\/td>\n8.704<\/td>\n4.608<\/td>\n<\/tr>\n
Texture units<\/td>\n512<\/td>\n304<\/td>\n304<\/b><\/td>\n336<\/td>\n328<\/td>\n320<\/td>\n272<\/td>\n288<\/td>\n<\/tr>\n
Tensor cores<\/td>\n512 4. Gen.<\/td>\n304 4. Gen.<\/td>\n304 4. Gen.<\/b><\/td>\n336 3. Gen.<\/td>\n328 3. Gen.<\/td>\n320 3. Gen.<\/td>\n272 3. Gen.<\/td>\n576 2. Gen.<\/td>\n<\/tr>\n
RT cores<\/td>\n128<\/td>\n76<\/td>\n76<\/b><\/td>\n84<\/td>\n82<\/td>\n80<\/td>\n68<\/td>\n72<\/td>\n<\/tr>\n
GPU base clock<\/td>\n2,235 MHz<\/td>\n2,205 MHz<\/td>\n2,205 MHz<\/b><\/td>\n1,560 MHz<\/td>\n1,395 MHz<\/td>\n1,365 MHz<\/td>\n1,440 MHz<\/td>\n1,350 MHz<\/td>\n<\/tr>\n
GPU boost clock<\/td>\n2,520 MHz<\/td>\n2,505 MHz<\/td>\n2,565 MHz<\/b><\/td>\n1,860 MHz<\/td>\n1,695 MHz<\/td>\n1,665 MHz<\/td>\n1,710 MHz<\/td>\n1,770 MHz<\/td>\n<\/tr>\n
Memory bus<\/td>\n384 bit<\/td>\n256 bit<\/td>\n256 bit<\/b><\/td>\n384 bit<\/td>\n384 bit<\/td>\n384 bit<\/td>\n320 bit<\/td>\n384 bit<\/td>\n<\/tr>\n
Memory bandwidth<\/td>\n1,008 GB\/s<\/td>\n716.8 GB\/s<\/td>\n716.8 GB\/s<\/b><\/td>\n1,008 GB\/s<\/td>\n936 GB\/s<\/td>\n912.4 GB\/s<\/td>\n760 GB\/s<\/td>\n672 GB\/s<\/td>\n<\/tr>\n
Video memory<\/td>\n24 GB GDDR6X<\/td>\n16 GB GDDR6X<\/td>\n16 GB GDDR6X<\/b><\/td>\n24 GB GDDR6X<\/td>\n24 GB GDDR6X<\/td>\n12 GB GDDR6X<\/td>\n10 GB GDDR6X<\/td>\n24 GB GDDR6<\/td>\n<\/tr>\n
Power consumption<\/td>\n450 W<\/td>\n320 W<\/td>\n320 W<\/b><\/td>\n450 W<\/td>\n350 W<\/td>\n350 W<\/td>\n320 W<\/td>\n280 W<\/td>\n<\/tr>\n<\/table>\n<\/div>\n
\n

Our test system features a Gigabyte X670E Aorus Master motherboard paired with a high-performance AMD CPU (Ryzen 9 7950X<\/a>), which is cooled by an Alphacool AIO solution (Alphacool Eisbaer Aurora 360 mm<\/a>). The system has 32 GB of G.Skill memory running on an appropriate EXPO profile and is connected to a Gigabyte M32U monitor.<\/p>\n<\/div>\n

\n

We used the following system for the benchmarks and tests in this review:<\/p>\n