The Nvidia RTX 4070, launched in April 2023, faced mixed reviews on YouTube, with criticisms centered around its pricing, modest performance improvement over the previous generation, and its high cost relative to other cards. Despite these concerns, it has become Nvidia’s most popular RTX 4000-series desktop GPU according to the December 2023 Steam Hardware Survey.
The recently released RTX 4070 Super, priced at $599, aims to address these concerns by offering performance more aligned with the RTX 3090 and maintaining impressive power efficiency. Despite the high cost compared to gaming consoles or AMD’s Radeon 7800 XT, the 4070 Super is seen as a positive development in the market. A bit pricey but it’s alright if it can prove its worth.
Overview
The RTX 4070 Super outperforms the RTX 4070 by about 15% in DLSS and 10–20% in gameplay and visuals. With optimal performance at 1440p, the RTX 4070 Super is regarded as a well-rounded gaming option, offering everything from entry-level 4K gaming and video editing to high-quality 1080p gaming. Although there have been improvements, it is still debatable if the RTX 4070 Super is better than the RTX 4070 because the former is still slightly less expensive.
The RTX 4070 Super features 20% more cores and computing components than its predecessor, but other than that, not much has changed. One of the small improvements is the Founders Edition card’s new gloss, which gives it a more reflective look. Due to its size and power requirements, the RTX 4070 Super is considered appropriate for midrange systems.
In 4K gaming, the RTX 4070 Ti performs far better than the 4070 Super in comparison, and the impending 4070 Ti Super is anticipated to close the gap even more, albeit at a higher price point—at least $200 more than the Super. After the Super is released, the earlier 4070 Ti might still be accessible, but the GPU model will no longer be produced. The older 4070 Ti is recommended if it costs $650 or less, as it is deemed to be a better option than the 4060 Ti, unless there are system or financial limits that need a different decision.
What You Get In RTX 4070 Super Founders’ Edition
Memory | 12GB GDDR6X |
---|---|
Memory bandwidth (GBps) | 504 |
Memory clock (GHz) | 10.5 |
GPU clock (GHz, base/boost) | 1.98/2.475 |
Memory data rate/Interface | 21Gbps/192 bits |
RT cores | 56 |
CUDA cores | 7,168 |
Texture mapping units | 224 |
Streaming multiprocessors | 56 |
Tensor cores | 224 |
Process | 4nm |
TGP/min PSU (watts) | 220/650 |
Max thermal (degrees) | 194F/90C |
Bus | PCIe 4.0×16 |
Size | 2 slots |
Launch price | $600 |
Ship date | Jan. 17, 2024 |
Boost In CUDA Scores & Cache Memory
The inclusion of 1,280 more CUDA cores, approximately 22% more than the standard 4070, is what makes the 4070 exceptional. In terms of both absolute core counts and percentage increases, that is a greater boost in processing power than either the 4080 Super or 4070 Ti Super witnessed. Although it’s not always a 22% increase in game performance, it’s still a pleasant improvement and much closer to the AD104 GPU die’s maximum core count of 7,680.
The cache, which is increased from 36MB in the 4070 to 48MB in the 4070 Super, and the card’s total graphics power (TGP), which can now reach a maximum of 220 W, are the only other characteristics that truly improved. The difference is negligible enough that Nvidia’s power supply wattage requirement stays at 650 W.
RTX 4090 | RTX 4080 | RTX 4080 SUPER | RTX 4070 TI | RTX 4070 TI SUPER | RTX 4070 | RTX 4070 SUPER | |
---|---|---|---|---|---|---|---|
CUDA CORES | 16,384 | 9,728 | 10,240 | 7,680 | 8,448 | 5,888 | 7,168 |
BOOST CLOCK | 2,520 MHz | 2,505 MHz | 2,550 MHz | 2,610 MHz | 2,610 MHz | 2,475 MHz | 2,475 MHz |
MEMORY BUS WIDTH | 384-bit | 256-bit | 256-bit | 192-bit | 256-bit | 192-bit | 192-bit |
MEMORY CLOCK | 1,313 MHz | 1,400 MHz | 1,437 MHz | 1,313 MHz | 1,313 MHz | 1,313 MHz | 1,313 MHz |
MEMORY SIZE | 24GB GDDR6X | 16GB GDDR6X | 16GB GDDR6X | 12GB GDDR6X | 16GB GDDR6X | 12GB GDDR6X | 12GB GDDR6X |
TGP | 450 W | 320 W | 320 W | 285 W | 285 W | 200 W | 220 W |
Other Improved Features of the Nvidia 4070 Super Graphics Card
Memory and GPU core rates remain the same, while the 4070 Super maintains the same 192-bit memory interface and 12GB bank of memory as the standard 4070. Although cards from Nvidia’s partners may have modest overclocks and 8-pin power connectors, we tested the Founders Edition version of the card for this evaluation, which adheres to Nvidia’s baseline specs and uses the 12VHPWR connector.
Now that the 40-series has been around for a while, you have more options than a year ago. The included 12VHPWR adapter only needs two 8-pin connectors attached, or you can connect it directly to a compatible power supply. ATX 3.0 supplies support 12VHPWR natively, but I recently bought an adapter cable for my gaming PC’s modular Cooler Master PSU that plugs into the 8-pin connectors on the power supply.
Unlike the 4080 and 4090 Founders Edition cards and a plethora of triple-fan, triple-slot third-party cards, Nvidia’s Founders Edition card is the same size and shape as the original 4070, thus it should fit in most tiny ITX designs with little difficulty. The black accents have replaced the silver ones, which is the only noticeable design change. This gives the card a more of “this is my final form” appearance. In the same way, we anticipate that the majority of Nvidia’s partners will mostly stick to their 4070 card designs.
RTX 4070 Super: Gaming Performance Beast?
The 4070 Super aims to compete with more expensive 3000-series cards such as the RTX 3080 Ti and RTX 3090. In many games, both of these 30-series cards perform quite similarly; however, since the 3080 Ti also has 12GB of RAM rather than 24GB, it may be a better reference card for the 4070 Super. Based on our testing, the 4070 Super achieves that performance with almost two-thirds less power consumption.
We’ve tested these GPUs using a range of DirectX 11 and DirectX 12 games, concentrating on 1440p and 4K resolutions because that’s what the card is meant for. In addition, we have charts for games like Cyberpunk 2077 and Returnal that use a lot of ray-tracing effects, as well as charts that show how well DLSS/FSR and DLSS Frame Generation perform. Take a reference from below mentioned charts,
Though you may not often reach that 22 percent figure (we did once, in a 4K Cyberpunk 2077 test with ray tracing enabled), adding 22 percent more CUDA cores without altering most of the card’s other characteristics generally makes the 4070 Super about 15 percent faster in most games. However, our test suite also has a large number of samples with lower performance benefits, in the 8–10 percent range. Overall, this is a pleasant improvement.
In-depth comparisons of the 4070 Super and 3080 Ti reveal many titles in which the card speeds are comparable. In fact, at 1440p, the 4070 Super typically matches or barely beats the 3080 Ti in terms of frame rate, as evident in above graphs
Benefit of DLSS Frame Generation In RTX 4070 Super
However, when the resolution is increased to 4K, the 3080 Ti’s edge becomes more reliable. The 4070 Super maintains its performance for the most part, plus it can rely on the added advantage of DLSS Frame Generation. However, at higher resolutions, the 4070 Super’s 48MB of cache is unable to fully bridge the gap with the 3080 Ti’s 384-bit memory interface.
Regarding DLSS FG, the 4070 Super is a rather perfect card for it. Your GPU should generally be able to achieve a respectable base frame rate to achieve acceptable results; this is especially true for the 4070 Super at 1440p and 4K, as opposed to the 4060 Ti or 4060. While there will still be some odd visual effects (Returnal’s blinking rainstorm is still a good, reproducible example), it’s worth enabling in more demanding games, such as Cyberpunk 2077, to see if the increased frame rate is compensated for by more input lag or glitches in the visuals.
Let’s last compare the AMD Radeon RX 7800 XT, which is the 4070 Super’s strongest rival in this segment. The 7800 XT can generally match or surpass the 4070 Super’s performance in games without ray-tracing effects enabled, and it is typically available for just a little bit more than its $500 MSRP. Though generally true, it’s not always so. As is customary, however, AMD performs worse in titles that make extensive use of ray tracing; in Cyberpunk, Returnal, Hitman III, and the 3DMark ray tracing benchmarks, it lags much below the 4070 Super (or even the standard 4070).
However, 4070 Super Remains Power-Efficient GPU
The 4070 Super is still a very power-efficient GPU, just like the 4070. In our tests, it used about 10 to 15 percent more power than the regular 4070 when playing games, which is about the same amount of speed boost you would normally expect. Running the same games at the same settings on the 7800XT uses about 30 W more power. As always, software-reported power usage comparisons between manufacturers should be taken with a grain of salt, but these numbers match what both Nvidia and AMD say in their spec sheets.
To Summarize It Up
RTX 4070 Super is a clear winner for most GPU customers. With stronger performance than the 4070, this $600 card competes with the 3080 Ti and 3090 ($1,199 and $1,499 at launch). Games still suffer from the 4070 Super’s 192-bit memory interface, especially at 4K, compared to the 3080 Ti. But architectural improvements and more processing power let the 4070 Super catch up.
Most GPU buyers are relieved that we’re over the transitional phase of current graphics generation when you had to consider both the competition and the last-gen GeForce and Radeon cards remaining on shelves at reasonable rates. It’s not worth considering 6800 XT or 3080 Ti open-box stock at this point.
The 4070 Super still leaves a large gap in Nvidia’s GPU lineup, since none are cost-effective. Between the 4070 Super and the $299 RTX 4060 are the $399 8GB 4060 Ti, $449 16GB, and $549 standard RTX 4070.
For 33 percent more money, the 4060 Ti is 10–20% quicker than the 4060, and for 9 percent more money, the 4070 Super is 12–22% faster. Either card is one good promotional reduction away from being a good purchase, but at their regular pricing, try the 4060, spend extra on the 4070 Super, or go with AMD.
AMD continues its competitive pricing: pay less for a card with strong rasterized performance but higher power utilization and average ray tracing. Despite its $500 price, the 7800 XT can deliver high-frame-rate 1440p or 60 fps 4K. Power consumption, improved ray-tracing performance, and DLSS compatibility prefer the 4070 Super or ordinary 4070, but if you don’t care about those things, go with the AMD card and save money.
Also Checkout our reviewed comparision betweeen RTX 4070 Super vs RTX 4070,
Leave a Reply