NVIDIA GeForce RTX 5070 Content Creation Review

Introduction

At CES, NVIDIA announced its next generation of consumer desktop graphic cards: the GeForce RTX™ 50 series, based on the Blackwell architecture, including the recently launched NVIDIA GeForce RTX™ 5070. As we covered a couple of weeks ago in our RTX 5070 Ti review, these GPUs promise improved performance in gaming and content creation and feature more and improved CUDA cores, fourth-generation Ray Tracing and fifth-generation Tensor cores, and the latest NVIDIA NVENC/NVDEC media engines. These hardware updates are packaged alongside a multitude of new technologies like DLSS 4, RTX Mega Geometry, and new NVIDIA Broadcast features.

In this article, we will be reviewing the new midrange NVIDIA GeForce RTX 5070. Replacing the outgoing NVIDIA GeForce RTX 4070 SUPER, the 5070 is likely to be the lower end of NVIDIA’s mid-range graphics card stack. 70-class cards usually are the everyman’s card, offering a good balance between performance and price, though whether the -70 or -70 Ti is a better value seems to change with each generation.

NVIDIA has introduced a number of new features and capabilities with this GPU generation, such as Neural rendering, Mega geometry, and DLSS 4 plus MFG. While we won’t be covering those in this article, we hope to do so in the future. Nonetheless, many of them are exciting for both gamers and professionals, and we continue to be impressed by NVIDIA’s software and feature support. If you are interested in the new features for content creators, we have a blog post available that summarizes everything NVIDIA has announced to date: NVIDIA GeForce RTX 50-Series | Features for Content Creators.

Below, we have listed the most relevant GPU specifications from AMD, Intel, and NVIDIA. For more information, visit Intel Ark, NVIDIA’s 40-series GeForce page, NVIDIA’s 50-series GeForce page, or AMD’s Radeon RX Page.

GPU Model	MSRP	VRAM	Shader Units	Boost Clock	VRAM Bandwidth	TDP	Release Date
NVIDIA GeForce RTX 5090	$2,000	32 GB	21760	2.41 GHz	1792 GB/sec	575 W	Jan. ’25
NVIDIA GeForce RTX 3090 Ti	$2,000	24 GB	10752	1.86 GHz	1001 GB/sec	450 W	Jan. ’22
NVIDIA GeForce RTX 4090	$1,600	24 GB	16384	2.52 GHz	1001 GB/sec	450 W	Oct. ’22
NVIDIA GeForce RTX 4080	$1,200	16 GB	9728	2.51 GHz	736 GB/sec	320 W	Nov. ’22
NVIDIA GeForce RTX 3080 Ti	$1,200	12 GB	10240	1.67 GHz	912 GB/sec	350 W	June ’21
AMD Radeon RX 7900 XTX	$1,000	24 GB	6144	2.5 GHz	960 GB/sec	355 W	Dec. ’22
NVIDIA GeForce RTX 5080	$1,000	16 GB	10752	2.62 GHz	960 GB/sec	360 W	Jan. ’25
NVIDIA GeForce RTX 4080 SUPER	$1,000	16 GB	10242	2.55 GHz	736 GB/sec	320 W	Jan. ’24
NVIDIA GeForce 2080 Ti	$1,000	11 GB	4352	1.55 GHz	616 GB/sec	250 W	Sept. ’18
NVIDIA GeForce RTX 4070 Ti SUPER	$800	16 GB	8448	2.61 GHz	706 GB/sec	285 W	Jan. ’24
NVIDIA GeForce RTX 5070 Ti	$750	16 GB	8960	2.45 GHz	896 GB/sec	300 W	Feb. ’25
NVIDIA GeForce RTX 4070 SUPER	$600	12 GB	7168	2.48 GHz	504 GB/sec	220 W	Jan. ’24
NVIDIA GeForce RTX 3070 Ti	$600	8 GB	6144	1.77 GHz	608 GB/sec	290 W	May. ’21
NVIDIA GeForce RTX 5070	$550	12 GB	6144	2.51 GHz	672 GB/sec	250 W	March ’25
NVIDIA GeForce RTX 3070	$500	8 GB	5888	1.73 GHz	448 GB/sec	220 W	Oct. ’20
Intel Arc B580	$250	12 GB	2560	2.67 GHz	456 GB/sec	190 W	Dec. ’24

Unlike the prior 50-series cards, the RTX 5070 seems, in some ways, more like an upgraded 3070 than an upgraded 4070 SUPER. Indeed, from the 4070 SUPER, the RTX 5070 features 15% fewer CUDA cores alongside a marginal 37 MHz higher boost clock. The base clock is 345 MHz higher, so we may see higher average clocks, but in most cases we expect the boost clock to be a better predictive specification. On the positive side, the 5070 features an improved VRAM bus, with a throughput of 672 GB/s, compared to the 4070 SUPER’s 504 GB/s. However, this comes along with 30 W more power draw and the same amount of VRAM. In this day and age, 12 GB of VRAM on a $550 card is too little for too much, and we would have hoped for 16 GB. Overall, despite a nominal $50 price reduction, we are unimpressed with the 5070 on paper—we don’t expect much from the card over the 4070 SUPER in either performance or value.

Test Setup

For our GPU testing, we have shifted to an AMD Ryzen 9 9950X-based platform from our traditional Threadripper platform. The 9950X has fantastic all-around performance in most of our workflows and should let the video cards be the primary limiting factor where there is the possibility of a GPU bottleneck. This means the results are more comparable to our recent Intel Arc B580 review but less so to our past GPU reviews. However, at this point in our 50-series testing, we have benchmarked most of the GPUs from the last few generations, so there should be plenty of comparative data for most needs. For testing, we used the latest available GPU drivers, though as we are re-using data from our previous 50-series testing in this review, we do have a few different drivers in the mix; we have not seen large performance differences between them. We tested everything on the “balanced” Windows power profile, while Resizeable BAR and “Above 4G Decoding” were enabled for every GPU as well.

In this article, our primary focus will be the new 5070, so we tested it against the most comparable cards from the last few generations: the RTX 4070 SUPER and RTX 3070. This is in addition to all the cards we tested in our previous 50-series reviews, so we still have most of the RTX 30- and 40-series cards, in addition to the 2080 Ti and 7900 XTX. We think that the 4070 SUPER is the best comparison from the last generation, and from the 30-series cards, we’re interested in seeing how it compares to both the 3070 Ti and 3070.

Unlike the 5070 Ti, the 5070 has a first-party FE (Founders Edition) model available from NVIDIA. We will be testing with that card as it should be representative of the baseline performance you get from any 5070, without additional factory overclocking or possibly-questionable coolers. We are still using the ASUS GPUs for the 5070 Ti, 4070 Ti, and 3070 Ti we mentioned in our previous review, as we have been generally impressed with the quality and reliability of ASUS’ cards and use them extensively in our own systems and in past testing.

In terms of applications, the new NVIDIA Blackwell cards have some lingering compatibility issues at present as we await developers’ integration of the new CUDA 12.8 and TensorRT 10.8 toolkits. As a result, the RTX 50-series of graphics cards is not supported in Redshift (Cinebench) or Octanebench—though the latest version of the core Octane renderer does support them—and has performance issues in V-Ray CUDA rendering. Due to this, we have slightly fewer results than is typical for our GPU reviews: PugetBench for Premiere Pro, After Effects, DaVinci Resolve, and Lightroom Classic, as well as Unreal Engine, V-Ray, and Blender.

Raw Results Tables

We choose our benchmarks to cover many workflows and tasks to provide a balanced look at the application and its hardware interactions. However, many users have more specialized workflows. Recognizing this, we like to provide individual results for benchmarks as well. If a specific area in an application comprises most of your work, examining those results will give a more accurate understanding of the performance disparities between components. Otherwise, we recommend skipping over this section and focusing on our more in-depth analysis in the following sections.

Photography: Lightroom Classic

One of the more frequent requests we get for software testing is Adobe’s Lightroom Classic. Although we have a benchmark for the application, at present, it tests relatively few GPU-accelerated features and so is typically a lower priority for us during GPU reviews. Nonetheless, we understand that even seeing things like GPU-accelerated image export times can be useful in purchase decisions for heavy LRC users, so we do our best to include Lightroom testing where possible. Additionally, we hope to have a revamped Lightroom Classic benchmark finished later this year, which should include many more tests and be more in line with modern Lightroom workflows.

In the overall score (Chart#1), the 5070 is on-par with the RTX 4090, although the majority of the tests in Lightroom Classic are unaccelerated, so the overall score really isn’t all that meaningful since almost every GPU performs within the margin of error.

When looking at a GPU-accelerated portion of the benchmark, image exporting (Chart #2), there is a larger difference between cards. Here, the 5070 is well ahead of the 30-series cards, leads the 4070 SUPER by 3%, and falls behind the 4070 Ti SUPER by 5%. The latter two are within the margin of error, especially since Lightroom tends to have a larger amount of variance in its tests.

Video Editing: Adobe Premiere Pro

Premiere Pro is a much more GPU-accelerated application than Lightroom Classic. In it, the RTX 5070 is essentially equivalent to a 4070 Ti SUPER and leads the base 4070 SUPER and 3070 by 6% and 18%, respectively. This is even given that Adobe has yet to fully add support for the new media engines of the 50-series cards, meaning we expect this gap to grow in the future.

Our LongGOP tests (Chart #2) are where we would expect the media engines to play a large role in elevating the performance of the 5070. Although NVIDIA’s next-generation NVENC/NVDEC are faster than last gen, making the 5070 6% faster than the 4070 SUPER and 17% faster than the 3070, when support for the new H.264/HEVC 4:2:2 10-bit acceleration is implemented by Adobe, we expect this lead to grow.

Intraframe codecs (Chart #3) are not GPU-accelerated, and predictably, we see no real differentiation between GPUs. However, we do see some in our RAW tests (Chart #4). RAW codecs use a mix of CPU and GPU, with the exact ratio changing based on the individual codec. On average, for RAW codec processing, the 5070 outperforms the 4070 SUPER by 2%; well within our margin of error. This does put it 27% ahead of the 3070, though, making it a good upgrade if you are currently a few generations behind.

The final set of tests in our Premiere Pro benchmark is for GPU Effects (Chart #5). Here, the 5070 nearly matches the 4070 Ti SUPER, with a score 15% ahead of the RTX 4070 SUPER and nearly three times the performance of the 3070. Overall, the 5070 offers a small upgrade over the outgoing 4070 SUPER, for, nominally, $50 less. This makes it a fine value at MSRP, and especially compelling as an upgrade from a midrange 30-series GPU.

Motion Graphics: Adobe After Effects

Though traditionally a CPU-based application, After Effects has added 3D capabilities that rely heavily on GPU acceleration. We recently updated our Pugetbench for After Effects benchmark to test these capabilities, and we are excited to be able to include After Effects in our GPU reviews.

In terms of overall performance (which includes both CPU and GPU-focused tasks), the 5070 scores 7% higher than the 4070 SUPER and 18% higher than the 3070. Two of our subscores, 2D (Chart #2) and Tracking (Chart #4), are principally CPU-bound, so we see no real difference in them. However, in 3D workflows (Chart #3), the 5070 extends the leads to 14% and 52% over the 4070 SUPER and 3070, respectively. These are both great generational and skip-generation uplifts.

For traditional After Effects usage, the graphics card is still largely irrelevant, and a cheaper card is typically preferred so that more of your budget can be allocated to CPU, RAM, or storage. However, if you are looking to explore the new 3D workflows in After Effects, the new 5070 can offer a compelling upgrade over older cards and is especially strong compared to AMD’s Radeon cards.

Video Editing / Motion Graphics: DaVinci Resolve Studio

As an application, DaVinci Resolve tends to be more sensitive to GPU performance than Premiere Pro or After Effects. Additionally, NVIDIA has provided a pre-release version of DaVinci Resolve to reviewers which is compatible with 50-series cards and properly uses the updated NVENC/NVDEC engines to accelerate additional “flavors” of LongGOP codecs, such as HEVC 4:2:2 10-bit.

Overall, this means that the 5070 scores 14% higher than the 4070 SUPER and 34% higher than the 3070. Much like Premiere Pro, though, there are some subtests that do not effectively make much use of the GPU, such as Intraframe (Chart #3), RAW (Chart #4), and Fusion (Chart #5). In the first of these, GPU has no effect, while in the latter two, the 5070 is about 11-15% faster than the 4070 SUPER and 30% faster than the 3070.

Our LongGOP tests (Chart #2) are perhaps the most interesting results in DaVinci Resolve, as this includes codec variants that were previously only accelerated with Intel Quick Sync. With the new support for HEVC 4:2:2 10-bit, the 5070 is able to have a 38% performance advantage over the 4070 SUPER, extending to 66% over the 3070. We will note that if you regularly work with H.264/HEVC codecs, it may be worth investing in the higher tier 50-series cards, as they offer additional NVENC (encoding) and/or NVDEC (decoding) chips.

GPU Effects (Chart #5) and AI (Chart #7) are our final categories of tests. In GPU Effects, the new 50-series card once again flexes its increased compute capabilities to establish a 15% lead over the 4070 SUPER and 51% over the 3070. However, these workloads benefit greatly from a more powerful GPU, so if you make use of a lot of GPU effects, the higher-end cards may be a better value depending on how much time they could save you. We continue to see odd behavior in the AI features of DaVinci Resolve, with the 5070 merely performing on par with the 4070 SUPER. This still allows it to outperform the 3070 by 28%, but we definitely hope future app or driver updates improve the neural engine performance in Resolve.

Topaz Video AI

Topaz Video AI is an application that uses AI models to upscale lower-resolution video to a higher-resolution output or low framerate to a higher framerate. We use the built-in benchmark to test performance across all of the included models at both 1080P and 4K and then combine those together into a singular overall score.

Unfortunately, Topaz Video AI has yet to be updated to fully enable 50-series support, so performance is disappointing compared to last-gen hardware. In fact, the new 5070 is slower than the 4070 SUPER by 22% but still manages to beat the older RTX 3070 by 22%. We would not recommend a 50-series card for Topaz or if you use it in your day-to-day work until the software is updated.

Game Dev / Virtual Production: Unreal Engine

Our Unreal Engine benchmark combines several scenes at varying resolutions and enabled features (e.g., Ray tracing) to see how various common factors affect GPU performance. We combine those FPS results together to get composite scores. Here, we decided to pull out Ray-tracing and Rasterized scenes as individual sub-scores to see how the new RT cores perform.

The Overall FPS (Chart #1) has the 5070 performing on par with a 4070 SUPER, which is a much worse showing than the other 50-series cards. It can still be a good upgrade over older 30-gen cards like the 3070, however, where it is 60% faster. Much like the rest of the 50-series cards, we see little relative change between ray-traced and rasterized performance. Once again, the 5070 is a great upgrade from the 30-series, but much less necessary if you already have a 4070 SUPER.

Rendering: V-Ray & Blender

For offline, GPU-based rendering, we could only test with two of our usual four benchmarks. At present, the 50-series cards have only experimental support in Redshift 2025.13 (but no compatibility with Cinebench 2024, which uses the Redshift renderer) and support in the OctaneRenderer 2025.1-beta 2 (but not in OctaneBench); as such, we were not able to test with either. Additionally, there is currently a known issue with CUDA rendering in V-Ray, resulting in low performance. We expect that this will be fixed as the implementation of the CUDA 12.8 toolkit is rolled out to applications, but it is definitely a note of caution for early adopters of the 50-series.

In V-Ray RTX rendering (Chart #1), the RTX 5070 scores 18% higher than the 4070 SUPER and nearly 2x higher than the 3070. This places it on par with the 4070 Ti SUPER. The 5070 is less impressive in Blender, only matching the RTX 4070 SUPER, although it still leads the 3070 Ti by 90%.

How good is the NVIDIA GeForce RTX 5070 for Content Creation?

Overall, the RTX 5070 seems to provide moderately better performance than the 4070 SUPER while offering better features at a nominally reduced price. However, MSRP is quickly becoming meaningless, which can quickly erode any advantages the card has over last-gen options and make price-to-performance comparisons something that can change daily.

In video editing and motion graphics, the RTX 5070 is only marginally faster than the RTX 4070 SUPER in most cases, but in GPU-specific workflows, it can be around 15% faster than the previous generation. It is much more attractive if you have an older card like the RTX 3070, which can be 12-34% faster overall and up to 50-60% faster in GPU-heavy situations.

In rendering applications, the 5070 is a mixed bag. It is often only on par with the previous generation RTX 4070 SUPER, although it is 60% to 2x faster than an older RTX 3070. However, there is still the lingering issue of compatibility and performance quirks, so we would recommend buying with caution or holding off for a bit before committing to a 5070 for a rendering system. We are currently maintaining a list of known issues in content creation applications that you can check in on to see when these are resolved.

In total, NVIDIA’s new GeForce RTX 5070 is an OK mid-tier GPU that promises decent performance and features for the price. If you are currently on a 30-series card or older and aren’t looking to break the bank, this could be an excellent option for professional use. The 12GB of VRAM is going to be a limitation if you want to get into heavier GPU-centric workflows like 6K+ video editing, or using Unreal Engine, but if what you are doing is more CPU-bound, or at low enough resolution to not need a ton of VRAM, this can be a great entry- to mid-level option.

---

If you need a powerful workstation to tackle the applications we’ve tested, the Puget Systems workstations on our solutions page are tailored to excel in various software packages. If you prefer to take a more hands-on approach, our custom configuration page helps you to configure a workstation that matches your exact needs. Otherwise, if you would like more guidance in configuring a workstation that aligns with your unique workflow, our knowledgeable technology consultants are here to lend their expertise.