GPU NVIDIA GeForce RTX 4090

amazon GPU NVIDIA GeForce RTX 4090 reviews

All in all, the new NVIDIA GeForce RTX 4090 24GB GPU has shown outstanding power and this is yet another big step forward in terms of NVIDIA GPU performance. The degree of performance improvement is highly dependent on each different application.

For video editing, the RTX 4090 can be 40% faster than the previous generation RTX 3090 and 3090 Ti, or nearly twice as fast as the older RTX 2080 Ti. The RTX 40 Series also offers a small performance boost for those using the GPU for hardware decoding or encoding of H.264 and HEVC files.

Working with the Unreal Engine, the RTX 4090 showed even greater performance gains, with the RTX 4090 giving content creators an approximately 85% FPS increase over the RTX 3090 and RTX 3090 Ti in all tests. Depending on the exact use case (ArchViz, virtual production, etc.), that means faster rendering, smoother performance, or increased detail.

Ultimately, GPU rendering is really where you’ll get the most out of your GPU, and the RTX 4090 has pushed through a lot in terms of performance power for complex rendering requirements. When using the RTX 4090, the GPU rendering of this graphics card will often be nearly twice as fast as the previous generation RTX 3090 or 3090 Ti or four times faster than the older RTX 2080 Ti.

With these performance gains, however, there are two major concerns. First, the new RTX 40 series (and their Quadro variants) seem to do away with NVLink entirely. This won’t affect the majority of users, but if you have a workflow that requires or benefits from an NVLink bridge, be aware that it won’t be an option. Second, the RTX 4090 requires four PCIe power plugs and has three full-width slots. This will limit you to a maximum of two external GPUs.

NVIDIA GPUs have long been a key component and a key element of workstations for content creation, as they offer a great combination of performance and stability. It is very reliable when used for a long time. Certain workflows are simply a requirement due to their proprietary CUDA technology. Now, they’re launching their latest GeForce RTX lineup: the NVIDIA GeForce RTX 4090 24GB.

While the RTX 4090 only has 24GB of VRAM like the previous generation RTX 3090 and 3090 Ti, NVIDIA is touting a huge boost in overall performance. However, most of the focus so far has been on gaming, so we’re looking forward to seeing how much faster the RTX 4090 is for content creation tasks like video editing, game development, and rendering using GPUs.

While the RTX 4090 costs a bit more than the previous-generation RTX 3090, the specs on paper are incredibly impressive, with a 56% increase in CUDA cores and a 46% increase in clocks. We’ll have to see how this translates to real-time performance, but it’s going to be significantly faster than any other NVIDIA GPU currently on the market.

Another change NVIDIA is making with the RTX 40 Series is the Dual NVIDIA Encoder (NVENC). This should allow these graphics cards to be significantly faster when encoding supported versions of H.264 and HEVC media, and we’re curious to see how well it performs.

Test Setup

To see how the RTX 4090 performs, we’ll see a comparison of the RTX 4090 with all of the RTX 3000 series GPUs as well as the older NVIDIA GeForce RTX 2080 Ti, plus the AMD Radeon RX 6900XT to get a feel for what is performance compared to NVIDIA’s main competitor. The test system that the Puget Systems team will use is one of the fastest available for most of the applications Puget Systems is testing and is built around the AMD Threadripper Pro 5975WX to minimize any unexpected CPU performance bottlenecks.

As for the nature of the tests, Puget Systems will primarily use their PugetBench benchmark series using the latest versions of the hosted applications. Most of these benchmarks include the ability to upload results to an online database of performance testers, so if you want to know how your system compares, you can download it yourself. Download and run the benchmark. Puget Systems’ testing was also added with Blender to demonstrate the GPU rendering performance of these graphics cards.

Video Editing: DaVinci Resolve Studio

To get a first look at the new RTX 4090 graphics card, we’ll test its performance in DaVinci Resolve Studio. More than any other NLE (Non-Linear Editor) currently on the market, Resolve can take full advantage of the power of high-end GPUs and even multi-GPU setups. The four main areas where GPUs are used are GPU effects processing, debugging (and optionally decoding) RAW media, H.264/HEVC decoding, and H.264/HEVC encoding.

For starters, performance testers want to look at the overall score, which is a combination of everything they test for Resolve. This is usually a good indicator of the kind of relative performance that a “typical” (if that exists) Resolve user might see with different hardware. Here, the RTX 4090 scores are about 9% higher than the RTX 3090 Ti or 10% higher than the RTX 3090.

However, this is the overall score, including some tasks that are either completely CPU controlled or often clogged by the CPU. If we look at GPU Score, we see how GPUs perform for tasks like OpenFX and noise reduction where the performance of the GPU itself is often the limiting factor. In this case, the RTX 4090 was 34% faster than the RTX 3090 Ti or 42% faster than the RTX 3090. The RTX 4090 couldn’t keep up with the dual RTX 3090 setup, but the dual RTX 4090 was 40% faster than the Dual RTX 3090.

This is also a good point to mention the older RTX 2080 Ti. The Puget Systems team put this graphics card (RTX 4090) to the test in their test to give a reference of the RTX 4090’s speed compared to the slightly older GPU, and for GPU effects, the RTX 4090. just over 2x faster than the RTX 2080 Ti.

Finally, we want to look specifically at the H.264 encoding. There’s no easy way to test this in a full vacuum (since you always have to decode or process the source clip), so we’ll take a look at the results of exporting the ProRes project to H.264 where the main bottleneck is the H.264 encoding part. The RTX 40 series includes dual encoders, which are said to significantly reduce output times when using NVDEC. Puget System didn’t see a huge increase in performance, but the RTX 4090 was able to complete “ProRes to H.264” almost exactly 25% faster. It should also be pointed out that this, and H.264 decoding, is one area where AMD can keep up with NVIDIA. Everywhere else (RED/BRAW processing, GPU Effects, etc.), NVIDIA has a strong lead.

The last major area where the GPU is used is RED/BRAW debug, but performance for that isn’t significantly different from the RTX 30 series in Puget System’s testing. Chances are, we’re simply CPU congested, so until we have even more powerful processors, having a faster GPU like the RTX 4090 won’t be able to show any benefit.

Video Editing: Adobe Premiere Pro

Adobe Premiere Pro may not be as GPU-intensive as DaVinci Resolve (and doesn’t make good use of multi-GPU setups), but having a powerful GPU can still make an impact depending on your workflow.

Again, we’ll start with the overall score from the test team’s Premiere Pro benchmark, which shows that the newly released RTX 4090 is only 7% faster than the RTX 3090 and RTX 3090 Ti. However, you’ll notice that there isn’t much of a performance difference between the RTX 3080 and the RTX 3090 Ti. There is a clear benefit to using NVIDIA, but most of the performance differences between different NVIDIA GPU models are true by generation line. RTX 40 is faster than RTX 30, faster than RTX 20 series. Most of this is because most of the tasks that Puget Systems tested in their benchmarks were affected more by the performance of the NVENC/NVDEC (NVIDIA encoder and decoder) on the card, rather than the raw performance of the GPU itself.

One area where a more powerful GPU from within a single generation can give you a boost is GPU-accelerated effects like Lumetri Color, most blurs, and more. For these tasks, the RTX 4090 is about 14% faster than the RTX 3090 Ti, or about 17% faster than the RTX 3090. That still doesn’t come anywhere close to the gains the Puget Systems team has seen in Resolve, but keep in mind that this is only with the base version of Premiere Pro. If you use plugins that take advantage of the GPU (such as many noise reduction plugins), the RTX 4090 can give you a much bigger boost.

Finally, Puget Systems is looking at the average of all their H.264 and HEVC results. The current version of their Premiere Pro benchmark doesn’t give a single score based on the source codec type, but this is something they are starting to do more and more often and will likely be included in an update. The next big thing in Puget systems is benchmark scores. As mentioned earlier by Puget systems, the performance for H.264/HEVC decoding/encoding usually depends more on GPU generation, than on raw performance because the NVENC and NVDEC parts of the GPU tend to be the same across all machine models. There’s certainly a small, but noticeable performance boost from the RTX 40 series, though, with the new RTX 4090 scoring about 13% higher than the RTX 30 series graphics cards.

Motion Graphics/VFX: Adobe After Effects

Adobe After Effects nearly failed to do the GPU testing part of the puget system team as a whole, as most projects in After Effects would be CPU-bound. However, the GPU sometimes comes into play.

In terms of overall performance, you won’t notice too much of a difference between the RTX 30 series and the RTX 4090. The RTX 4090 is just 2% of the RTX 3090, with the 3090 scoring higher overall. This is not to say that the 3090 is necessarily faster as the margin of error for this kind of real-world testing is around 5%. Instead, it means that any GPU in the RTX 30 or RTX 40 series should work pretty well for most users.

However, just like with Premiere Pro, we can dive into the “GPU Score” section of our benchmark. These are test projects specifically designed to put as much load on the GPU while minimizing the load on the rest of the system. This is a set of tests compositing at the frontier as no one has used After Effects this way but provides a “best case” scenario for what a GPU upgrade could bring for you in the base version of After Effects without taking into account plugins, GPU-based rendering engines, etc.

Even for GPU scores, the RTX 4090 is still within the margin of error compared to the RTX 3090 Ti and RTX 3090. With the powerful AMD Threadripper PRO 5975WX processor, we can see a benefit from RTX 2080 Ti or AMD Radeon 6900XT to RTX 3080, but beyond that, performance testers once again witnessed CPU performance drops.

where can you get a GPU NVIDIA GeForce RTX 4090 online

NVIDIA GeForce RTX 4090 Founders Edition Graphics Card 24GB GDDR6X – Titanium and Black: Buy it now

Game Dev/Virtual Production: Unreal Engine

Switching to the Unreal Engine, things are a bit more nuanced. While some niche use cases can take advantage of multiple GPUs, this benchmark focuses on single GPU performance – specifically for timing graphics. This score looks at three different Unreal Engine scenes, with and without ray tracing, and at a variety of resolutions. For this test, the team of Puget systems performance testers did not use DLSS. Currently, only the 4090 supports DLSS 3.0, while the 3000 series graphics cards use DLSS 2.0, so it’s hard to get an apple vs apple comparison. Instead, puget systems are looking at raw horsepower. Furthermore, not all professionals will use DLSS in their daily work as it has an impact on the final pixels each user will need to check to see if it is acceptable for their project or not.

Overall, the RTX 4090 showed a frame rate increase of about 78% on all scenes compared to the 3090 Ti and 93% when compared to the 3090. For those working at a fixed 30-60 FPS for virtual production, This means you can do a lot more in your scene at the same frame rate. If you’re taking that advantage on a 3000 series GPU, you’ll have plenty of headroom with the RTX 4090, allowing you to push realism even further. Many in the virtual production space will be waiting for the new RTX 6000 Ada to gain access to Quadro Sync and 48GB of VRAM, but based on previous generations, the 4090 should give us a really solid look at what to expect.

An important note about this generation of video cards (both the RTX 40 Series and the Quadro “Ada Lovelace” cards) is that Nvidia has removed NVLink. It’s now required if you want to use multiple GPUs for tasks like GPU lightmass, internal/external failure in nDisplay, and multi-GPU usage for path trackers in Unreal 5.1. Technically, nDisplay doesn’t claim NVLink, but the performance hit when not using it is so great that every one the Puget Systems team spoke to, including the folks at Epic, said it did not worth a try.

There are some potential fixes, such as Epic recently adding SMPTE 2110 in 5.1, improving PCIe handling (though that may not appear until PCIe 5.0), or some other solution. Not many people will use GeForce graphics cards for this because you lose the Quadro Sync option, so there’s still time to find a possible solution. Worst case scenario, the RTX 4090 has the same performance as two RTX 3090s, but without the hassle of getting the dual GPUs to work.

GPU Rendering: OctaneRender

Switching to Octane from Otoy, we see that the new Nvidia GeForce RTX 4090 once again nearly doubles the speed of the previous generation’s RTX 3090 and RTX 3090 Ti, being 92% and 83% faster, respectively. It’s practically on par with the dual RTX 3090s. If you opt for a dual RTX 4090 system, the RTX 4090 balances out almost perfectly, allowing for truly impressive render times.

Otoy has made some huge improvements recently and these results are mainly for the standard rendering system. You’ll see similar performance improvements to their real-time preview, or when using Octane internally and DCC, Unreal Engine, or plugins like EmberGen. Additionally, Otoy is working on improving out-of-core performance and other ways to make large scenes work better on the GPU if you don’t have enough VRAM. This will help minimize the lack of NVLink and the inability to pool VRAM.

GPU Rendering: Redshift

Next is the GPU rendering test of puget systems Redshift from Maxon. Here, puget systems see not a huge improvement, with the RTX 4090 being only about 60% faster than the RTX 3090. To a certain extent, this may be a limitation of the Redshift benchmark. Both Octane and V-Ray look at the amount of work that can be completed in a given amount of time, while Redshift renders a single frame and reports the time taken. This has been a good approach for a long time, but modern GPUs are evolving so quickly that the cost of rendering, in the beginning, will matter to the end. The performance test team went from 251 seconds for the RTX 2080 Ti, to 145 seconds for the RTX 3090, to just 87 seconds for the RTX 4090. Amazingly, the dual RTX 4090s finished rendering in 45 seconds. Again, there’s no NVLink bridge, so you’ll be limited to the VRAM of a single graphics card.

We must also mention that while this is a result of Redshift, Maxon has also started adding GPU-based physics simulations to Cinema 4D. These new features are too new for puget systems to test before the RTX 4090 launches, but the puget systems performance team hopes to have some concrete data soon. However, we fully expect to see similar results, with the RTX 4090 performing 60-90% faster than the RTX 3090.

In conclusion, the RTX 4090 is still a very powerful graphics card, and it has been continuously improved and upgraded by NVIDIA in terms of performance, however, with a price of more than $2000, this will indeed still be the case. is the biggest barrier that makes technology users shy and hesitant to upgrade from RTX 3080, RTX 3090 to RTX 4090, the performance that RTX 4090 brings is still really good and this makes it still worth it. Some people are willing to spend a large amount of money to upgrade to ensure performance and work efficiency.