Nvidia’s next-generation graphics cards have finally been officially unveiled, and if Nvidia’s own claims are anything to go by, they’re setting new standards for power and performance. At the top of the heap is the new RTX 3090, a monster card that perhaps looks more like a Titan than a Ti. But since it’s the most powerful graphics card Nvidia has ever made, it’s worth comparing it to the latest generation flagship card.
When it comes to the best of the best, the 3090 and 2080 are two (not) titans in the graphics card game, but how different are they really? Is it still worth buying a 2080 Ti if you can find it at a good price?
Prices and availability
The RTX 2080 Ti originally launched on September 28, 2018, priced at $1,199 for the Nvidia Founders Edition. The “Standard Edition,” which served as the basis for third-party cards, had a base price of $999, though Nvidia’s board partners typically added better cooling, higher clock speeds, and other features that pushed the price up. at $1,900. in extreme cases.
Although technically still available, the 2080 Ti is no longer being made and supplies are very limited. There are still some more expensive options available, and the used market is likely to be extremely busy as RTX 3000 graphics cards become available, but finding a new 2080 Ti isn’t easy.
The RTX 3090 launched on September 1, 2020, but is not yet available for sale. It will hit store shelves on September 24, priced at $1,499. This is significantly more expensive than the original price of the 2080 Ti, making it an even more niche card. Still, it’s much more capable than the Titan RTX, which debuted at $2,500 just a few years ago, so if you want the best of the best, it might be a price worth paying when it goes on sale.
Performance
Nvidia RTX 2080 and 2080Ti
Nvidia posted some clear performance numbers for its RTX 3090 during its September press event, giving us a first glimpse of what the next-gen card can do. We also have some real-world performance results for Digital Foundry’s weaker 3080 that hint at the full power of the even more capable 3090. We’ll have to wait for third-party reviews to be sure of the new flagship’s capabilities compared to the RTX 2080 Ti, but Nvidia’s own numbers and the raw specs of the two cards give us an idea of what to expect.
| RTX 3090 | RTX 2080Ti | |
| GPUs | GA102-300 | TU102-300-K1-A1 |
| Interface | PCI Express 4.0 | PCI Express 3.0 |
| CUDA cores | 10,496 | 4,352 |
| tensor nuclei | Determined | 544 |
| RT core | Determined | 68 |
| basic class | 1.395MHz | 1350MHz |
| turn up the clock | 1.695MHz | 1.545MHz |
| Memory | 24GB GDDR6X | 11GB GDDR6 at 14Gbps |
| memory speed | 19.5Gbps | 14 GPS |
| memory interface | 364 bit | 352 bit |
| Broadband | 936GB/s | 616GB/s |
| TDP | 350W | 260W |
The RTX 3090 has an incredible number of CUDA cores – over 10,000. On paper, that’s nearly 2,000 more than the RTX 3080 and more than double the RTX 2080 Ti. This helps it produce a total of 36 TFLOPs in shader performance, which is almost a triple increase over the 2080 Ti’s 14.2 TFLOPs. While it sounds like it should mean the 3090 is more than twice as fast as the 2080 Ti, it’s not that simple.
All of the CUDA cores on the RTX 2080 Ti are designed to handle floating point math (FP32), with an additional 4352 cores (unspecified) designed to handle integer math (INT32). Games use them both in different ways, but each can only do one job. By comparison, the RTX 3090 has 5,238 FP32 CUDA cores and 5,238 cores that can run either FP32 or INT32. This should mean that all of its cores are used at the same time, but if those cores that can handle INT32 are used for INT32 in the same way as with the 2080 Ti, then the 3090’s performance boost increases the number of cores. it could be more than 20%, instead of the 140% you might otherwise suggest.
That being said, there are other aspects of the 3090 that mean it will be significantly faster than the 2080 Ti. Their cores run at a faster base and increase the clock. It also has more than double the amount of memory with 24GB of next-generation GDDR6X, along with a wider memory bus that leads to nearly 1GBps in memory bandwidth—more than 50 percent increase with compared to the 2080 Ti.
Nvidia claims the 3070 is more capable than the 2080 Ti, so it’s no surprise that Digital Foundry has tested the RTX 3080 in a number of games (certainly handpicked by Nvidia), seeing regular results of 60% to 90% performance improvements. on the RTX 2080 — in 4K with ray tracing enabled and disabled. That would mean it could be up to 40% faster than the 2080 Ti. Based on the number of CUDA cores and similar clock speeds, the RTX 3090 could be 70-80% faster than the 2080 Ti, even faster in real-world games.
We’ll have to wait for more detailed tests of the 3090 in a wider range of games at different detail settings, but it looks like the 3090 will be significantly more capable than even the best of the last generation, especially in scenarios where it can take advantage of its variable CUDA cores to plus FP32 operations per second.
While the 2080 Ti is designed as a 4K graphics card, Nvidia did show off a demo where the 3090 has the power and performance to handle games at resolutions up to 8K at 60 FPS with ray tracing enabled. DLSS probably played a big part in making this possible, and we’ll have to see this replicated in third-party sources before we take it for gospel, but it’s an impressive demo nonetheless.
Gamers React to 8K Ray Tracing | GeForce RTX 3090 [Presented in 8K]
If it’s anything to go by, 100 FPS at 4K seems quite possible with the 3090.
power supply and cooling
The 2080 Ti launched with what was at the time a revolutionary cooling design for Nvidia’s Founders Editions, introducing a dual-fan setup for the first time. Third-party alternatives tended to be varied, with two- and three-fan variants, as well as some water-cooling and hybrid-cooling solutions.
Like most cards of its generation, the 2080 Ti used dual 8-pin power connectors to handle its base TDP of 260 W. A heavily overclocked 2080 Tis could draw over 500 W through three 8-pin connectors. , although these were rare, even and for such a high class card.
The RTX 3090 uses Nvidia’s new push-pull vapor chamber design and reconfigured PCB in the concave “V” shape it introduced with the RTX 3080; it is much bigger. The fins are bigger, the fans have longer blades, and the entire card is bigger. It measures 5.4 inches wide and 12.3 inches long (one inch wider and longer than the 3080) and therefore could cause clearance issues in smaller cases.
The large cooler also makes for a three-slot design, though Nvidia promises this new configuration will help the card perform better at lower noise levels than previous-generation Titan GPUs, despite higher power requirements. .
Powercolor and Palit RTX 3090s
Third-party card alternatives have already been announced and the new PCB shape doesn’t seem to have disrupted the typically long and thick three-fan designs, nor water-cooled alternatives. However, most seem to have a dual eight-pin power connector design, rather than a 12-pin one. This may be limited to Nvidia Founders editions.
Lightning tracing, DLSS and more
As with the launch of the RTX 2000 series graphics cards, Nvidia spent much of its press event for the new RTX 3000 series talking about ray tracing performance. Second-gen RT cores are said to deliver up to 1.7x Turing performance, allowing the next-gen RTX 3080 to run at 4K at 60 FPS with ray tracing enabled. The 3090 will be even better, offering 8K ray tracing using DLSS technology.
The RTX 2080 Ti was unable to play 4K games with ray tracing enabled without extremely low frame rates, even with DLSS enabled. The 3090 seems to have solved that problem.
Some of that progress is due to a general improvement in the performance of the 3090’s GPU core, so until we have third-party testing to confirm all aspects of the new card’s performance, we can’t say for sure what capacity has. will be the traceability. However, it’s much more capable than the 2080 Ti, making it a much better card for those who want to game at ultra-high resolutions with ray tracing enabled.
For a full list of games that support ray tracing, check out our guide.
Second-generation DLSS is a major component in Nvidia’s ray-tracing and 4K gaming program, making the 3090 much more capable than it otherwise would be. It’s come a long way since the first generation, and while it’s not perfect, it should help give Nvidia a performance advantage at the top of the market when it competes with AMD’s upcoming Big Navi and RDNA 2 alternatives.
The RTX 3090 also uses faster second-generation tensor cores, allowing it to handle even more detailed sharpening and denoising tasks in real time, allowing for more detailed gaming, with less performance loss than we saw with the 2080 You.
The king is dead, long live the king
The best of the most incredible of the last generation is not so unique anymore, thanks to the 3080 and 3070 offering performance that exceeds what the 2080 Ti was capable of. Unlike that generation, the 3090 feels like a significant upgrade. We can’t provide you with the proper statistics at this time, but we can assure you that preliminary analysis offers promising prospects for unrivaled performance ratings that far exceed products from AMD and Nvidia.
The 3090 has an excellent HDMI 2.1 compatible power connector, so you can display your game consoles in unsurpassed quality. Its innovative internal fan technology has an efficient and quiet design that will protect your console from excessive heat without muting your sound. It’s even priced significantly lower than a typical (Titan-Esque) flagship card, even if it’s out of most gamers’ budget.
Once you’re in the market, you’ll want to get the RTX 3090, so you can have the fastest 4K and 8K gaming, even with ray tracing enabled. While we can’t argue with the quality of the 2080 Ti card, it no longer stands out as we enter the new year.
editor’s recommendations
Categories: GAMING
Source: newstars.edu.vn