The gaming video card market today is on the verge of big changes. NVIDIA is preparing to release consumer versions of Ampere silicon, and AMD will soon break into the upper price segment, still occupied by the “green”, with accelerators on the large Navi chip. In addition, we must not forget that the next generation of game consoles is coming - PlayStation 5 and Xbox Series X, and these will be the first consoles that will receive hardware-accelerated ray tracing functions, and in general they will be much stronger than their predecessors. All this means that not only flagship offerings, but also video cards of the mid- and mid-high price echelons will see a major increase in performance. Unless AMD will not disturb the existing Radeon RX 5000 lineup, which, with the exception of the very top, is already fully equipped (although some intermediate upgrade may occur, following the example of the Radeon RX 500 family).
Of course, the hopes that AMD will return the golden days, when the GeForce and Radeon brands competed on equal terms in the entire performance range, and the gaming FPS was rapidly becoming cheaper, have turned into a complete disappointment more than once. But now, it seems, the “reds” have every chance to push, if not the latest accelerators on Ampere chips, then at least the GeForce RTX 2080 Ti. And most importantly, this is not so important anymore: since the prices for top models have risen to $700 and more, for most gamers sitting at screens with a resolution of 1920 × 1080, such video cards are only of theoretical interest. Another thing is accelerators one step lower, which have recently occupied a niche from $400 to $500. It was to them that all the attention was riveted last year, when the Radeon RX 5700 XT appeared, and NVIDIA in response was forced to almost completely redraw the GeForce RTX 20 series. These models, and before that their predecessors, have always enjoyed well-deserved popularity, because they are sold for quite feasible sums, and a serious performance margin even at a relatively low resolution is now more than ever in demand by new resource-intensive games like, for example, .
It is these devices that manufacturers combine with the term Performance (as opposed to the flagship Enthusiast) that we will deal with in the second part of the retrospective review (if anyone missed it, here ). In it, we intend to cover the most striking models presented in the eight years since NVIDIA introduced Kepler logic and AMD introduced the GCN architecture. We will again omit earlier devices of the GeForce 500 and Radeon HD 6000 series due to the severe shortage of RAM in most of them.
When choosing test participants, we had to be guided by several criteria. First of all, the position of the device in the NVIDIA product line. It is NVIDIA, because the "green" numbers of all the models we are interested in end with 70, and among the "red" analogues, the range of which is constantly changing, we put forward devices that are close in speed and price. Another sign that all video cards of the test pool have in common is that almost all of them were based on second-tier chips of their time: Gx-104/204 from NVIDIA or Tahiti, and then Hawaii/Grenada from AMD. Even the Radeon RX Vega 56 and Radeon RX 5700 XT don't stand out from the crowd, as the Vega family has the flagship Radeon VII product, and the Navi line will soon receive a natural continuation as well. The only exception was the GeForce RTX 2070, which NVIDIA spared the TU104 chip, although the GeForce RTX 2070 SUPER is already based on it.
The price corridor of all listed devices lies within $329-500 (the only exception on the chart is the GeForce RTX 2070 in the Founders Edition modification, which NVIDIA priced at $100 more than the recommended amount), although you can see that such video cards were the cheapest between 2013 and 2016 , when prices were pressed by heightened competition between NVIDIA and AMD. Since then, even the "red" accelerators, which are traditionally considered the choice of economical gamers, have steadily risen in price. So let's find out whether the increase in prices is justified by the corresponding increase in performance, or, conversely, as we have already stated for the flagship models, new devices give more FPS, but each frame per second is now paid at an increased rate.
How We Tested
Before we start analyzing the test results, it's worth explaining once again why we chose as benchmarks exactly those games whose names you will see on the diagrams, and not any others. With the flagship models behind us this time around, the issue of scaling performance across devices that are seven years of rapid progress (like the GeForce GTX 680 and GeForce RTX 2080 Ti) is no longer so acute. Nevertheless, all the obstacles that initially stood in the way of comparative testing remained in place.
The first difficulty is related to the extremely limited amount of memory on board old video cards. So, the standard version of the GeForce GTX 770 participating in the second review series has only two gigabytes of VRAM, while the Radeon HD 7950 and Radeon R9 280X have three. In the comments to the last article, readers noticed that some older models have varieties with twice the amount of memory, but we are bound by the capabilities of the reference devices, which make up the lion's share of the test fund. At the same time, any modern game eats up at least 4 GB, but not always its appetites can be tempered with reduced detail settings. For the same reason, we had to limit all tests to 1920 × 1080 screen mode, because resolution is always positively related to VRAM consumption: the larger the picture, the more memory it requires.
The next obstacle was the ability of the game engine to unleash the potential of modern accelerators, increasing the frame rate beyond a hundred or even two hundred FPS. This is exactly what is required in a situation where older devices start from low positions, and we have reduced the load on the GPU in order to keep within 2-3 GB of VRAM. But fortunately, among the games that we constantly use for GPU tests, several projects - Battlefield V, Borderlands 3 DiRT Rally 2.0, Far Cry 5 and Strange Brigade - have the right properties. However, we do not guarantee that the current versions of NVIDIA and AMD drivers, as well as the games themselves, are well optimized for legacy silicon. To compensate for this factor, we added several old games from 2011-2013 to the benchmarks collection - Crysis 2, Metro Last Light and Tomb Raider, and in order to ensure the correct frame rate scaling, we had to, on the contrary, increase the graphic parameters to the maximum and enable resource-intensive full-screen anti-aliasing.
| Games | |||
|---|---|---|---|
| Game (in order of release date) | API | Settings, test method | Full screen anti-aliasing |
| Crysis 2 | Direct3D 11 | Adrenaline Crysis 2 Benchmark Tool. Max. graphics quality, HD textures | MSAA 4x + Edge AA |
| tomb raider | Direct3D 11 | Built-in benchmark. Max. graphics quality | SSAA 4x |
| Metro Last Light | Direct3D 11 | Built-in benchmark. Max. graphics quality | SSAA 4x |
| Far Cry 5 | Direct3D 11 | Built-in benchmark. Low graphics quality | Off |
| Strange Brigade | Direct 3D 12/Vulkan | Built-in benchmark. Low graphics quality | AA Low |
| Battlefield V | Direct3D 11/12 | OCAT, Liberte mission. Low quality graphics. DXR off, DLSS off | TAA High |
| DiRT Rally 2.0 | Direct3D 11 | Built-in benchmark. Average graphics quality | MSAA 4x + TAA |
| Borderlands 3 | Direct3D 11/12 | Built-in benchmark. Very low graphics quality | Off |
Despite all the efforts to select games and optimize settings, in the previous, flagship part of the review, we could not avoid scaling artifacts in the final segment of the timeline - from the GeForce GTX 1080 Ti and Radeon VII to the GeForce RTX 2080 Ti. As a result, we had to exclude a large part of the data from the general performance and FPS cost graphs. For devices of the next price category, which we will focus on today, this problem is not so acute, and the results of most test games, and under different APIs (Direct3D 11, Direct3D 12 and Vulkan), will be taken into account in the conclusion of the review.
Performance testing in Crysis 2 was performed using the timedemo function and the Adrenaline Crysis 2 Benchmark Tool. DiRT Rally 2.0, Far Cry 5, Metro Last Light, and Strange Brigade used the built-in benchmark for testing and collecting results, while Borderlands 3 and Tomb Raider used the built-in benchmark in combination with the OCAT program. Battlefield V required manual testing using OCAT over a repeated portion of the Liberté mission.
| Test stand | |
|---|---|
| CPU | Intel Core i9-9900K (4,9GHz, 4,8GHz in AVX, fixed frequency) |
| Motherboard | ASUS MAXIMUS XI APEX |
| RAM | G.Skill Trident Z RGB F4-3200C14D-16GTZR, 2×8 GB (3200 MHz, CL14) |
| ROM | Intel SSD 760p, 1024 GB |
| Power supply unit | Corsair AX1200i 1200W |
| CPU cooling system | Corsair Hydro Series H115i |
| Chassis | CoolerMaster Test Bench V1.0 |
| Monitor | NEC EA244UHD |
| Operating system | Windows 10 Pro x64 |
| AMD GPU software | |
| All video cards | AMD Radeon Software Adrenaline 2020 Edition 20.4.2 |
| NVIDIA GPU software | |
| All video cards | NVIDIA GeForce Game Ready Driver 445.87 |
Test participants
- ;
- ;
- ;
- ;
- ;
- ;
- ;
- ;
- ;
- ;
- ;
- .
Note. In brackets after the names of the video cards, the base and boost frequencies are indicated according to the specifications of each device. Video cards of non-reference design are brought into line with the reference parameters (or close to the latter), provided that this can be done without manually adjusting the clock frequency curve. Otherwise (GeForce RTX Founders Edition accelerators), manufacturer settings are used.
Test results (old games)
Crysis 2
The graph with the test results in the first game shows how much easier it is to compare the performance of devices over time that belong to the same category (albeit a fairly broad one in this case) based on their price and position in the manufacturer’s product line. Over the past eight years, the capabilities of accelerators for enthusiast gamers have grown at a vigorous, almost linear pace, and Crysis 2, despite its venerable age, does not hold back performance scaling from the starting positions of the GeForce GTX 670 and Radeon HD 7950 up to the GeForce RTX 2070 SUPER and Radeon RX 5700 XT.
But any conclusions about historical trends will have to be made with great caution - we are no longer talking about the flagship products of NVIDIA and AMD, which reflect the best achievements of companies. This time we chose for review the models that are closest in terms of overall performance in each time period, but the advantage of a particular device in terms of frame rate does not mean that it is definitely better than its direct competitor, for the reason that the difference in performance during in many cases, it was included in the price of video cards. This question is best answered by the graph of the average cost of FPS, which we will provide at the end of the article.
However, certain expectations are associated with the model numbers of devices in the nomenclature of NVIDIA and AMD. In particular, that is why the composition of the test participants is exactly this, and not any other. If we focus on a narrow class of product, as manufacturers understand it, then in Crysis 2 AMD's high point was the Radeon R9 390 (an extremely popular - and for good reason - model of 2015). Up to this point, the game, due to obvious sympathy for the Kepler architecture compared to the first generation of GCN, works better on “green” hardware, and after that it is impossible to hide that AMD, as in the case of the flagship models that we studied in the last part of the study , has faced purely technical obstacles that prevent it from playing on an equal footing with NVIDIA.
Metro Last Light
Metro Last Light is a fairly heavy game by modern standards, and even more so with “fair” full-screen anti-aliasing SSAA 4x. It is not surprising that in this test NVIDIA products did not go beyond 125, and AMD - 100 FPS. Here we see that clashes between the two chipmakers over the course of eight years often ended in conditional parity (especially when adjusted for the price of the devices). Indeed, Metro Last Light equates the Radeon R9 390 and the GeForce GTX 970, and then between the Radeon RX Vega 56 and the GeForce GTX 1070, and has narrowed the gap between the GeForce GTX 770 and the Radeon R9 280X.
tomb raider
The first game in the relaunched Tomb Raider series from 2013 was the only one among the three old projects we selected that showed AMD devices in the most favorable light. The first video cards based on GCN architecture chips work more efficiently in it than the “green” Kepler chips, and even the enormous overclocking of the GeForce GTX 680 that NVIDIA performed in order to get the GTX 770 did not allow it to snatch the championship from the Radeon R9 280X at that time. The GeForce GTX 970 and Radeon R9 390 are, by and large, equivalent here, as are their rivals in the next pair - GeForce GTX 1070 and Radeon RX Vega 56. Finally, the Radeon RX 5700 XT is not much inferior to the original, not SUPER, version of the GeForce RTX 2070.
Test results (new games)
Battlefield V
Battlefield V gave us a lot of trouble in the first part of the GPU retrospective because its graphics engine behaves so differently in Direct3D 11 and Direct3D 12 environments, especially at the high frame rates that flagship devices develop. Nevertheless, we did not discard this test and, as the results showed, we did the right thing. In the performance range we're focusing on today, Battlefield V does not hinder FPS scaling when running both versions of the Microsoft graphics API, while still reflecting the significant difference between Direct3D 11 and Direct3D 12.
Contrary to popular belief, the transition to Direct3D 12 does not in all cases have a beneficial effect on the performance of AMD accelerators. Last time we noticed that the Radeon HD 7970 GHz Edition in Battlefield V was faster under Direct3D 11, and now the same thing happened with two related models - the Radeon HD 7950 and the Radeon R9 280X. All other testers benefit to some extent from the migration to the progressive API, and this is clearly seen in the different slopes of the curves in the diagrams.
As a result, early AMD graphics cards (Radeon HD 7950 and Radeon R9 280X) and NVIDIA (GeForce GTX 670 and GeForce GTX 770) are swapped depending on the current API, and the GeForce GTX 970 is being pulled up to the Radeon R9 390 thanks to Direct3D 12. How do we It has already been noted more than once that the latter has the best effect on the results of large AMD chips. Under Direct3D 11 conditions, the Radeon RX Vega 56 and GeForce GTX 1070 Ti, on the one hand, and the Radeon RX 5700 XT and GeForce RTX 2070, on the other, showed almost the same result. Thanks to Direct3D 12, these graphics cards are clearly faster.
In general, we can say that in Battlefield V, the “red” accelerators hold up perfectly for an eight-year time period, and if we adjust for the prices of rivals, then in total it is AMD that wins.
Borderlands 3
Borderlands 3 is another illustration that Direct3D 12 doesn't always benefit GPU performance. In this game, only the older models of NVIDIA (GeForce RTX 2070 and RTX 2070 SUPER) and AMD (Radeon RX Vega 56 and Radeon RX 5700 XT) accelerated thanks to the modern API. The change in the software layer did not affect the Radeon R9 290 in any way, and relatively low-power video cards only lost FPS.
Nevertheless, in all the test results of Borderlands 3, it is worth focusing on Direct3D 12, since Direct3D 11 from a certain point simply does not allow scaling performance in accordance with the processing power of the GPU. The new API here almost always plays in favor of AMD. Due to it, the Radeon R9 280X is approaching the GeForce GTX 770, the next two models (Radeon R9 290 and Radeon RX Vega 56) are ahead of all their rivals (GeForce GTX 970 and GeForce GTX 1070, GTX 1070 Ti, respectively) and even the Radeon RX 5700 XT caught up with a formally stronger GeForce RTX 2070 SUPER graphics card.
DiRT Rally 2.0
Among the games that we use now or have ever used in the past to compare video cards, there are not many that, in principle, can demonstrate the full range of performance between today's powerful video cards and their eight-year-old predecessors. DiRT 2.0 is one such project, but it has a specific problem that will not allow the results of this benchmark to be included in the final graphs and tables. For some reason, AMD's Hawaii-based accelerators (Radeon R9 290/390 models) are slower here than the Radeon R9 7950/7970 and Radeon R9 280/280 X.
Otherwise, DiRT 2.0 ranked old and modern video cards from two manufacturers according to their average performance, which we installed at the time and will reassure you in the final section of the retrospective review. Here, AMD's early GCN architecture devices, the Radeon R9 7950 and Radeon R9 280, outperform their GeForce GTX 670 and GeForce GTX 770 rivals in terms of frame rates, while the Radeon RX Vega 56 falls between the GeForce GTX 1070 and GeForce GTX 1070 Ti. Finally, the Radeon RX 5700 XT has a slight advantage over the GeForce RTX 2070.
Far Cry 5
The results of all video card benchmarks in Far Cry 5 also look pretty typical, but again with the exception of the Radeon R9 390 - the difference between the latter and the Radeon R9 280X is too small. However, in this case this is not explained by the frame rate deficit of the Radeon R9 390 (it is on par with the GeForce GTX 970), but by the unexpectedly high results of accelerators on Tahiti chips - Radeon HD 7950 and Radeon R9 280X. More recent models are in their usual places: the Radeon RX Vega 56 sits next to the GeForce GTX 1070 Ti, and the Radeon RX 5700 XT sits next to the GeForce RTX 2070.
Strange Brigade
Strange Brigade is a rare game that gives you the choice not between two versions of the Microsoft API, but between Direct3D 12 and Vulkan. The latter generally provides higher performance, but not always for the video cards from which it is customary to expect. Vulkan in Strange Brigade favors older AMD models (Radeon HD 7950 and Radeon R9 280X) and NVIDIA accelerators starting with GeForce GTX 1070. For more powerful AMD devices (Radeon R9 390, Radeon RX Vega 56 and Radeon RX 5700 XT) along with GeForce GTX 970 is useless, while the GeForce GTX 670 and GeForce GTX 770 only hurt.
Strange Brigade, true to its reputation, is more of a "red" than a "green" project. Three early AMD models (Radeon HD 7950, Radeon R9 280X and Radeon R9 390) outperform their closest rivals (GeForce GTX 670, GeForce GTX 770 and GeForce GTX 970) in FPS, especially under Vulkan. But the Radeon RX Vega 56 and Radeon RX 5700 XT perform better in Direct3D 12. The former is in any case ahead of the GeForce GTX 1070 Ti, but under Direct3D 12 the difference is greater. In turn, the Radeon RX 5700 XT under Vulkan is inferior to the GeForce RTX 2070, but thanks to Direct3D 12 it is able to catch up.
Conclusions
Just like , dedicated to top-end video cards from AMD and NVIDIA, we placed the benchmark results of several games on a summary chart and drew average frame rate lines through the points of individual devices. But this time we managed to avoid the performance scaling artifacts that plagued flagship testing in most games. All projects were included in the final calculations, and under different APIs, with the exception of DiRT 2.0 and Far Cry 5, in which there is no expected distance between AMD accelerators on Tahiti and Hawaii chips, and Borderlands 3 in Direct3D 11 mode, where performance growth is limited after Radeon RX Vega 56 and GeForce GTX 1070.
Looking at the graph, we realized that we had made no mistake either in the selection of video cards for comparison or in the list of test games. The products of each of the two manufacturers lined up, and the rival models took quite predictable positions. All this means that, even if the performance of flagship solutions stalls over time - at least in the most popular resolution of 1920 × 1080 - you can rest easy for accelerators one step lower at a price in the range of $400–500. In addition, there is no such gap between “red” and “green” devices as in the highest category. Here, NVIDIA only took the lead in the last two years with the birth of the GeForce RTX 2070 and GeForce RTX 2070 SUPER, but this is completely logical if you take into account the high starting prices of both models.
By the way, about prices. Unlike top-end video cards, more affordable accelerators have demonstrated a steady decline in the specific cost of gaming performance. On the “red” side, FPS fell in price by 4,26 times taking into account inflation, and on the “green” side by 3,66. Only the GeForce RTX 1070 Ti and GeForce RTX 2070, which in our test is represented by the expensive Founders Edition modification, broke away from the general downward trajectory. The GeForce RTX 2070 SUPER, which appeared on the market under pressure from the Radeon RX 5700 XT, returned NVIDIA products to their previous course. The two competing models offer FPS for similar amounts - $1,9 for the Radeon RX 5700 XT and $2,1 for the GeForce RTX 2070 SUPER, but AMD's slight advantage in this case is completely balanced by hardware-accelerated ray tracing on NVIDIA chips. The sad thing is that after the GeForce 10 series, gaming video cards do not slow down the pace of performance growth, but the changes in the price of “green”, and with them “red” FPS, frankly speaking, are not noticeable. It looks like the chipmakers (or one of them, as caustic commentators are sure to correct) intend to accustom the public to the idea that it’s time to wean off the “free” speed increase every two years. If you want to still play without brakes when the old hardware has outlived its useful life, please pay the same amount. The only hope again is that someday Ryzen will appear among gaming video cards.
In two series of historical testing, we have already covered a total of 23 devices introduced between 2012 and 2019. There are models that belong to, perhaps, the most demanded middle price category, whose names in NVIDIA's nomenclature end in 60 (and, of course, their "red" counterparts). We intend to deal with them next time and sum up the general results of the entire study - do not miss it.
| Release date | Average frame rate, FPS | Recommended price at the time of issue, $ (without tax) | Recommended price adjusted for inflation, $2012. | $/FPS | $'2012/FPS | |
|---|---|---|---|---|---|---|
| AMD Radeon HD 7950 | January 2012 of | 56 | 450 | 450 | 8,1 | 8,1 |
| AMD Radeon R9 280X | August 2013 | 67 | 299 | 295 | 4,5 | 4,4 |
| AMD Radeon R9 390 | 2015 of June | 107 | 329 | 319 | 3,1 | 3 |
| AMD Radeon RX Vega 56 | August 2017 | 155 | 399 | 374 | 2,6 | 2,4 |
| AMD Radeon RX 5700XT | July 2019 city | 192 | 399 | 358 | 2,1 | 1,9 |
| Release date | Average frame rate, FPS | Recommended price at the time of issue, $ (without tax) | Recommended price adjusted for inflation, $2012. | $/FPS | $'2012/FPS | |
|---|---|---|---|---|---|---|
| NVIDIA GeForce GTX 670 | May 2012 | 52 | 400 | 400 | 7,7 | 7,7 |
| NVIDIA GeForce GTX 770 | May 2013 | 64 | 399 | 393 | 6,2 | 6,1 |
| NVIDIA GeForce GTX 970 | September 2014 of | 92 | 329 | 319 | 3,6 | 3,5 |
| NVIDIA GeForce GTX 1070 | 2016 of June | 143 | 379 | 363 | 2,7 | 2,5 |
| NVIDIA GeForce GTX 1070 Ti | November 2017 | 157 | 449 | 421 | 2,9 | 2,7 |
| NVIDIA GeForce RTX 2070FE | 2018 October, the | 190 | 599 | 548 | 3,1 | 2,9 |
| NVIDIA GeForce RTX 2070 SUPER | July 2019 city | 209 | 499 | 448 | 2,4 | 2,1 |
Source: 3dnews.ru