Hard times for NVIDIA video game companies with AMD in the back

The ups and downs of NVIDIA's price

The NVIDIA (NVDA) stock, like all other semiconductor stocks, has had its ups and downs since October 2018 due to a variety of business, industry and investment factors. the weakness of the macroeconomic environment. In October 2018, the demand for cryptography disappeared, creating a glut of GPUs (graphic processing units) of the occasion in the market. At the same time, the United States imposed tariffs on Chinese exports worth $ 200 billion. In a low-demand environment, NVIDIA launched the first real-time gaming graphics processor based on its Turing architecture. GPUs failed to create spark because of their high prices, lack of an ecosystem supported by ray tracing and improvement marginal performance of existing AAA games.

These headwinds drove the NVIDIA stock down to 55 percent in the fourth quarter. The company experienced some growth in March and April, followed by a decline when the US-China trade war resumed in May, as higher rates impacted NVIDIA's earnings guidance for the US. the rest of the year. Signs of slackening trade war somewhat relieved the stock in July.

NVIDIA is desperate to increase its gaming revenue

NVIDIA is a leader in the discrete GPU market and earns more than 50% of its revenue from gambling. However, its gaming revenue has fallen to a high double-digit rate over the past two quarters. Its gaming revenue for the first quarter of fiscal year 2020 dropped 39% year-over-year, while the chain was gearing toward the elimination of excess GPU inventory after the burst of the cryptographic bubble. However, its gaming revenue grew 10.6% sequentially thanks to sales of its GTX 1600 graphics processor supported by ray tracing, starting at $ 149.

NVIDIA's GPU market share increased from 81.2% in the fourth quarter of 2018 to 77.3% in the first quarter of 2019, reflecting the two consecutive quarters of declining gaming revenue, according to data from Jon Peddie Research. NVIDIA lost this market share to Advanced Micro Devices (AMD).

Declining gaming revenue and loss of market share led NVIDIA to strengthen its gaming business and maintain its lead. NVIDIA's despair was visible in its Turing GPU lineup, which was a little complicated, with too many GPUs of game launched in too short a time to cover each price range.

NVIDIA GPU Turing Portfolio

The NVIDIA GPU Turing range has ten GPUs, compared to four in the Pascal range. NVIDIA began with the launch of Turing with standard versions of the RTX 2080 Ti, 2080, 2070 and 2060. NVIDIA has also followed this range of traditional GeForce GPUs in the Maxwell and Pascal generations.

Turing's GPUs dramatically improved ray tracing performance, while delivering performance similar to that of its predecessor Pascal in existing games that did not support real-time ray tracing. The lack of games supported by ray tracing and the high price of RTX GPUs have discouraged players from upgrading.

To drive the adoption of its Turing GPUs, NVIDIA has launched the Turing-based GTX 1660 Ti, 1660 and 1650 models, with no ray tracing function, in the price range of 149 to 279 USD. These GPUs offer up to 50% performance improvement over their Pascal-based predecessors. These are the GPUs that increased NVIDIA's gaming revenue in the first quarter.

In July, AMD launched its Navi-based Navi-based GPUs, which it says exceed the performance of the NVIDIA RTX 2070 game (with the exception of raytracing). To compete with AMD's new range of GPUs in terms of price and performance, NVIDIA is launching an updated version of the super GPU lineup at the same price as its standard version.

NVIDIA Super GPU Series

NVIDIA unveiled on July 2 the Super version of its RTX 2060, 2070 and 2080 GPUs, a few days before the launch of AMD's Navi GPUs. NVIDIA has improved the product specification in its updated series by adding more cores, increasing clock speed and improving Tera's floating point performance per second. The Super Series will offer an average speed gain of 15% over standard versions, maintaining the performance advantage of the NVIDIA GPU over AMD's GPU.

According to AnandTech, the specifications of the Super 2070 and 2060 cards show that NVIDIA is hitting its margins by offering these specifications at the same price as its standard versions. For example, NVIDIA switched to a larger chip for the RTX 2070 Super so that additional transistors could be inserted at the same price and increased the RTX 2060 Super's 8GB random access video memory compared to the RTX 2060's 6GB for only $ 50 more. .

Although NVIDIA has improved the specifications of its RTX Super GPUs, it has compromised its advantage in terms of TDP (thermal design power) advantage. Until now, NVIDIA's GPUs had a significant TDP advantage over AMD GPUs. With the GPU Super RTX lineup, NVIDIA has increased its TDP compared to standard versions, reducing its energy efficiency and narrowing the gap between NVIDIA and AMD GPUs.

NVIDIA will launch its Super RTX 2060 and 2070 GPUs on July 9, two days after the launch of AMD's RX 5700 Series GPUs. NVIDIA said the RTX 2070 Super and Super 2080 would replace their standard versions. The new Super GPUs will be packed with two free games supported by Ray Trick: Remedy Entertainment's Control and Bethesda Wolfenstein: young blood.

As the Super GPU lineup comes on the market, retailers will automatically reduce GPU prices to the standard version. The RTX 2080 Ti will remain the company's premium offering, with the RTX 2060 being the lowest option in the gaming space.

NVIDIA wanted to put pressure on AMD with its Super GPU range, but it has played and reduced the prices of its Navi-based GPUs. AMD was able to reduce its prices due to the cost advantage offered by the 7 nm node (nanometer). With its Navi GPUs, AMD seeks to gain a discrete GPU market share. NVIDIA is now looking to enter the 7 nm game.

NVIDIA switches to Samsung for the 7-nm gaming GPUs

Over the past year, NVIDIA has been bombarded with questions as to when it will move from the 12nm node to the 7nm node, while its rival AMD would have surpassed it in node technology. NVIDIA said it was in no hurry to switch to 7-nm GPUs, as its Turing GPUs consume less power than AMD's 7-nm GPUs.

However, a June 5 DigiTimes report indicated that NVIDIA Korea had announced that it would use Samsung's extreme 7nm ultraviolet (SSNLF) node for its new GPUs, likely for the entire range. Until now, NVIDIA manufactured most of its GPUs on TSMC nodes (TSM). He used the Samsung smelter for the last time for his Pascal-based GTX 1050, 1050 TI and GT 1030 cards. Last year, AMD upgraded to TSMC for its full 7nm product line. It may be for this reason that NVIDIA opted for Samsung.

While Samsung struggles to overtake TSMC in the 7nm node space, the former would offer good discounts to attract customers. This competition between foundries will help NVIDIA produce its new GPUs at a lower cost, which it will eventually benefit for its customers.

NVIDA 7 nm gaming GPU coming next year

NVIDIA has already launched an updated version of its Turing-based GPUs on the 12-nm node. Citing market sources, DigiTimes said NVIDIA's 7-nm GPUs will be built on its next generation of architecture. Samsung would have started sampling its 7nm node with customers such as NVIDIA by the end of the year and starting volume production by mid-2020. NVIDIA launched its Pascal series in mid-2016. With regard to the production and launch schedule of Samsung and NVIDIA, we can expect NVIDIA's 7-nm GPUs to be launched in mid-2020.

Next year will be exciting for the GPU market. AMD will launch its Navi 20 GPUs, NVIDIA will launch its first 7nm GPU and Intel will launch its first discrete Xe GPU.

Risks related to NVIDIA's gaming activities

Until now, NVIDIA was unbeatable in the GPU space, which gave it the power to command a higher price for its products. However, this scenario is changing.

NVIDIA's GPUs are in the top ten on the list of most favorite GPUs by users of Steam, a digital video game distribution platform that allows users to install and update games. NVIDIA's Pascal-based GTX 1060 is the most used GPU for Steam users. AMD's Polaris-based RX 580 graphics processor ranks eleventh. Steam data shows that PC gamers do not upgrade their Pascal-based GPUs.

Players stick to their old cards probably because of the lack of games supported by ray tracing. That's why NVIDIA's decision to unlock the Turing radius tracing feature in Pascal maps did not favor Turing adoption. AMD is preparing to integrate a ray tracing feature with next-generation Sony and Microsoft game consoles, which means that NVIDIA will lose its technological edge. In addition, the adoption of cloud gaming could slow down the adoption of GPUs by players, but this would open a new revenue opportunity through subscriptions.

All of the above factors have affected GPU demand. According to data from Jon Peddie Research, GPU shipments decreased overall by 18.6% in the first quarter.

With its strong balance sheet and profit margins, NVIDIA could engage in a price war with AMD, but that would hurt its margins. NVIDIA's strong margins on gaming and data center activities pay for its research and development activities related to future technologies, such as AI and autonomous driving. Until these future markets mature and start generating sufficient revenue, NVIDIA must protect its margins and market share in gaming and data centers by taking calculated risks.