Even if it rises more than 210% within a year, Nvidia's share price still has 50% room for growth compared to the current level. This is the view of Hans Mosesmann, a semiconductor analyst at Rosenblatt, who raised Nvidia's target price to $200 per share, a historical high. #AI boom# #technology# #macro outlook for 2024#

Mosesmann rated Nvidia as a "buy", with a previous target price of $140.

This bullish expectation will bring Nvidia's valuation to $5 trillion.

Mosesmann said this is what the chip maker deserves.

"We see Nvidia's Hopper, Blackwell and Rubin series driving 'value' market share in one of Silicon Valley's most successful platform product cycles," Mosesmann said.

However, Nvidia's real profit point in the future is not in its hardware-centric GPU business, but in its software business, which is driven by Nvidia's popular CUDA platform.

Millions of developers use Nvidia's CUDA to build large-language models and other programs on the company's AI-supporting GPU chips.

"The real highlight is the software, which complements all the advantages of hardware. We expect software to increase significantly over the next decade in terms of overall sales mix, and valuation will rise due to sustainability," Mosesmann said.

If Nvidia can generate substantial recurring revenue from software, it can reduce the company's risk by increasing revenue predictability.

Nvidia has historically relied on hardware sales to drive its revenue, and hardware sales are typically cyclical and experience periods of highly unstable "boom and bust."

Currently, Nvidia's hardware business is experiencing an unprecedented boom period.

But Mosesmann said it is Nvidia's software business that can help the company achieve its goal of $5 earnings per share by 2026 after the stock split, with a target price of $200 calculated at a 40x PE ratio.

Last month, I/O FundTech analyst Beth Kindig also expressed optimism about Mosesmann's view of Nvidia's software platform.

"The CUDA software platform is the platform developers learn on. So, similar to iOS, this is actually what locks people into an iPhone, because developers are building applications for the iPhone. The same thing is happening with Nvidia where the CUDA platform is what AI engineers learn to program GPUs on, so it helps lock them in, and this combination, now, I call that an unbreakable moat," Kindig explained.

Kindig believes that Nvidia's valuation could eventually reach $10 trillion by the end of 2030, which means its valuation has a potential upside of 205% compared to the current level.