Skip to main content
  • Home
  • Library
  • Read
    • Popular
    • Charts
    • Browse
    • Categories
    • Wikis
    • Glossary
  • Request
  • Activity
  • Leaderboard
  • Help
Our Apps
  • DaoSearch
  • Narratorr
Social
  • Discord
  • Patreon
Tech Nation: My Creations Have Upgrade Panels
Chapter 243

Powerful Starlight S80 Graphics Card Draws Gasps from the Entire Crowd!

1,178 words 4 min read 1 view 2mo ago Source ​
Skip to the end of the chapter
Prev
Next

Traditional DVU lithography machines are limited by the resolution of the light source. Once they reach 28 nanometers, the manufacturing cost, time cost, and labor cost to continue to go down are very high.

For example, if the world wants to use DUV lithography machine to produce 14-nanometer chips, the only option is to use repeated exposure technology.

This results in excessively high overall manufacturing costs, time costs, and effort costs.

Conversely, EUV lithography machine are different because they only require one exposure, which naturally increases production efficiency and reduces the amount of effort wasted.

What's even more remarkable is that EUV lithography machine, due to their resolution advantage, are naturally adept at chip production using processes below 7 nanometers.

Therefore, the domestically developed EUV lithography machine we have developed is a very good lithography machine, and its upper limit is actually not just 10 nanometers.

Of course, to conquer the future 7nm, 5nm, and 3nm processes, we ultimately need to work diligently and practically in research and development!

Upon hearing Lin Feng's words, the people present once again applauded excitedly, cheering for Lin Feng's inspiring speech.

It has to be said that after Lin Feng's words, they had high expectations for the future of Great Xia chips.

Xingtu Technology spent 50,000 people and more than half a year to finally succeed in the localization of the 10-nanometer process technology.

At this point, people at the scene, including those on the internet, believed that Lin Feng's claim of mastering even stronger 7-nanometer, 5-nanometer, and 3-nanometer technologies had a high probability of success.

Under such circumstances, how could they not feel hopeful?

After the applause subsided slightly, Lin Feng continued to explain the 10-nanometer process technology in conjunction with the video.

The article recounts who made significant contributions during the entire 10-nanometer process development process, the challenges encountered, and so on.

Through Lin Feng's explanation, people at the scene and online had a more realistic understanding of the research and development of the 10-nanometer process technology.

Therefore, they expressed their deep admiration and gratitude to the more than 50,000 R&D personnel from Xingtu Technology and various research institutions across the country.

It was through their collective efforts that they finally created a miracle in just over half a year, successfully breaking through foreign blockades and developing numerous chip manufacturing equipment using the 10-nanometer process.

Time passed slowly, and in the blink of an eye, more than twenty minutes had gone by. Lin Feng, having finally finished discussing the development process of the 10-nanometer process technology, waved his hand.

Then, a graphics card image appeared on the projection screen behind Lin Feng, with a note next to it.

The graphics card shown above is codenamed 'Starlight S80 Gaming Graphics Card'!

Indeed, the naming convention for the graphics cards released this time is clearly a tribute to the S80 graphics cards of Moore Threads.

After all, the team responsible for developing the graphics card was the original team behind Moore Threads in the previous life.

When Lin Feng wanted to enter the graphics card business, in addition to acquiring VIA's graphics card division, he also bought their graphics card business and R&D team.

Their biggest breakthrough was successfully poaching the founding team of Moore Threads, getting them to leave Nvidia and join Xingtu Technology.

This is because graphics cards, or computing cards, are extremely important, given the immense computing power required by the Artificial Intelligence Jarvis.

Therefore, Lin Feng provided the team that switched jobs with maximum resource support, not only offering them extremely high salaries, but also providing them with a Hawking- level Gene Enhancement Potion boosted their IQ to 160+.

As a result, Xingtu Technology has successfully developed two graphics cards in less than a year, from the initial project to the present.

Of course, Lin Feng also contributed a lot to this, solving many problems that could have held them back for months.

Therefore, these two graphics cards were developed by both them and Lin Feng.

That's why they were able to develop so quickly, successfully launching two graphics cards simultaneously in less than a year.

Of course, let's get back to the scene.

At this moment, the people at the scene and those watching the live stream were quite curious about the 'Starlight S80' graphics card.

I don't know what level of graphics card performance this 'Starlight S80' can achieve, or what kind of Nvidia graphics card it can actually rival.

"As you can see, this is our Starlight S80 Graphics Card, which comes in two versions, or two configurations."

The biggest difference between these two configurations is that the high-end configuration has two graphics card chips, while the low-end configuration only has one.

Their names are Starlight S80 and Starlight S80 Plus, respectively.

Let me first introduce its minimum configuration, which is the version with only one graphics card chip.

While Lin Feng was speaking, the projection screen behind him displayed the performance parameters of the low-end version of the 'Starlight S80'.

Production capacity name: Starlight S80

Manufacturing process: 10-nanometer process technology.

Number of transistors: 46.2 billion transistors.

Core frequency: 3.2GHz~4.5GHz.

Number of stream processors: 9728.

Memory bus width: 256 bits.

Graphics memory capacity: 16GB.

Memory type: GDDR6 memory (Xingtu Technology's custom GDDR6 memory standard).

Memory bandwidth: 716GB/s.

3D API: Supports DirectX 12.

Maximum resolution: 7680×4320 (8K resolution)

Bus: PCIe 5.0 (Xingtu Technology's custom PC bus standard, compatible with PCIe 3.0 bus standard.)

Rated power consumption: 350W.

Featured technologies: Supports ray tracing, supports the StarPath MUSA architecture, supports multi-GPU technology, supports Deep Learning Super Sampling, and perfectly supports the Unreal Engine 4.

Output interfaces: 1 x HDMI port...

Looking at the performance parameters above, everyone on site widened their eyes, their expressions filled with disbelief.

Because the above parameters are so luxurious, is this really a low-end configuration or a high-end configuration?

The hardware specifications mentioned above are somewhat exaggerated, to the point that they are almost unbelievable.

People don't really care about other things like memory bus width and memory type, but they are very sensitive to parameters that can significantly improve game performance.

Take stream processors, for example. In games, the more the better, because the game screen is constantly being calculated in front of people.

Because the game's graphics are too complex, the only way to speed up the graphics is to use synchronous calculations.

This means assigning computational tasks to various stream processors to handle the computation.

Therefore, the more stream processors the better, as more stream processors mean less need for queuing.

Ultimately, when players are playing the game, they can clearly feel a significant increase in frame rate, with no lag whatsoever.

Therefore, the number of stream processors is definitely a key metric affecting game performance.

But now this so-called low-end S80 graphics card has 9728 stream processors. Doesn't that mean there will be no lag when playing games?

End of Chapter
Prev
NovelSummaries
Next

Explore the wiki

Lin Feng
Xingtu Technology
Great Xia
Jarvis

Comments

No comments yet.
AboutStatsAPIChangelogStatusContactPrivacyTermsDMCA
© 2026 RandomTranslator·v3.26·RSS