The exercise method adopted by Li Qingsong was to intentionally delay the time of issuing orders, so that both fleets had a communication delay of about 100 seconds, but one side did not have Combat AI assistance, while the other side did.
Furthermore, the Combat AI was deployed near the battlefield, with no delay.
As soon as the two sides clashed, Li Qingsong realized something was wrong.
Is this artificial intelligence? Or is it artificial stupidity?
In the midst of a chaotic battle, a warship should clearly advance to provide fire support for its allies and create conditions for them to launch Interstellar Missile.
However, during that 100-second communication delay, before it could issue the next command, the Combat AI preemptively steered the warship to retreat hastily.
It did indeed dodge an enemy Interstellar Missile, but it directly tore apart its own defenses, disrupting its own battle plan. This allowed the enemy to seize the opportunity and launch a salvo, firing millions of bullets and causing heavy losses to its side.
On another occasion, this side should have launched an attack, but under the control of the Combat AI, their own warships remained stationary, ultimately causing them to miss their opportunity.
This was a relatively major mistake. There were countless other, lesser mistakes as well.
Ultimately, in this exercise, the side without Combat AI assistance achieved a great victory, "annihilating" the side with Combat AI assistance, even as Li Qingsong endured approximately 100 seconds of communication delay.
"More than ten years, more than 100 million tons of fuel consumed, trillions of bullets used, and this is the result?"
Li Qingsong simply couldn't accept it.
"That shouldn't be happening! How could this be?"
Li Qingsong recalled the Cargo Ship carrying the supercomputer from the battlefield, retrieved the Combat AI's computational data, and dispatched thousands of clones to analyze its code operation and decision-making logic.
After this analysis, Li Qingsong gradually discovered the problem.
It seems I misjudged this Combat AI...
This is because I underestimated the difficulty of real-time battlefield control of warships, as well as the amount of data that needs to be analyzed and processed.
It's not that this AI is stupid, but rather that at the beginning of the battle, the amount of data and computation was too large, which caused the supercomputer to overload and become unable to keep up, resulting in a large number of delays, calculation errors, and logical errors, which ultimately led to this situation.
Li Qingsong breathed a sigh of relief.
"Fortunately, it wasn't a problem with the Combat AI. It seems that the investment of the past ten years or so has not been in vain..."
Once the problem was identified, things became much easier.
Since the computing power of supercomputers is insufficient, then we should develop supercomputers with even greater computing power!
With parallel algorithms and architectures nearing their limits, there is only one way left to build supercomputers with even greater computing power.
Develop more advanced chips!
At present, Li Qingsong has mastered the large-scale manufacturing technology of 45 Nanometer process chips.
Following the technological path already practiced in the human world, Li Qingsong should develop 28 Nanometer process chips for the next generation.
However, based on years of continuous research and technological advancements, Li Qingsong decided to bypass the 28 Nanometer chip and go directly to research the next generation of 20-nanometer chips!
Once the 20-nanometer chip is successfully developed, the number of Transistors on a single chip will increase dramatically from approximately 1 billion to 4 billion.
The overall performance of a single chip will also be improved by approximately 60% to 100%.
But that's not the most important thing. The most important thing is that chips with higher process technology can be adapted to more advanced parallel algorithms and supporting hardware. In a single supercomputer, Li Qingsong can use more chips to perform parallel operations, and the computing power of a supercomputer can be increased by about 40 times on the existing basis!
The prospects are so promising, but going directly from 45 Nanometer to 20 nanometers, bypassing 28 Nanometer, obviously means huge obstacles and extremely high difficulty.
At the Computational Technology Research Base, Li Qingsong once again mobilized a large number of clones to begin research on more advanced chips, building upon the existing research.
After a short period of preliminary research, the technical difficulties he needed to overcome were divided into several categories and presented to Li Qingsong.
"First of all, the wavelength of the Deep Ultraviolet Light Source I am currently using is too long, and the resolution has reached its limit. The diffraction effect causes the pattern to be blurred, and it is not possible to write smaller Transistors structures."
Secondly, traditional Transistors at this size exhibit significant Short-Channel Effect, making it difficult to control power consumption and performance.
Furthermore, the physical properties of traditional silicon gate dielectrics do not meet the requirements, and these also need to be improved...
Li Qingsong divided the clones into several large teams, each tackling challenges in their respective areas.
Through tireless research day and night, a breakthrough was first achieved in the light source.
Li Qingsong has developed Extreme Ultraviolet Light Source with a shorter wavelength, reducing the wavelength of light to only about 13.5 5 Nanometer, thereby greatly improving the resolution.
In addition, Li Qingsong changed the previous process from single exposure to multiple exposure, which greatly improved the processing accuracy by increasing the number of process steps.
After that, Li Qingsong found a material with even better performance and used Metal Gate to replace the Silicon Dioxide gate, achieving another technological breakthrough.
Six months have passed. Li Qingsong has overcome most of the other technical difficulties, but the most difficult and important technical obstacle still stands in Li Qingsong way.
Short-Channel Effect.
When the length of Transistors is shortened to a certain extent, the electric field distribution and carrier behavior will change due to physical laws. This will lead to a series of problems such as a decrease in the threshold voltage of the Transistors and a reduction in the drain-induced barrier, which will seriously affect the chip performance and reliability.
To solve this problem, Li Qingsong has tried thousands of solutions, including optimizing Transistors distribution and design, improving material performance, and improving manufacturing processes, but all of them have ultimately proven to be ineffective.
"All the other problems have been solved, but this one can't be solved, it's really..."
Li Qingsong sighed.
At this moment, the time and resources consumed by the 20 Nanometer Chip Research Institute have exceeded expectations, and have already affected Li Qingsong's overall research plan.
But there was no other way; giving up halfway was impossible. The 20-nanometer chip was far too important to Li Qingsong. Not only would the Combat AI need it, but all other AI, including intelligent factory equipment, intelligent spacecraft control, and even research in Basic Physics, mathematics, chemistry, and engineering, would be inseparable from supercomputers with greater computing power.
On this day, Li Qingsong was still controlling tens of thousands of clones to carry out related experiments, and he also mobilized thousands more brainpower to think about this problem.
Under these circumstances, a thought suddenly arose in the brain of one of the clones and was transmitted to Li Qingsong's consciousness.
"Since Planar Transistor cause Short-Channel Effect, then... could we change the Transistors from a two-dimensional structure to a three-dimensional one?"