Nvidia's cost-effective card GTX 970 in her previous life was severely defeated due to the video memory gate incident and was boycotted, resulting in AMD's big sale!

This time, I can't escape such a fate!

Even Wang Yi plans to wait for Xingyi Technology's RTX 960 to test the performance after the successful RTX 960 chipping.

If you have strong performance, you can change your name directly to RTX 970 and compete with Nvidia's GTX 970!

Take his life while he is sick!

If the GTX 970 does not fail, it will be stressful after the failure.

The GTX 970 that has failed is undoubtedly the best opportunity for Xingyi graphics card to become a god!

Opportunities must not be missed, and no more mistakes will come.

Wang Yi quickly made arrangements to let the graphics card department continue to work hard.

Nvidia's video memory door crashes is a once-in-a-lifetime opportunity. If you miss it, it will not be tolerated by nature!

"Also, the optimization and compatibility of mainstream games are also a problem." Wang Yi remembered this: "You all have deployments, right?"

The compatibility of the CPU is actually pretty good, but the compatibility of the graphics card is really a problem.

Williams nodded: "Chairman, we have always attached great importance to compatibility issues. We have established a compatibility optimization department long ago, not only the CPU optimization department, but also the graphics card optimization department. Of course, our graphics card has not yet been leaked, and the graphics card optimization department's main job is to optimize the graphics card."

"After Xingyi CPU streaming, the graphics card optimization department and the CPU optimization department have already started optimization work day and night. There are two teams on both sides working together to optimize. For mainstream applications and games, all-round optimization is carried out, and one bug is found to solve the other bug."

"It is precisely because of this that our CPU compatibility is very good, and the compatibility of core graphics is also very good."

"As for the compatibility of graphics cards, it is a little more difficult. However, in the past few months of optimizing graphics card verification, our team has accumulated a lot of experience and is well aware of the characteristics and performance of mainstream games, and has established corresponding databases based on similar games or similar games."

"After the graphics card is being streamed, based on the data of these databases and the experience of core graphics optimization, we can quickly promote the optimization of graphics cards."

Wang Yi nodded: "Very good, this arrangement is very good. However, the graphics card streaming will start to be optimized in July and will be launched in September. Two months is not enough, right?"

"Yes, Chairman, two months are really tight."

"Then recruit more people and use the number of people to make up for the time gap!" Wang Yi said decisively: "Especially experienced talents, even if they are digging, they will be at any cost."

"Okay, Chairman." Williams smiled, familiar operations.

What Mr. Wang likes to do the most is to invest money in research and development and invest money to poach talents!

Under this kind of money offensive, talent is not a problem, and under the money + talent offensive, time is not a problem.

A team of 60 people will do a graphics card driver optimization. One person is responsible for many games and software, which may take half a year to optimize.

But a team of 300 people works to optimize the graphics card and multiple people are responsible for a game, which is almost the same for more than a month.

Graphics card optimization is designed for different games, and can be deployed in different teams. It can be fully promoted in the violent army mode and multiple lines at the same time.

Currently, Nvidia's graphics card driver team has only a thousand people in total, and only three or four hundred people are responsible for compatibility optimization.

Nvidia releases five or six different graphics cards a year, and only 50 or 60 optimization teams are assigned to each graphics card.

Xingyi Semiconductor recruits a team of 300 people, and is currently only responsible for the optimization of one graphics card, so the efficiency is naturally several times that of Nvidia fifty or sixty people responsible for a graphics card.

In addition, Nvidia's latest graphics card GTX 970, 980 was only launched in September, and the optimization was also average in the past few months since it was launched.

If they optimize well, there will be no video memory gate incident.

In this way, Xingyi Technology's RTX 960 may not be at a disadvantage with the full optimization of gathering hundreds of people, and may even have better compatibility!

Nvidia Headquarters.

Lao Huang is in a good mood recently: "Wang Yi can play, but he actually goes to make a computer processor. Hahaha, it's so good, I'll have a headache next time!"

The subordinates agreed very much: "Yes, Wang Yi turns the direction and makes efforts to force the computer CPU and abuse Intel, so we will be relaxed. It just so happens that Intel can also experience the fear of being critically hit by Xingyi Technology and being dominated by Xingyi Technology, hahaha!"

"That's right, how can we lose Intel after Nvidia? We have to go through it once. But I didn't expect that Xingyi Technology is powerful. She just made a computer processor and then broke I3, I5 and AMD, especially nuclear graphics, which is really powerful, not as powerful as nuclear graphics. But it's okay, Intel is the one that needs headaches, hahaha."

Old Huang gloated and felt refreshed.

Previously, Nvidia entered the mobile chip market in order to expand its business.

In 2012, it was favored by Xingyi Technology with the world's first quad-core mobile chip Tegra 3. The sales of tegra 3 were so popular. Nvidia made money and made sluggish money. At one time, it felt like it was OK and wanted to make great strides in the field of mobile phone chips and kill all sides.

But the good times didn't last long. Soon Xingyi Technology launched its self-developed Kunpeng chip, which not only replaced the Nvidia chip market, but also stole many Nvidia chip customers!

As you can imagine, the tegra 4 didn’t sell much at all last year, and it was a big loss!

It can be said that how did tegra 3 make money the year before yesterday, and how did tegra 4 lose money last year!

After comprehensive calculation, not only did it not make any money, but it also made serious losses.

In addition, the baseband problem is really unsolvable, so Huang simply gave up his mobile phone chips and turned to car chips, tablet chips, and even TV chips that do not use baseband.

This year, the tegra K1 was launched!

I thought this track was easier to get along, but unexpectedly, Xingyi Semiconductor turned around and launched a chip dedicated to flat panels, TVs, and car machines, Kunpeng 902C, Kunpeng 706C, Kunpeng 508C...

Not only does it have powerful configuration, lower power consumption, better optimization, better compatibility, and has a 64-bit architecture, and covers all high, medium and low-end ranges, allowing customers to choose on demand. It is not comparable to a Nvidia chip.

As you can imagine, Nvidia's tegra K1 is also in a state of disgrace.

It was mass-produced in the first half of this year, but by July, no one was sold!

This annoyed Lao Huang!

"How is Audi? Do you want to use our K1?" Lao Huang couldn't help asking, feeling sad when he thought of this.

The subordinate frowned: "Chairman, Audi originally planned to use our K1 chip to make car mobile chips, but I heard that Xingyi Technology will soon launch the Kunpeng 910 SOC and Kunpeng 910C, so I changed my mind."

"Fake!" Old Huang couldn't help but curse: "Kunpeng 910 SOC is dedicated to mobile phones. Kunpeng 910C has no baseband and is just used for car machines. Therefore, Audi plans to give up our tegra K1 and choose the Kunpeng 910C of Xingyi Technology?"

"It's possible, but there's another possibility. They plan to use the Kunpeng 910 SOC with baseband directly!"

"???" Old Huang was confused: "There is a baseband for the car chip?"

"Chairman, it is a trend for luxury cars to come with communication functions. For example, when encountering traffic accidents, emergency SOS services, fault repairs, etc., baseband communication is required."

"For example, Mercedes-Benz, their fuel vehicles are equipped with these services. If something happens, it's okay if the phone can't be found. You can ask for help from Mercedes-Benz service center with one click. The service center will immediately locate the vehicle and call the police for rescue."

"As for trams, future car-machine navigation, etc., baseband online navigation is also needed."

"With our tegra K1, there is no baseband, and it can only be shared through mobile phone traffic and Bluetooth. But with Kunpeng 910 SOC, you can directly get a traffic card, online navigation, entertainment..."

"Fake! So, future car machines cannot be separated from baseband? Then our tegra K1 is also useless?"

“It’s theoretically.”

"No wonder Audi has agreed to place an order, and they all started to hesitate! Farke! Farke!"

Lao Huang cursed: "I still can't get around the baseband!"

"Wouldn't that mean that our tegra K1 can only make tablets?"

"It's such a chairman. But the problem is that most tablet manufacturers have directly purchased Kunpeng 902C, Kunpeng 706C, Kunpeng 508C..."

Lao Huang's previous gloating was gone, leaving only the corner of his twitching mouth: "!!!"

"Forget it, then let's learn from Xingyi Technology, develop Nvidia tablets, and make our own tablets!"

"Okay, Chairman, this project is progressing smoothly. NVIDIA Shield Tablet is expected to be released in the third quarter and will be listed in the fourth quarter."

"Well, with the strength of the tegra K1 desktop-level GPU, NVIDIA Shield Tablet can crush the 2nd and 2pro generations of Xingyi tablets, which will definitely be popular all over the world!"

"This is for sure. Although the performance of Xingyi Tablet 2pro is good, it is still far worse than the tegra K1. As for their third-generation tablet, it will be next year. Our NVIDIA Shield Tablet is destined to be popular."

"Okay, make the prototype as soon as possible. There is no problem. Just mass-produce it as soon as possible, and strive to release it in September and launch it in October!"

"yes!"

"We lost too badly before, too much. This time, I must win one game in Xingyi Technology! Win one game in Wang Yi!"

Old Huang looked into the distance with a bright look.

But what Huang didn't know was that when Nvidia was defeated repeatedly and was preparing to turn to the tablet market and sneak attack on Xingyi tablet 2nd generation, its own graphics card base was in danger!

After the X7 4800 processor streaming was successful, Xingyi Semiconductor immediately began streaming work on its own graphics card RTX 960.

After several days of preparation, the filming officially began.

Without any surprises, the RTX 960 streaming was successful and a series of tests began immediately.

On the day the test results came out, Wang Yi personally came to Xingyi Semiconductor R&D Center to wait.

The current R&D factory is independent and has become a slitting factory, specifically used for slitting of new chips.

Whether it is a mobile phone chip or a computer processor graphics card, it is all here to cut the movie.

Yu Longxing, director of Xingyi graphics card department, took the test results and walked out:

"Chairman, all the test results are out!"

“The yield rate of RTX 960 reaches 85%!”

"85%!" Wang Yi was in a good mood: "Very good, the yield rate of 85% is 85%. The official mass production hopes to reach more than 90%. It's very good! How is the performance?"

"Chairman, Nvidia GTX 960 and 970 have not been released yet, and we don't know the specific performance. We can only compare the high-end dessert gaming card GTX 770 released last year with the high-end flagship gaming card GTX 780."

“Because our RTX 960 memory adopts a more powerful 4G GDDR5, it is still performing well ahead of Nvidia GTX 770’s 2G GDDR5 and GTX 780’s 3G GDDR5.”

"According to our tests, our RTX 960 performance is 10%+ than the GTX 770 in many games, but it is still a bit worse than the GTX 780, with a gap of about 15%.

"Of course, it's not that our RTX 960 is not as good as the GTX 780, but because our graphics card has not been optimized yet, and the 780 is the flagship high-end graphics card released by Nvidia last year, and the optimization is already very good."
To be continued...