return to bolang life

Chapter 392 Scode Beta Version Launched

Probably due to the tradition of Bolang, or because several of the main co-founders are from Huagong, Lao Yin, a postgraduate student, also put the main development part of the Scode project in Huagong.

Laboratories were also set up.

It's called the Programming Language Discovery Lab.

Although Bolang and Huagong have cooperated a lot in research projects, only the three aspects of systems, chips, and programming languages ​​have really been established independently.

After a series of developments, it has been upgraded.

For example, the original mobile device operating system laboratory has been upgraded to an operating system exploration laboratory.

The integrated circuit design laboratory has also been upgraded to a very large-scale IC design exploration laboratory.

Coverage has been maximized to the relevant domain.

Mainly also brought the word 'exploration', which means infinite possibilities.

At present, under the overall needs of Bolang, the pressure on the programming language exploration laboratory is relatively high, because the time period is very tight.

Wen Liang walked to the laboratory with Li Ze.

On the way, Li Ze was also a little bit embarrassed: "During this time, Lao Yin should be under a lot of pressure. Bowen is in Moscow now, and Lao Yin has to take time to take care of his business, mainly Bolang Lab. It is also difficult for ordinary engineers to make decisions, and it is really not easy to make Scode in line with such a fast efficiency."

"Yes." Wen Liang deeply agreed, "Everyone has worked hard."

"That is to say, it's not the time yet. Otherwise, I should really pull you co-founders out to show off. You are my most proud 'work'."

"What I am most gratified about is that the development of Bolang Group so far, whether it is key technology or key administration, is all handled by our co-founders."

"I have to say, I have very good eyesight."

Walking in the familiar yet unfamiliar campus, Li Ze looked at the kind-hearted Wen Liang next to him, and wanted to refute a few words, but felt that there was nothing wrong with what he said. He himself was indeed very good~
Now in the business world, who doesn't know him, Mr. Bo Langze.

Therefore, Li Zechun expressed his agreement with a full face: "It is indeed the case. Leaving aside other aspects, your vision is really unique in the world."

"Good job." Wen Liang glanced at Li Ze, tsk tsk with contempt.

Although the Huagong campus was a few kilometers deep, the two quickly walked to the building where the laboratory was located.

After brushing his face seriously, he was able to finally enter.

Soon, Wen Liang and Li Ze, who were still sighing and sighing just now, were 'shocked' by the scene in front of them.

Especially when both of them saw Sun Baoyin in the main position at a glance, his refreshed and relaxed appearance was even more... indescribable.

high pressure?
Hard?

not easy?
You can go to Nima's.

Boss Yin doesn't even have a dark circle under his eyes!
I didn't stay up all night at first glance, probably because I didn't even have 996.

However, there is no need to ask too much about this matter, and there is an answer. The people in the laboratory alone have clearly shown the answer.

The so-called need for 996, need to stay up late, need to work overtime, in fact, there is only one reason:
Not enough staffing.

一些还算有点良心的996也无非是用1个人来干3个人的活，给1.5~2个人的工资，已经显得很不错了。

From top to bottom, Bolang Group has been rejecting 996, rejecting overtime culture, and rejecting the so-called 'hard work'.

Expenditure on labor costs is very loose.

Therefore, Bolang Group is always short of people, and there are always not enough people.

The solution of Bolang Group has always been: each employee should complete a reasonable allocation of tasks within five days and seven hours. If the total amount of tasks exceeds, then add people.

At the same time, Bolang Group pays employees about 1.5 times the average salary of the same position in the market.

That is to say, employees can not only have a relaxed working schedule, but also get ideal salary.

This also makes Bolang's creativity in many positions beyond the imagination of the outside world.

In the current simple view of the earth, human creativity is endless. The model of Bolang has shown good creativity at the current stage, and it has also brought very rare industry competitiveness.

The most rumored sentence about Bolang in the outside world is: the employees who can work overtime in Bolang are all the kind of awesome.

Bolang's model is actually volume, but it is a benign volume.

At present, although it has not yet reached the point where everyone has a bright future in the true sense, and everyone who has joined the Bolang Group or newly joined the Bolang Group can obtain the means of production, but at the shareholder meeting a few years ago The decisions finalized above will be implemented, and the final direction will definitely be like this.

There were just a few people who had just been idle and greeted Wen Liang and Li Ze, and the atmosphere in the laboratory was not at all disturbed by their arrival.

Li Ze nodded with a smile, and then said to Wen Liang: "No wonder it can be completed three working days ahead of schedule."

"Mr. Wen has to have a vision. The efficiency of this model is really impressive, and it's also really cool."

Wen Liang put his hands behind his back at the moment, pretending a little bit, just nodded slightly, and then scanned the environment of the laboratory with an unhurried look.

Sun Baoyin has come over to report: "Mr. Wen, Mr. Ze, the first stable beta version of Scode has been offline."

Li Ze made a gesture: "Then let's have the first actual demonstration."

Sun Baoyin nodded, made a gesture, and invited the two of them to an open workstation.

There are several young people gathered together.

At this moment, someone naturally introduced it.

"This is a very rough demonstration environment. The parts have not been assembled yet. This is the host part, including a new chip that has not yet officially passed the test..."

"Of course, the new chip has passed the initial screen lighting experiment in the extremely large-scale IC design and exploration laboratory next door..."

"What we are going to conduct now is also a relatively elementary experiment. Can the star system recompiled through Scode be run in this environment..."

"Then run the recompiled Little Orange Book and the international version of Little Orange Book..."

"..."

In this young man's eloquent introduction, both Wen Liang and Li Ze understood the current stage.

forgot to say.

As early as January, the extremely large-scale IC design and exploration laboratory led by Qin Zheng had carried out in-depth cooperation with Huahong Hualiwei on tape-out.

Because the cooperative relationship between the two parties is very harmonious.

The first fully self-developed 40nm process SoC went through a series of pre-preparation, tape-out, and post-testing, and finally rolled off the assembly line two days ago.

Of course, there will be a lot of debugging and testing work in the follow-up, so that it can finally go to mass production.

This is completely different from the previous ARM chip. That thing is so simple that Bolang can't really blow it up. It doesn't even move anything at all. Just hand over the drawings to TSMC and it can be offline without debugging.

It's just that in order to optimize and integrate more modules on one SoC, Bolang spent some time and invested some funds in it to do some research and drill the design team by the way.

Now this SoC, whether it is CPU, GPU or other necessary modules, is from Bolang, and the overall time-consuming is one year, three months and eleven days.

The current rough calculation of the overall cost is a little over 23 billion.

Some include overlays on ARM chips.

Just as Qin Zheng said, we cannot negate the nearly one-year-long research in the early stage just because the instruction set was formed in December 12, and because it took only more than a month to connect the tape-out process at the end of January 12. with try.

In the end it looks very efficient.

But this is based on a series of massive preparatory work such as re-starting several times in the early stage, hands-on training, study visits, in-depth research, and recruiting.

Numerous controlled group experiments were also conducted in the middle.

More recently, controlled experiments have been pushed to the limit.

For example, the initial determination is relatively smooth, and the docking with TSMC's 28nm process is also being carried out simultaneously, including the tapeout of another different CPU core (core).

Just like ARM v8 has different cores of A57, A53, and A35, Bolang's self-developed CPU also has two cores, one big and one small, and the number of transistors, performance, etc. are different.

The current SoC is not only a 40nm process, but the CPU part is only the design of the small core used, and the tape-out of the large core is still in progress.

On the contrary, the GPU on the mobile phone is not so particular. With different CPUs, it is only to adjust some aspects such as the ALU (arithmetic operation unit) of the GPU...

Now back to the experiment itself.

What needs to be verified is the support for compilation of the Xinghe instruction set in the IDE (Integrated Development Environment) of Scode, so all software-level parts including the Xinghe system need to be recompiled through Scode.

Yes, the star system is just a super-large complex software, and it is also composed of codes and then compiled.

Before Wen Liang and Li Zelai, the laboratory had only conducted split-body experiments, but no overall experiments.

Now, after introductions by several youngsters, the operation process begins.

On a pile of 'scattered' parts... let's call it a mobile phone, the initial installation of the star system began.

In addition to the display on the small screen, it was also output to the large display screen so that Wen Liang and the others could see it more clearly.

The process is completed quickly.

Then it's officially launched...

no problem.

The system drove all the hardware, and immediately someone started touching the small display, tried every angle, and there was no problem.

Then, the software installation that was compiled earlier was performed directly by touching the screen.

Then comes violence.

It's all fine.

Finally, Sun Baoyin breathed a sigh of relief: "At least for now, Scode has no major problems in terms of basics, and it supports the compilation of major mobile phone programming languages ​​such as Java and C."

"Try a small software written and compiled in the 'Starlight' language."

Naturally... no problem.

Then the laboratory couldn't help but burst into warm applause.

The process after that is a bit simpler.

Sun Baoyin initiated an online review report meeting, with Wen Liang and Li Ze present, the efficiency was naturally full.

"...The current situation is like this. Scode can only be regarded as an internal test version. Scode itself is almost the same. After all, it has the foundation of the previous ecS and accumulated there. It is only through the accumulation of open source projects that a self-developed version has been launched. Before and after Made some improvements.”

"The main problem now is the improvement of massive documentation on the new programming language of Starlight, and the establishment of forums. After all, our ultimate goal is to hope that global developers can use Starlight to write programs that adapt to the Starlight system to achieve the best user experience. Effect."

"On the other hand, the new language of Starlight may have some flaws and needs to be perfected. It was originally planned to be used among the hundreds of thousands of programmers in the company, but now it is a bit too late."

Li Ze also expressed some opinions.

In the end, Wen Liang made a decision on the spot: "Time is tight now, but it's not that urgent. The original plan was to launch on the 16th, but now it can be released on the 16th and 17th. It will be used internally in the next few days, and it may be very simple to try to solve some bug."

"After all, this thing is only for the developer group, and the trial sample within the company is large enough."

This is of course no problem.

…………

…………

November 3th, at two o'clock in the afternoon.

Bolang and Bolang Terminal promoted the internal beta version of Scode through their respective official channels, and announced that the starlight forum, which has been used internally for a short period of time, is open to the public.

The official description is to the effect:
"eclipse-Stars (ecS), an open source integrated development environment, will stop updating from now on, and will stop technical support on March 2013, 3, and will be replaced by the newly launched Scode;

Scode includes the high-level programming language "Starlight" independently developed by Bolang, which has become the official recommended programming language of Stars App Store and Stars Apps, and is recommended for global developers to download and use. "

On the Starlight forum, the official documentation is very rich.

Among them, it systematically introduces the specific problems that Starlight, a new programming language, is mainly developed to solve, and mentions the syntax advantages of Starlight, and the most important point: the most perfect support for Starlight system, etc.

At the same time, relevant news has also been promoted in overseas markets.

The English version of the Starlight Forum was established directly and synchronously.

Relevant news quickly became a hot topic in the international edition of Xiaochengshu and Xiaochengshu under the vigorous promotion of Bolang.

Domestic netizens were shocked for 1 years.

Those active IDs all came out.

The machine circle is holding the handle: "??????? Don't ask me why I have so many question marks. I can't bear it when the Star system is launched. After all, it's just Unix-like. It's not too much work to change it. Did you come up with the Starlight programming language without notice? Completely independent research and development? Targeted adaptation of the Star system? Solved [-] specific problems? Bo Lang, you have gone too far! How can you be so awesome!"

"Forget it all, the forum is so complete? There are so many discussions and replies on specific issues, and there are so many Cao Nima documents on the horse. How long has Bolang been doing this behind our backs? ?”

"Holy shit, let's not talk about anything else, Scode does a great job on the front-end UI, it's easier than ecS, and sure enough, only programmers who are on horseback understand what programmers need. As for Starlight, I really created a I tried the project, and it has some meaning, and the cost of learning grammar is not high, so it can be regarded as learning from others. As for the problems solved, it is hard to say, but it does seem to have solved the problem of star application development efficiency."

"Bolang can really play with making wheels. The key point is that you can't say that it is invalid. The star system is now the third largest smartphone system independent of iOS and Android. It can have an official language, but ... Well, I understand why it is rumored that Bolang has 10 development engineers, so many projects need developers."

Overseas markets also have their own opinions.

Some well-known master programmers have their own views on Bolang's move.

Some people dismiss it, thinking that the problems Starlight solves are all pediatrics.

Some people are more in favor of this behavior.

Most of them are indifferent.

After all, it is not domestic.

The launch of the Scode beta version naturally promoted another thing in overseas markets...
-
(End of this chapter)