Why Tử Vi AkiNet was created

The moment the fire was lit

In busy modern lives, many people face dead ends and lose direction. They do not need empty comfort; they need a tool of self-understanding to know where to head next.

The project began with a vision to bring ancient Tử Vi knowledge, through modern technology, into a transparent mental anchor, helping users recognize their innate potential and shape their own lives.

Why AkiNet needed to exist

We live in an age flooded with information, yet wisdom is rare. Young people get lost among endless choices; adults are worn down by repeated failure. I wanted to hand them a tool that could reach the depth of the soul, to know they were born not to endure, but to build.

Tử Vi AkiNet is designed as an analytical system built on data, artificial intelligence, and Eastern philosophy. By keeping analytical angles transparent and objective, it provides users with a solid grounding for personal decisions.

A promise to every companion

Every feature and content here is developed with respect for the user experience. We commit that every Tử Vi perspective aims toward understanding and freedom of choice, rather than fear or anxiety.

Khởi Nguyên is a commitment: leveraging technology to improve precision, connecting traditional wisdom to broaden perspectives, and maintaining strict honesty in every reading.

Before casting a single chart, we went looking for which chart was the right one

There is a deceptively simple question no Tử Vi site in Vietnam answers in public: which school was the chart you just saw cast under?

We asked that question seriously. Not by reading a few more articles, but by pulling down the live source code of the major chart-casting sites, prying open their reference tables, calling their own data endpoints, then comparing every palace, every star, every brightness rating one by one. Thirteen Vietnamese reference sites were read in full. Six chart engines went into the comparison set. Four sample charts ran through all of them, plus two extra cases built to test exactly one convention: the Tý hour.

What we found was not comfortable. Given the same birth data, the sites produced charts that diverged in ways that mattered: the ruling star of Mệnh disagreed across all four test cases, the Thiên Khôi–Thiên Việt pair swapped places for people born in a Canh year, brightness tables changed depending on a school setting the user was never asked about. One site advertised one school but its numbers matched another. One site could not even render a chart, stopped by a code error sitting in plain sight. One site drew the chart as an image, handsome enough, but not a single character of it could be copied, and a screen reader had nothing to read.

None of this was deliberate wrongdoing. Tử Vi has no single universal answer, it has many lineages, and each has its own reasoning. What stood out was the silence: no reader was ever told which lineage they were standing in.

We chose to say it out loud. Every point of dispute, how the Canh/Tân/Nhâm year stems transform the Four Transformations, whether the Trường Sinh cycle starts at the four growth branches or the four cardinal ones, whether Hỏa Tinh and Linh Tinh are placed by year or by hour, how the Tý hour is handled, how leap months are handled, was written down as a named choice, together with the variants we rejected and why. A chart shipped without its own convention is a chart nobody can reproduce.

We rewrote it from scratch, then caught our own mistakes

A hand-copied reference table is the single biggest source of error in modern Tử Vi software. We know because we caught mistakes inside the printed tables of some of the most trusted sites: a "Thìn" misprinted as "Thì", a "Hợi" misprinted as "Hơi", one Lưu Niên Văn Tinh cell printed into an entirely different palace from the one the other nine year stems all follow.

Catching them was simple: instead of copying a table, we went looking for the formula that generates it, then had the machine rebuild the whole table and check it cell by cell. The five-cục, thirty-day Tử Vi placement table: 150 of 150 cells matched. The fourteen main stars table: 156 of 156. The Cục assignment table: 120 of 120, reproduced completely enough that the printed table became redundant. Five stars that used to live in five separate printed tables all turned out to derive from one star, Lộc Tồn, correct across all ten year stems.

The lunar calendar got the same treatment. We borrowed no library, partly because the available ports carry no clear license, partly because what we cannot compute ourselves we cannot verify ourselves either. The astronomical algorithm was rewritten from first principles, just over a hundred lines, zero dependencies. Checked against a reference chart engine across four charts and five fields each: twenty of twenty matched, plus all three cross-checked Lunar New Year dates came out correct.

The Tý hour is the one hour that crosses midnight, so it is also the one hour where the date used to cast the chart can differ from the date on the calendar, and we measured it separately for that reason. The new day starts at 23:00: someone born between 23:00 and 23:59 is cast under the next day's date, someone born between 00:00 and 00:59 is cast under that same day. Same hour, two different charts, and that rule is stated here so you know which convention your own chart was cast under.

We also wrote down what we are still not sure of: places backed by only one source, edge cases nobody has tried yet, the timezone boundary when northern Vietnam still ran on UTC+8 until 8 July 1967 (and the south until 1975), where the wrong timezone shifts the lunar date by a full day and the chart along with it. The whole algorithm passed 671 internal checks and matched 572 data cells against our measured reference, all exact; a Tử Vi project worth trusting has to be willing to say what it does not yet know, because that is exactly what you need to know before you trust the result.