The mark you need, per subject
Both pass conditions, not one: the 40% separate end-semester minimum as well as the combined total. Spreadsheets miss the first, which is how a subject that “looked passed” comes back as a supplementary.
KTU B.Tech · 2024 scheme · Regulations 2024
Every other calculator tells you what you already scored.
This one tells you what you still need.
TargetX reads your marks and attendance, applies the KTU rules as written, and answers the only question worth asking before an exam: what do I have to score in this paper?
Free for students, forever — written into the licence.
Your record is a file on your own machine, and no account is needed
to use any of it.
RegulationR 7.5.ii
76% attendance is not “fine”. It is two marks you have already lost.
Under Regulations 2024, attendance is not only an eligibility gate — it is a scored component of your internal marks, worth up to 5 of them, in every subject, in bands. Full marks start at 85%.
So a student sitting comfortably above the 75% line is still forfeiting marks, in every subject, quietly, all semester. No portal shows this. No other calculator computes it. It is the single largest number most students never see.
1
60%+
2
70%+
You, at 76%
3
75%+
4
80%+
5
85%+
CIE marks earned by attendance alone
R 7.5.ii
And below 60%, the door shuts. R 6.2 lets the Principal
condone attendance under 75% only as far as 60%. Below that the exam
cannot be sat, and attending from there on does not change it. TargetX
says so on the day it becomes true — not after results.
What itdoes
How many classes you can still miss
Computed from real attended-over-held counts, with duty leave credited the way R 6.3.ii actually caps it — at 10% of classes held, which grows as you spend it. Answered as a number of classes, not a percentage.
The cheapest route to your CGPA
Name a target. TargetX works backwards to the SGPA each remaining semester has to carry, then to which subject is cheapest to push and how far — balanced so the plan is one a person can actually attempt, rather than four subjects at 95.
Your marks, in without typing them
Sign in to your college’s etlab portal and it pulls every semester, attendance, series marks, grades and published SGPA in one pass. Your password is used for that one request and never stored — or skip it entirely and type things in by hand.
Getthe app
Download
A desktop app — Windows, macOS and Linux. It installs for your user account only, so it needs no administrator rights, and it runs entirely offline apart from the portal you point it at.
Recommended for you
Windows · installer
| Platform | File | Size | Download |
|---|---|---|---|
| Windows | .exe installer, and .msi for deployment unsigned | — | Releases |
| macOS | .dmg, universal not notarised | — | Releases |
| Linux | .AppImage and .deb | — | Releases |
Windows will say the publisher is unknown. The
installers are not yet code-signed, so SmartScreen shows a warning —
choose More info, then Run anyway. Rather than ask you
to take that on trust, every release publishes
SHA256SUMS-<platform>.txt beside the installers, so
you can check that what you downloaded is what was built. In
PowerShell:
Get-FileHash .\TargetX_<version>_x64-setup.exe -Algorithm SHA256
A certificate is on the list. A checksum proves the file did not
change in transit, not who built it — it is not a substitute, and this
page will say so until one is bought.
macOS will refuse to open it outright. The build is
not notarised by Apple, so Gatekeeper does not offer a way through —
it reports that TargetX is damaged, or that the developer cannot be
verified, and stops. It is neither; that is the message macOS gives
any unnotarised app. Open System Settings → Privacy &
Security, scroll to the note about TargetX being blocked, and
choose Open Anyway. If it does not appear there, clear the
quarantine flag the browser attached:
xattr -dr com.apple.quarantine /Applications/TargetX.app
Check the SHA-256 above first. Running that command on software you
have not verified is exactly how the protection is meant to be used
against you, and it is worth being suspicious of any page that tells
you to type it — including this one.
Updates come to you. The app checks for a new version
a few seconds after it opens and offers it; you accept or decline.
Update payloads are signed, and a build only installs one signed by the
key compiled into it — an unsigned or wrongly signed update is refused
rather than installed.
Where yourdata lives
On your machine. That is the whole answer.
Your marks are a file in your own user profile. Nothing about them is transmitted anywhere, and there is no telemetry, no analytics and no crash reporting — the app makes network requests to exactly four places, and the fourth only if you choose to sign in:
- github.com
- Once on launch, to ask whether a newer version exists. Block it and the app works unchanged.
- raw.githubusercontent.com
- Once on launch, to refresh the KTU course catalogue, so a curriculum revision does not need a new build.
- Your college portal
- Only when you press Sync, and only to the address you typed. Your password is held for the length of that one request — never written to disk, never put into the saved record, never in the export, never in a log.
- The question box
- Only if you sign in, and only for a question the app could not answer by itself — most are answered on your machine and never leave it. What is sent is your question and the code and title of your subjects. No marks, no attendance, no CGPA, no name, no register number. What comes back is a destination, not an answer: one of five screens, or one of your own subjects, and anything else is rejected. Every figure you then read was computed here.
Sync has been proved against one college’s etlab so far.
If yours lays its pages out differently, the sync fails and shows you
what it could not read rather than recording half a semester. You can
check whether it will work without giving anyone an account:
save your academics page from a browser you are already signed in to,
and drop the file onto Data → “Will sync work at my college?”.
The page never leaves your machine; what the app offers you to send is
a separate block of table shapes with every digit blanked out.
Forcolleges
Students never pay. Institutions do.
TargetX is licensed under BUSL-1.1, and the right of an individual student to use it free, forever, is written into the licence rather than offered as a policy that could be withdrawn. Deploying it across an institution is a commercial use.
- Install scope
- Per user, not per machine — no administrator rights, and a student’s academic record stays out of a shared location on a lab machine.
- Silent deployment
- NSIS /S and msiexec /qn, for Intune, SCCM or a login script.
- Reads only
- The app has no code path that writes to a portal. The session cookie lives in the native process, where nothing in the web layer can reach it.
- The detail
- DEPLOYMENT.md answers “what is this and what does it touch?” — every claim on this page is checkable there against named files.
Somethingwrong?
A wrong number is a bug, and it is worth reporting.
TargetX has been proved against one college’s portal and one set of regulations read carefully. Neither of those is the same as being right everywhere. If a figure it shows you disagrees with what your college recorded, the college is right and this is a defect — and it is one worth knowing about, because the same arithmetic runs for everyone.
Report a problem
GitHub issues
What makes a report actionable. The subject code, the
figure TargetX showed, the figure you expected, and where the real one
came from — a portal page or a grade card. If a sync failed, the app’s
Data screen names the log file; its contents are error messages and
nothing about your marks, and you can read it before you send it.
Never paste your portal password into an issue, and
blank out your roll number: an issue is a public page.
Questions and “will this work at my college?” belong in discussions rather than issues. Security problems should not be filed in public at all — use GitHub’s private advisory form, which is only visible to the maintainer.