Production and Conversion on Your Own Computer with CartaX Studio

Not every job needs the cloud. If you have a suitable computer, CartaX Studio produces the same product package — orthophoto, DSM/DTM, classified point cloud, class layers — on your own machine, converts your LAS/LAZ and GeoTIFF files into the format the browser opens, and uploads only the result to your project. No tokens are spent; no 40 GB of raw data is uploaded. This guide walks through the local route and the hardware it needs; for what Studio is and the installation details see What is CartaX Studio?.

Installation: Windows and Linux

• Windows 10 / 11. Get the installer with Download on the CartaX Studio card in your dashboard. The file is not signed yet, so Windows shows "Windows protected your PC"; this is expected: More info → Run anyway. Nothing else is needed on Windows 11; Windows 10 needs Microsoft Edge WebView2.
• Linux. Download and install the .deb package. On Linux Convert works; the engines for production from photos are published for Windows only, so Create jobs on Linux are produced in the cloud.
• There is no macOS version.

Studio is part of the Professional plan; free accounts cannot sign in. At sign-in Studio opens your browser; once you sign in with your CartaX account you return to the app.

The production engines do not ship with the installer; they are downloaded once from the Components card in the wizard: base engine ≈0.3 GB, object recognition (segmentation) ≈6 GB. Conversion needs no engine download.

Conversion: LAS/LAZ and GeoTIFF

Press Convert and pick the file; Studio detects the format itself.

Input	Output	What happens in CartaX
LAS / LAZ point cloud	Ready octree	Smooth 3D viewing in the browser, measurement, sections
GeoTIFF orthophoto / DEM	COG	Georeferenced raster layer on the map, elevation readout

Conversion needs no graphics card; when it finishes the result uploads to the project you chose automatically. Converting the same file in the cloud would cost tokens; local conversion does not.

Production: from a photo folder to products

The Create button opens the production wizard:

1. Under Assets, pick the flight's root folder; subfolders, photos (JPG, PNG, TIF) and RTK logs (.MRK) are found automatically.
2. Under Position accuracy, add a GCP file if you have one; the points are marked in the photos and the model is fitted to them.
3. Under Advanced, set the delivery coordinate system and optional extra search terms.
4. Under Record, enter the project, layer name and capture date (all three are required).
5. Press Produce.

There is no output or quality selection: every job produces the same package at the same high quality — orthophoto, DSM/DTM, classified point cloud and class layers. A textured 3D model is not part of the package.

Studio makes the call: once the photos are selected the computer is checked; if it is up to the job it runs on this computer, otherwise the line reads "Will be produced in the cloud" with the reason, and Produce sends the job to the cloud. Even when the computer is capable you can send the job to the cloud with Produce in the cloud.

What you need to produce on this computer

Requirement	Needed
Operating system	Windows
Graphics card	NVIDIA, at least 8 GB of graphics memory
Memory (RAM)	At least 16 GB; more as the photo count grows
Engines	Downloaded once: base engine ≈0.3 GB, object recognition ≈6 GB
Disk	14 GB for object recognition plus room for the job's temporary files
Job size	At most 800 photos and 20,000 megapixels
CPU	Fewer than 8 cores gives a warning: the job will take long

Memory decides how many photos a single job can take:

Memory	Roughly per job
16 GB	130 photos
32 GB	260 photos
64 GB	530 photos
96 GB and above	800 photos

Object recognition (segmentation) is part of the production package and runs on the graphics card; AMD, Intel and Apple GPUs and integrated graphics are not supported. Your photos' resolution does not affect graphics memory: the model scales the image down to a fixed size.

Today the dense point cloud step uses GPU acceleration best on the RTX 40 series; on RTX 30, RTX 50 and other NVIDIA cards this step runs on the CPU and production can take several times longer. This support is in beta.

If your computer does not qualify, Studio says why (e.g. "at least 8 GB of graphics memory is required" or "the segmentation engine is not installed") and routes the job to the cloud; the cloud has no hardware requirement and is billed in tokens.

No tokens spent, no raw data leaves

Local conversion and local production spend no tokens. Raw photos and raw LAZ files never leave your computer; only the produced outputs are uploaded to the project and count against your storage quota. Outputs are also written on your computer to product folders under Documents / CartaX Studio: Ortofoto (orthophoto), SYM (DSM/DTM), Nokta Bulutu (point cloud), Sınıf Katmanları (class layers), Rapor (report). The report folder holds the run log and, if applicable, the accuracy report — your delivery files stay in your hands.

The Jobs screen: progress, Pause / Resume, Retry upload

The Jobs screen shows overall progress on a single card; expand it to see the job's stages, elapsed time and percentage. One production and one conversion run at a time; the rest wait as Queued.

• Pause / Resume — suspend a running job to free the computer for something else; it continues from where it stopped (it stops immediately on Windows). A paused job keeps its place: the job behind it does not start until it is resumed or cancelled.
• Shutdown safety — if Studio closes or the computer restarts, the job on this computer continues the next time Studio opens.
• Retry upload — if the upload fails on a network error, the products are ready on your computer; this button repeats only the web upload, production does not run again.
• Open log — the detailed run log; the answer to "which stage is it in and how long is left" lives here.

When it completes the layer opens on the map; Measurement elevation: Surface / DSM / DTM appears in the layer settings.

Compared with the cloud

	Studio (this computer)	Cloud
Cost	No tokens	Tokens (by megapixels)
Hardware	Windows + NVIDIA graphics card (8 GB) + at least 16 GB RAM	Not needed
Computer	Busy for the whole run	Can be shut down once the photos are uploaded
Classes (object recognition)	Yes	Yes
Per-job limit	800 photos / 20,000 MP (fewer depending on memory)	2,000 photos
Raw data	Stays on your computer	Uploaded to CartaX

For the cloud route see Cloud photogrammetry production.

Related guides

• What is CartaX Studio?
• Cloud photogrammetry production
• Classes: automatic object recognition
• What are LAS and LAZ files?
• Supported formats and limits

Frequently asked questions

Question	Answer
Does it spend tokens?	No; local conversion and production are free, only the uploaded output counts against storage.
Are raw photos uploaded?	No; they stay on your computer.
Do I need a graphics card?	Not for conversion; production from photos needs an NVIDIA graphics card with at least 8 GB of memory.
Windows says "protected your PC"?	Expected: More info → Run anyway.
Is there a macOS version?	No; Windows 10/11 and Linux (on Linux, production runs in the cloud).
How many photos can I process?	At most 800 photos and 20,000 megapixels per job; your memory may allow fewer (see the table above).
Can I pause a job?	Yes: Pause / Resume; after a shutdown the job continues the next time Studio opens.
The upload dropped — does production start over?	No; Retry upload repeats only the upload.
Do classes come out here too?	Yes; object recognition is part of the package and runs on the NVIDIA graphics card.