What are the best specifications of the computer that I can buy to make a project consisting of 3000 images with a resolution of 24 mega pixels so that I can complete the process in a few days and get excellent accuracy and how much does it cost?
Take a look at the various Pix4D articles from Puget Systems.
If its for a once off, consider using a cloud compute solution - much cheaper than buying a whole new system.
We just ran 2000 45MP images on a virtual machine in Google Cloud in 2-3 days. We were using 24 cores, 120GB RAM and a Nvidia P4 GPU. I reckon you could probably do it for free (Google Cloud give you $300 free credit for signing up). I’m sure Amazon Web Service and Microsoft Azure do the same also.
@mike1, welcome to Pix4D Community!
The following description shows the recommended hardware and software requirements:
- Windows 10, 64 bits (Windows 11 is not supported yet).
- CPU quad-core or hexa-core Intel i9/Threadripper/Ryzen 9/.
- GeForce GTX GPU compatible with OpenGL 3.2 and 2 GB RAM.
- Hard disk: SSD.
- Small projects (under 100 images at 14 MP): 8 GB RAM, 15 GB SSD Free Space.
- Medium projects (between 100 and 500 images at 14 MP): 16GB RAM, 30 GB SSD Free Space.
- Large projects (over 500- 1000 images at 14 MP): 32 GB RAM, 60 GB SSD Free Space.
- Very Large projects (1000 - 2000 images at 14 MP): 64 GB RAM, 120 GB SSD Free Space.
There is also more information about this topic for PIX4Dmapper in the following article: Computer requirements - PIX4Dmapper
For larger projects, we suggest the use of the software PIX4Dmatic. You can find more information about computer requirements for this software in the following article. here.
I hope this information was useful for you.
Have a very nice day!
Thanks for your responses,
I have tried the process of 2900 images with a resolution of 24 MP using a server with a processor 2× Intel Xeon E5- 2620 V4 clock speed for one processor 2.10 GHz and core for one processor 16 cores
Using 64 GB RAM, using Nvidia Quadro RTX 8000 graphics card and an SSD Hard Driver, and after 96 hours the program stopped, and I had to close it
Therefore, I am hesitant and afraid to buy a new device and lose large sums of money, then let me down and not achieve what I wish from it.
I’d reiterate that rather than buying your own high powered hardware I’d rent someone elses.
Also, there are some settings that end up massively blowing out the processing time - if you’re mainly after the mosaic, for example, there is no need to max out the settings for point cloud densification. This step is normally the single most time consuming, so using optimal cloud density and a smaller image scale (1/2 or 1/4) will make things go faster, keep the program acting a bit more stably, and likely won’t materially impact the quality of the mosaic.
I’d suggest throwing more RAM at the problem, just as a guess.