YES Thanks! I was hoping that build 2 is now finally works on all platforms perfectly.
About one second per minute is below the measurement's error margin because you were running them in a multitasking environment. It could be easily the other way around depending how many other tasks are running and how the scheduler picks them. But if it turns out to be presistent, even then the guarantee that the data is physically written to the disk worth that extra half seconds per Gigabytes IMHOCaptainMidnight wrote:Well what can I say, your program can write a raw(dd) image to a 32GB uSD card in less that 1min more than the established competition and with a program size of just 239KB
My understanding is, that with compressed images USBImager is about twice as fast as etcher? Is that correct?
If there are no more bugs, then I'm hoping I can close the Release Candidate phase and go to the Stable Release. For stable I could finally bump the version to 1.0.0 which you'll see in window.CaptainMidnight wrote:Just download a new version again from GitLab - how do I know it's the new version 2 (presume it would be) - it still displays 0.0.1 in the program window?
Thanks! Thank you for your through testing, that was very very helpful! I'm sure I could not done without! I put your name in the READMECaptainMidnight wrote:Well done
So the conclusion:
- For raw images all programs performed equally, the difference being about 1 sec / 1 minute (within error margin of multitasking)
- Win32 Disk Imager does not support compressed images
- balenaEtcher does, but it takes about twice the time
- Of the three USBImager is the smallest, by a lot
- Only USBImager is dependency-free
- Win32 Disk Imager requires Qt
- balenaEtcher has hundreds of megabytes (!) of dependency
- balenaEtcher and USBImager are available on all desktop platforms, Win32 Disk Imager is Windows only
- Of the three only USBImager is available for ARM and Raspbian