Inaccurate SD Card maximum file writing times

[gerrievanzyl]

In the OpenBCI GUI you can select the SD file sized based on the expected duration of the trail.

It would appear that these sizes are based off a 8 channel data writing rates at 250Hz. With 16 channels at 1kHz a 1-Hour file is filled in 7.5 minutes.

We need to update the file size based on the sample rate and the board type settings.

I can tackle this if required.

A simple way to implement this would be in the code that allocates the SD space by simply multiplying the number of blocks defined (see below) by 2 for 16-Channel board and 2 or 4 for 500 or 1000Hz sample rate.

define BLOCK_5MIN 11000

define BLOCK_15MIN 33000

define BLOCK_30MIN 66000

define BLOCK_1HR 131000

define BLOCK_2HR 261000

define BLOCK_4HR 521000

define BLOCK_12HR 1561000

define BLOCK_24HR 3122000

[andrewjaykeller]

we should do this programatically but i 100% agree this needs to be done!

[agharibans]

I tried to double the BLOCK_24HR to get 24hr 16ch recordings, but then OpenBCI often fails to initialize through the GUI. Is it necessary to pre-allocate space? It would be ideal if it can just keep writing until the memory card is full.

[gerrievanzyl]

@aghariba it looks to me as if the limitation is on the SD library the code uses. I do not have any background to this, but I know that it is a good strategy in cases like these where we have something like a microcontroller trying to run a file operating system. I see there are newer SD arduino libraries available, that do not have this limitation, but I do not know how well they perform. For now, I propose that we make the small changes and @aj-ptw can evaluate whether it is worth rewriting this functionality.

I will give a shot at making some changes and submitting a pull request

[andrewjaykeller]

@gerrievanzyl I think this is a great idea!!

[gerrievanzyl]

@aghariba were you able to make the changes?

[agharibans]

I haven't tried to do this yet.

[retiutut]

closed by #102