how to speed up download

Tip / Sign in to post questions, reply, level up, and achieve exciting badges. Know more

cross mob
Anonymous
Not applicable
Downloading app takes 30-60 seconds at 115200 baud.  Why so slow?

Is debug info (symbols, etc.) include, which can be disabled?

Is the compilation code-space optimized?

Thanks.
0 Likes
1 Solution
MichaelF_56
Moderator
Moderator
Moderator
250 sign-ins 25 comments on blog 10 comments on blog
As a rule of thumb today, we have found that a 15K application typically takes about 12s to erase the EEPROM and download then verify that it was downloaded successfully. Roughly 80% of this time scales linearly with app size while the rest of the time is used for preparing the chip for download.

Note that we have found a method to improve the download speed and we are working on a solution which will be implemented in the next revision of the SDK (hopefully before the end of the year).

Once this new method is in place, a 28K download will probably take around 12-15 seconds.

View solution in original post

0 Likes
6 Replies
MichaelF_56
Moderator
Moderator
Moderator
250 sign-ins 25 comments on blog 10 comments on blog
Please describe your development environment.

My point of reference is the WICED SDK installed on my Windows 7 PC with a BCM920732 Tag board connected via USB (over UART) and when I compile and load a project such as "ROM.heart_rate_monitor-BCM920732TAG_Q32 download", which is described in the WICED Smart Quick Start Guide available here on the website, and I thought the process took 6 or so seconds.

Are you using the WICED SDK with the same dev board, or have you incorporated a header on your own custom board which provides direct access to the HCI UART?

What is the status of bleapp_trace_enable while you are programming the part?

Note that HCI based downloads and debugging are mutually exclusive of one another.
0 Likes
Anonymous
Not applicable
I can attest to this as well. Using the WICED IDE with the BCM920732 Tag board via USB, applications are taking 20-30 seconds to download onto the device. Its not a big deal, but if it can take 5 seconds instead, I would like that.

How would I check the status of bleapp_trace_enable? If it means anything, to download onto the board, I am double clicking the Make Target menu on the right-hand side of the IDE. Also, the sample programs Ive been playing around with are RAM hello_sensor as well as automation_io.

Thanks!
0 Likes
MichaelF_56
Moderator
Moderator
Moderator
250 sign-ins 25 comments on blog 10 comments on blog
My apologies. Im not sure what I was thinking when I quoted 6 secs for full compile and download.

I just ran ROM Heart Rate Monitor and RAM Hello Sensor and both took a little over 30 secs to load.

Regarding the location of bleapp_trace_enable, in the directory tree of the WICED Smart IDE on the left hand side, find the WICED-Smart-SDKWiced-Smartsparcommonspar_init.c file. Within it you can modify the SPAR_APP_SETUP to set bleapp_trace_enable.
0 Likes
Anonymous
Not applicable
Thank you.
0 Likes
Anonymous
Not applicable
Does Broadcom plan to speed up the download.

Its actually almost  up to 60 seconds once you approach 28k of object code.

Also, if the available RAM is increased, and we add more code, the slow download will become significantly distracting to the productivity of our developers.

Thanks for great product otherwise.
0 Likes
MichaelF_56
Moderator
Moderator
Moderator
250 sign-ins 25 comments on blog 10 comments on blog
As a rule of thumb today, we have found that a 15K application typically takes about 12s to erase the EEPROM and download then verify that it was downloaded successfully. Roughly 80% of this time scales linearly with app size while the rest of the time is used for preparing the chip for download.

Note that we have found a method to improve the download speed and we are working on a solution which will be implemented in the next revision of the SDK (hopefully before the end of the year).

Once this new method is in place, a 28K download will probably take around 12-15 seconds.
0 Likes