Is it possible to develop a Scroll view based on grlib?

jim wong said:

P.S. I am using TM4C123GH6PGE

 Did you ported GRlib or it is available to your platform?

Typically such scrolling demands that much (potential) screen data be, "at the ready" - awaiting user's gesture/command.  Most often this data is stored in RAM - for quick, easy access. 

Cortex M4 MCUs (here & elsewhere) usually fall far short both in RAM size and processing speed/power - and thus are not the (usual) suspects - revealed in cell/tablet, "teardowns."

This vendor has more advanced ARM MCUs - and also "Tablet App Notes" - which appear more suitable for your expressed objective. Cell phones & Tablets have "widened the gulf" between what can be done here - and where the "real" state of the art is practiced...

Outside your post's thrust - low-cost ARM-based Tablets are available - serious investigation on your part may uncover those most suitable for your, "poking/prodding" and save you the time/cost/effort of interconnecting display, MCU and memory.  (always a delight...)

cb1- said:

Cortex M4 MCUs (here & elsewhere) usually fall far short both in RAM size and processing speed/power - and thus are not the (usual) suspects - revealed in cell/tablet, "teardowns."

 CB1, I disagree this, TIVA has some part number dedicated to tablet market(LM4FS1 and TM4E), they are just not available to mass market and remain dedicated to large manufacturer. I don't know much more about them and compiler support is no more installed by default on latest compiler version.

Roberto - of course it's alright to disagree.  (no fun - if we all think/act the same - mais non?)

That said - I'd pile large stack of chips on "bigger/better" ARMs to be "resident" on most all "serious" Tablets.  Poster specifically targeted TM4C123 - which has vastly under-sized RAM - incapable of any (realistic) scrolling.  Thus - court must find in my favor - poster's MCU is far, "over-matched" for his intended application...

As you/I well know - mating the (proper) tool to task is "more" than 1/2 the battle!  Choosing a tool because it's convenient/known is unlikely to yield, "best/brightest" results - and sure to frustrate when "more" is demanded...

HI Roberto,

Yes, the GRlib is available on the platform. I can run the grlibdemo project on it. Thanks for your help.

Hi cb1,

Thank you for your help. I totally understand your point. However I have no control to the hardware, I am just a software guy.

What if the "scroll view" only contains few icons( in monochrome) or just few lines of text.(In order words, little/limited ram consumption) Then you think is there chance to make it work?

P.S. And if the MCU has 256KB ram (we can switch to TM4C129)

jim wong said:

P.S. I am using TM4C123GH6PGE

Hi Jim,

Aren't you confusing flash size (256KB) with ram size (32KB?) 

The techniques you offer (reducing the size & color depth) of the scrolled area surely will help - yet the issue remains - your task still "dwarfs" your (MCU) tool.  Ask your boss/firm leader this, "Should your product perform with markedly less capability & performance than a teenager's cell phone?"  Really?  Will that not be noted - and reflect negatively upon your entire product - no matter how excellent the other areas (may) be or perform?

Pushing a large/heavy stone up a steep incline can be done.  But should it be?  And once that effort is made - and found less than "Stellar" - then what?

Jim - you forced the clear contrast between a (relatively) weak MCU and far superior ones.  (via your desire to "gesture-scroll.")  Perhaps unwise.  Might you be able to better camouflage any such performance contrasts (not in your favor) - thus making your product (appear) more competitive?  If you're truly so "MCU restricted" - that would be my tack...

Nothing prevents your experimentation - the design & exercise of the "essence" of your program - and your independent test/verify to judge the MCU's (and your programming's) adequacy...

Hi cb1,

Sorry, misled about the ram a bit on the previous post. I was talking about if we switch to TM4C129 so that there will be 256KB ram. Switch to TM4C129 is the only option they offered. Again, thank you so much. 

Switch to a higher performing, better equipped MCU is what I've been advocating.  Whether that particular one will provide the necessary "boost" - you/I cannot say - at this point in time.

When our tech group embarks on product design/tweak we usually investigate the "best/brightest" competitive offerings.  We try to glean the, "What, where, how & why" of component & design choices - made by skilled others. With a reasonable survey - many key facts emerge - which best "educate, inform & guide" our group's design/development efforts.

As a closely related aside - our group designed a GUI using another's Cortex M4 - this one with external address/data bus & 180MHz system clock. (exceeds that of listed 129)  And - while this worked - teenager's cell phone (not the best one) clearly, "kicked our design's a**!"

Vendor here has far more advanced MCUs - and a neat application note - focused/guiding in Tablet design/development.

Perhaps a "middle ground" exists - multiple vendors have announced new Cortex M7 MCUs - appear to "fill the gap" between performance available from M4 and those far more advanced (and expensive) MCUs.  M7 is on our firm's radar - would be nice if (one day) such device would arrive this space...