This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

CCS/TM4C1290NCPDT: Bigger Font more than 100 pt !!

Part Number: TM4C1290NCPDT

Tool/software: Code Composer Studio

Hey i want to have bigger fonts more than 100 pt !!

How can i do the same?

  • smit majithia said:
    Hey i want to have bigger fonts more than 100 pt !!

    Do note - 'Hey' is far from a respectful salutation - it proves far too forceful - almost commanding.      Thus not the ideal means of encouraging others to ASSIST you!    (in our case - you/I 'meet again.')

    To your issue - have you checked how MS Windows (in general) creates & manages their 'scalable fonts.'     Examining the 'successful' methods of skilled others - often saves hours (days in this case) of your time & effort.    There IS a caution though - (many/most WIN Fonts are copyright protected - you must proceed w/care.)     There do exist 'free fonts' - usually these are not as visually compelling - yet deserve your search.

    Should you find an acceptable font - check to insure that it may be scaled to 30 'points'  (which may translate to 30 'pixels' - should your objective be some 'Flat-Screen Display.'     (BTW - the display vehicle for your GIANT FONT - is silent!)

    Assuming that you can find an adequate font - and it natively may extend to 30 points (or beyond) you should be able to "Magnify it' by a factor of four.     Each pixel of the base font would be processed to produce a 16 pixel total (4x4 pixel cluster.)      In this manner a 30 pixel font would expand to (30x4 = 120 pixel height.)     There IS a downside - each of your new font's individual characters would have a 'squared-off' (i.e. 'jagged edge')     In the past - for far smaller fonts - firm/I would engage the local 'gifted high school' - to manually 'clean up' our 30-40 (maximum) height fonts.    

    A far better method would be to employ a powerful computer (ideally a PC) to 'smooth & remove' - any/all - rough edges of your new font.    It is those 'PC Processed Characters' then - which would make up your Giant Font.    (I believe the ARM M4 is 'Not up to' such a task - would spectacularly SLOW your Screen Writes.    A 220MHz Cortex M7 (may) work - yet such achievement is unknown - at this time.)

    One of the 'horrors' of this vendor's Graphic Library - is their 'placement of all Font Data w/in the MCU's internal memory space!'      No (real) Graphic Designer would have done so - as that causes:

    • each/every program load to require much more time
    • while 'eating memory' - based entirely upon the Size & Number of Such Fonts
    • further - Graphic Images - also are stored (only) w/in the MCU's memory - causing (even) added delay & loss of resource

    Past (real) Graphic Designers would insure that such Font & Graphic Image Data - was stored in Fast Access, yet EXTERNAL MEMORY - which avoids the 2 'killer defects'  identified above.   (killer in my opinion).    To 'head-off' the opposing, "External Memory would ADD Cost" - that external memory may be, 'Employed ONLY during Development - during the (easily) HUNDREDS of Program Loads" yielding (surely) HOURS - but more likely DAYS - of Saved Time!      And how COST-SAVING THAT!    (due to vastly faster Program Loads!)      Once the code is deemed 'approved/acceptable' - then it may be placed ENTIRELY w/in the MCU's memory store - saving the cost of the external memory - yet  STILL ENABLING - a VASTLY MORE EFFICIENT DEVELOPMENT VEHICLE!

    Our poster here is SO VAGUE in his requirement - that (OTHER) Advantaged Methods - which (may) work - can not properly be considered.     

    Removing that 'Hey' - while providing meaningful detail (I'll leave that determination to you) potentially enables (other) resourceful approaches - to be considered & (perhaps) presented...

  • Hello smit,

    Please see this post from Amit that explains what would need to be done: e2e.ti.com/.../1483728
  • One hopes that past (claimed to be needed) - will ... someway ... somehow  'FIT w/in MCU's Flash store.'     (while leaving adequate room - for User's code - and not looking too awkward (due to the jagged font edges - in that process!)  

    I've just read Amit's past post - it makes no note of the (potentially) enormous size - which such Giant Font - most always - demands/extracts!     In the development of such a program - many, many program downloads is,  "Par for the course."     And each one will take NOTABLY LONGER - due to that HUGE Graphic FONT - being repetitively sent - time & again!    That's NOT GOOD!

    May it be noted too - that 'SO MANY Here' - report on-going 'JTAG Issues.'     As each/every download grows SO EXPLOSIVELY - do we not  (GRAVELY) - invite the dreaded 'JTAG LOCK-OUT?    And as witnessed here - many times - recovery is NO SURE THING!     What then?

    Burdening the MCU's internal memory - and download mechanism - multiple times - with such EXTREME storage needs - proves NOT - the 'normal/customary' means of  'LONG DEEMED EFFECTIVE - Font & Image EXTERNAL storage!'     NOT INTERNAL!    (as the earlier post greatly detailed...)

  • cb1_mobile said:
    I've just read Amit's past post - it  makes no note of the (potentially) enormous size - which such Giant Font - most always - demands/extracts!

    And - while the sun just begins to set over the glorious Pacific - a similar 'sun'  has already set over my (very) failed belief (quoted, above!)       We were taught to 'Parade our Weaknesses' (so they could be best defeated) rather than 'hide them.'     I was SO WRONG!

    The ftrasterize program which Amit referenced - provides 'spectacular compression!'      Each/every one of the (more than ten)  'ftrasterized' Font files - my marine crüe and I examined today - were unusually compact!     Thus - the 'squawk' that - 'program loads' will 'Age mens' souls' - must find (another) target...

    With that 'mea culpa' complete - there 'are'  (some)  'ftrasterize' weaknesses revealed - and my group's seaside efforts today - have uncovered (hopefully) some productive findings:

    • Along w/creating seriously 'shrunken' Font Data Files - ftrasterize runs at lightning speed  (that was also, unexpected - and (may I say) 'LIKED!')     (that word - unlike its (departed) icon - has (yet) to be banned.    (maybe)
    • We scanned multiple, '48 point font files' - we found (none) which exceeded 7KB!    That is outstanding.    (I expected beyond 250KB...)
    • For completeness - we examined (relatively) small font files, too.     As expected - these received substantially less benefit from the compression algorithm(s).    (i.e. a 6x8 font consumed 960 bytes - that's (almost) no compression - as was predicted...)

    Good as ftrasterize appears - some weaknesses were found.    These include:

    • The GRLib User Guide promises a 'Pictorial Review' of (many) of  ftrasterize's 'creations.'      One of the LOUD 'Knocks' we had past heard about such compression programs - they produce fonts which are (pardon) 'ugly.'     We thus raced to Section 16 of our Graphic User Guide - and met 'Page after Page' of  BLANK  SPACE - NOT ONE demonstration of ftrasterize's output - revealed!    Mine is marked, 'Literature Number: SPMU300E,  SW-TM4C-GRL-UG-2.1.4.178  - April 2013 - Revised February 2017.   (that's believed to be 'reasonably' current - is it not?    Why are ALL of the rasterizations BLANK?    We provide multiple 'page captures' - to illustrate that EXPANSE of Blank space.
    • The originating poster 'specifically requested' Fonts beyond 100 points.    And today we learned - that ftrasterizer - 'tops out' at 100 points!    (will NOT run - beyond that 100 point limit!)
    • We learned as well - that even during ftrasterizer runs of < 100 points - certain 'wide' chars - may not be properly rendered.   (they are revealed as, 'requiring beyond 256 bytes' - w/out further explanation!)
    • While the 'ftrasterizer command' (teases) by allowing an (apparent) 100 point char height - in reality - about two-thirds of that - is the best that may be realized!   (a good 1/3 of that 100 point total - is devoted to 'lower-case' character 'descenders' - and 'inter-line' spacing.    While (not) rising to a breach in, 'Truth in Advertising' - what's suggested - is NOT Delivered!
    • There IS mention of a 'wrapper' - which (may) enable font storage (as someone here earlier proposed) - off chip.   (MCU)    Yet the Graphic User Guide notes that such is not provided - and appears reasonably demanding - as well.
    • The data we've extracted - and share w/the group here - reveals (most) of the font compressions to produce a 'Character Aspect Ratio' of  '6:7.'     Now having worked in the Display Biz - it is known that '5:7' proves most pleasing - and the output of 'ftrasterizer' appears to produce a far more 'blocky' (near square) character.    We all wanted to see that output - blank pages (alone) - filled our 'salt encrusted screen.'

    Follows now - support documentation - of these findings...

    Our search for 'Red October'  (ftrasterizer's output) revealed (consistently):

    and

    and these BLANK IMAGES continued - upon EVERY SINGLE - 'Font Illustrative' Page...        Did I say ...  that I (really) liked the 'font compression'?'       HOW (really) may we review the compressed font's appearance?

  • So is there any way out to have >100 pt Font to be displayed on my 3.5 inch screen?
  • Here are two (very) large fonts we (past) developed for our defense clients:    (before even 'LMI' (founders of Stellaris) was born!)

    This particular screen was used 'LIVE' w/in the desert - during 'action.'     The 'frit-seal' atop the Lcd was pierced - due to intense heat & severe shock/vibration - yet still the display - and our firm's electronics - carried on.    I believe this Graphic Controller was our first to use ARM - this one an AT91SAM7S - out prior to LMI's ARM Cortex M3.     (may they rest in peace - but for the 600 - we retain in stock...)

    This one 76 pixels tall:   Screen is QVGA  -  320x240 pixels  - 4.7" diagonal

    and this one is 112 pixels tall  (both screens are QVGA - 320x240 pixels - 4.7" diagonal

    Note that the larger (4.7" diagonal) screen produces a (naturally) larger character - independent of the point or pixel number.    The 112 pixel font was 'clipped' (you can see that at its top) the '0' was a truly - 'Auto-Scaled Font' - the '2' was bit-mapped.    Note the base-font - approximately 30 pixel rows beneath the 112 pixel '0.'

    So yes - there are several 'ways out' - and you (are)  welcome...

  • Yes "larger (4.7" diagonal) screen produces a (naturally) larger character - independent of the point or pixel number."

    now the things goes weird when i would say that with the high density display like 320x480 haveing the higher DPI then this would make we have to make the fonts bigger for such high density screen !

    Am i right ?
  • As the font's pixel inventory increases - while the screen stays reasonably (size) consistent - then Yes - higher DPI does lead to a 'larger than usual'  font requirement.   (to yield the targeted 'viewable font spec.)

    You (may) be able to 'help yourself.'      'Ftrasterizer.c'  appears - in all of its 'naked' Source Code glory - awaiting your 'skillful' examination.    

    As my report clearly revealed - the code 'cuts off @ 100 points' - yet that's NOT explained nor detailed.     Maybe - you can extend that upper limit - enabling you to produce a more visually notable font.

    That said - my report (also) detected that as the font 'Enlarges' - it  'may pierce the 256 byte limit'  (Who knew?   Again that's w/in my report).    

    Your review of the source file should enable you to determine - 'How or (even) If' - (both) the '100 point limit'  AND  the '256 byte limit' (may) - be overcome...

  • Then i should make my library for my self instead of using the Ftrasterizer.c or i should learn how the Ftrasterizer.c works and make it happen for >100 pt its a pity issue with Ftrasterizer.c or what we can say is limitation! can it be escalte to the maker of Ftrasterizer.c so that this issue would not be for your clients !
  • smit majithia said:
    can it be escalte to the maker of Ftrasterizer.c

    That's up to this vendor - (I have (only) profitable, 'User-Specifier' relationship w/this vendor.)    [don't know if you had that awareness]

    You must consider - in ALL of my time here (now > 10 years) [prior to being 'chased away' - by heavy handed 'LIKE-less e2e crüe.]    I've NOT encountered - such a 'Giant' (not just bigger - to be fair) Font request.

    As you grow/mature - you will learn that 'LARGE SALES' - and those 'design elements' - which (surely) yield such' - receive  EXTREME Vendor focus.     (as they should - both here - and at (other) semi-vendors!)

    In that 'inside light' - how do 'you' rate  the chance for such an,  FTRasterizer 'escalation?'

    And - is not your issue - 'predominantly brought about' - by your (perhaps unwise) choice of a, 'Too Pixel Rich' Display?      (some 'slight' note of 'thanks' - somehow has been missed ... yet remains awaited - and  warranted.)

  • Hello Smit,

    Unfortunately we will not be making changes to ftrasterizer to support larger fonts at this time.
  • May I note - my 'very special Decoder Ring' ... resolves vendor's, '(not) at this time' ... to 'never!'      ('LIKE's return - enjoys - better odds...)