Libfixmath: Difference between revisions

Revision as of 19:39, 24 February 2011

Latest revision as of 10:11, 20 September 2024

(45 intermediate revisions by 20 users not shown)

Line 1:

{{Primary sources|date=January 2023}}

{{lowercase|title=libfixmath}}

{{Infobox software

| name = libfixmath

| developer = [[Ben Brewer]] (aka flatmush)

| latest_release_version = r7r64

| latest_release_date = {{Start date and age|20112012|02|2302}}

| programming language = [[C (programming language)|C99]]

| operating_system = [[Cross-platform software|Cross-platform]]

| genre = [[fixed-point arithmetic|fixed -point math library]]

| license = [[GNU Lesser General PublicMIT License|GNU LGPLMIT]]

| website = http://code.google.com/p/libfixmath

| status = active

}}

'''libfixmath''' is a [[platform independent|platform-independent]] [[fixed-point arithmetic|fixed point]] maths library aimed at developers wanting to perform fast non-integer maths on platforms lacking a [[Floating Point Unit|FPU]]. It offers developers a similar interface to the standard [[Math.h|math.h]] functions for use on [[Q (number format)|Q16.16]] [[fixed-point arithmetic|fixed point]] numbers.

'''libfixmath''' is a [[cross-platform software|platform-independent]] [[fixed-point arithmetic|fixed-point]] math library aimed at developers wanting to perform fast non-integer math on platforms lacking a (or with a low performance) [[Floating Point Unit|FPU]]. It offers developers a similar interface to the standard [[math.h]] functions for use on [[Q (number format)|Q16.16]] [[fixed-point arithmetic|fixed-point]] numbers. libfixmath has no external dependencies other than [[Stdint.h|stdint.h]] and a compiler which supports [[64-bit]] integer arithmetic (such as [[GNU Compiler Collection|GCC]]).<ref>{{cite web|url=httphttps://code.googlegithub.com/pPetteriAimonen/libfixmath|title=libfixmath Project Page|website=[[GitHub]] }}</ref>

WorkConditional iscompilation underwayoptions exist to remove the requirement for a [[64-bit]] capable [[compiler]] as many compilers for [[Microcontroller|microcontrollersmicrocontroller]]s and [[Digital signal processor|DSPs]] do not support [[64-bit]] arithmetic.<ref>{{cite web|url=httphttps://code.google.com/p/libfixmath/issues/detail?id=4|title=64-bit Compiler Support Issues}}</ref>

==History==

'''libfixmath''' was originally developed by [[Ben Brewer]] (aka flatmush) and first released publicly as part of the [[Dingoo A320|DingooSDK]] SDK.<ref>{{cite web|url=httphttps://code.google.com/p/dingoo-sdk|title=Dingoo SDK Project Page}}</ref> It has since been used to implement a (currentlysoftware closed source) 3d[[3D graphics]] library called FGL.<ref>{{cite web|url=httphttps://boardscode.dingoonitygoogle.orgcom/dingoo-development/platform-independant-dingoo-sdkp/fgl|title=DingooFGL SDKFlatmush/Fixed-Point Graphics ReleaseLibrary Project Page}}</ref>

==Q16.16 Functionsfunctions==

'''libfixmath''' was originally developed by [[Ben Brewer]] (aka flatmush) and first released publicly as part of the [[Dingoo A320|Dingoo]] SDK.<ref>{{cite web|url=http://code.google.com/p/dingoo-sdk|title=Dingoo SDK Project Page}}</ref> It has since been used to implement a (currently closed source) 3d graphics library called FGL.<ref>{{cite web|url=http://boards.dingoonity.org/dingoo-development/platform-independant-dingoo-sdk/|title=Dingoo SDK Release Page}}</ref>

==Q16.16 Functions==

{|class="wikitable"

|-

Line 35 ⟶ 33:

|-

|<code>fix16_cos</code> || [[cosine]]

|-

|<code>fix16_exp</code> || [[exponential function]]

|-

|<code>fix16_sin</code> || [[sine]]

Line 42:

|<code>fix16_tan</code> || [[tangent (trigonometric function)|tangent]]

|-

|<code>fix16_mul</code> || [[Multiplication|multiplication]]

|-

|<code>fix16_div</code> || [[Division (mathematics)|division]]

|-

|<code>fix16_sadd</code> || [[Saturation arithmetic|saturated]] [[Addition|addition]]

|-

|<code>fix16_smul</code> || [[Saturation arithmetic|saturated]] [[Multiplication|multiplication]]

|-

|<code>fix16_sdiv</code> || [[Saturation arithmetic|saturated]] [[Division (mathematics)|division]]

|}

==Other Functionsfunctions==

{|class="wikitable"

|-

Line 72 ⟶ 71:

|}

==LicensingPerformance==

For the most intensive function ([[atan2]]) benchmark results show the following results:

{|class="wikitable"

'''libfixmath''' is released under the [[GNU Lesser General Public License|GNU LGPL]], a [[permissive free software licence]], and is [[free software]].

|-

! Name !! Time Compared to Float

|-

|[[ARM Cortex-M0]] || 26.3%

|-

|[[Marvell Technology Group|Marvell]] [[XScale|PXA270]] ([[ARM architecture|ARM]]) @ 312&nbsp;MHz || 58.45%

|-

|[[List of Intel Core 2 microprocessors|Intel T5500]] || 120%

|-

|[[Intel atom|Intel Atom N280]] || 141%

|}

Note: These results were calculated using fixtest with caching optimizations turned off.<ref>{{cite web|url=https://code.google.com/p/libfixmath/downloads/detail?name=fixtest_r16.zip|title=fixtest Download Page}}</ref>

==External linksLicensing==

'''libfixmath''' is released under the [[GNU Lesser General PublicMIT License|GNU LGPL]], a [[permissive free software licence]], and is [[free software]].

* [http://code.google.com/p/libfixmath|title=libfixmath Project Page]

* [http://groups.google.com/group/libfixmath|title=libfixmath Group Page/Mailing List]

==See also==

{{Portal|Free Software}}

* [[Binary scaling]]

* [[Fixed-point arithmetic]]

* [[Floating-point arithmetic]]

* [[Q (number format)]]

==References==

{{Reflist}}

==External links==

* [https://github.com/PetteriAimonen/libfixmath Project Page]

* [http://codegroups.google.com/pgroup/libfixmath|title=libfixmath ProjectGroup Page/Mailing List]

[[Category:Numerical software]]

[[Category:C (programming language) libraries]]

[[Category:Free development toolkits andcomputer libraries]]

[[Category:Free software programmed in C]]