TweetFollow Us on Twitter

Oct 96 Challenge
Volume Number:12
Issue Number:10
Column Tag:Programmer’s Challenge

Programmer’s Challenge

By Bob Boonstra

Note: Source code files accompanying article are located on MacTech CD-ROM or source code disks.

DNA Match

This month’s Challenge is based on a suggestion submitted by Vicente Giles of the Universidad de Málaga. Vincente faces a real-world problem to look for all the genomic sequences that match certain

criteria, given a DNA database sequence and a problem sequence. A DNA sequence is a string of the four different nucleotides involved in the genetic code, denoted ‘A’, ‘C’, ‘G’, and ‘U’, which stand for adenine, cytosine, guanine, and uracil. The problem is to find all possible matches of the problem sequence in the database sequence, allowing a specified number of differences.

The prototype for the code you should write is:

long FindMatch(
 char *alphabet, /* legal characters for database and fragment strings */
 char *database, /* reference string to compare fragments against */
 void *storage,  /* storage preallocated for your use */
 char *fragment, /* string to match against database */
 long diffsAllowed,/* differences allowed between fragment and database */
 long matchPosition[]/* return match positions in this array*/
);

void InitMatch(
 char *alphabet, /* legal characters for database and fragment strings */
 char *database, /* reference string to compare fragments against */
 void *storage   /* storage preallocated for your use */
);

Because we would like our DNA-matching algorithm to be useful even if scientists discover an extraterrestrial genetic code based on other nucleotides, the algorithm accepts the genetic alphabet as a parameter. In the problem posed by Vincente, this would be the string “ACGU”, but in our Challenge it might include any of the characters ‘a’..’z’ or ‘A’..’Z’ (Extraterrestrial DNA is case sensitive). The null-terminated reference string contained in the database parameter can be up to 1000000000 (109) characters long. The fragment that you are to match is also null-terminated, but will be significantly shorter on average (up to 10000 characters) than the database string. You should compare the input fragment against database, finding all occurrences of fragment that differ in no more than diffsAllowed positions from a substring of database. Your code should populate one entry in the preallocated matchPosition array for each match found, storing the offset of the character in database that corresponds to the first character of fragment. The FindMatch function should return the number of matches found.

As an example, given the following input

alphabet: ACGU

database: ACGTACGTACGTAAAAAATACGTACGTATA

fragment: ACGTACGTAC

diffsAllowed: 5

your code should find 7 matches and store the following values in matchPosition:

-4 0 4 8 15 19 23


Notice that partial matches can occur at the beginning or the end of database, and as a result, the offsets returned in matchPosition can be negative or greater than strlen(database) - strlen(fragment).

To allow you to do some preprocessing, your InitMatch routine will be called once before a sequence of calls to FindMatch. InitMatch will be called with the same alphabet and database parameters provided to subsequent FindMatch calls. Both routines will also be given the same storage parameter that points to at least 1MB of memory allocated and initialized to zero by the calling routine. FindMatch will be called between 100 and 1000 times, on average, for each call to InitMatch. The winning solution will be the one with the fastest execution time, including the execution time for both InitMatch and FindMatch.

Other fine print: The alphabet characters will be provided in increasing ASCII order. The offsets you store in matchPosition need not be in any particular order. The value for diffsAllowed will typically be smaller than 50% of strlen(fragment). Finally, you should not allocate any dynamic storage in your solution beyond that provided in the storage parameter.

This will be a native PowerPC Challenge using the latest Symantec environment. Solutions may be coded in C or C++.

Two Months Ago Winner

Congratulations to Randy Boring for submitting the fastest entry to the A-Maze-ing Programmer’s Challenge. The Challenge this month was to write code that would find a path leading out of a three-dimensional maze. The solutions were provided with the maze size, an initial position, some storage for use in mapping the maze, and a callback routine. The callback provided the result of attempting to move in a given direction, indicating whether the attempt to move succeeded, failed because there was no opening in the specified direction, resulted in a fall down a shaft in the mine, or found an exit to the mine. Of the four entries submitted, only two successfully solved all of my test mazes; one of the entries crashed, and one went into an infinite loop.

The table below summarizes the results for each correct entry, including the language in which the solution was written, the size of the solution code, the amount of static data used by the solution, the total execution time for all test cases, and the number of moves needed to solve the mazes.

Name Language Code Size Data Size Time Moves

Randy Boring C++ 2792 484 343153 33519

Jay Negro C++ 1788 51 40945114 7120802

The test mazes used in the evaluation ranged in size from 10x20x30 to 100x100x200, and ranged in density (the percentage of open cells) from 10% to 20%. As indicated in the problem statement, a path to an exit was guaranteed to exist from any cell reachable from the starting position.

Randy’s winning entry spent more time processing each move than the second place entry from first-time Challenge contestant Jay Negro, but Randy’s code solved the maze using significantly fewer moves and executed two orders of magnitude more quickly. His code maintains a queue of what are believed to be the best moves to try. As long as there are moves in the best move list, it invokes the callback with the best move, checks for success, and then updates the map with what it has learned about the maze position it just tried. The CalcBestMove routine determines the best possible move (surprise!) by first moving toward an adjacent maze boundary if one exists, or moving toward the nearest maze boundary if nothing is known about the current position, or trying adjacent positions about which nothing is known, or finally by moving toward a position about which nothing is known. The CalcBestMove heuristics, along with judicious use of inline functions and some optimization of maze offset calculations, combined to make this an efficient solution.

Careful readers of the code will note one potential problem with the Initialize routine, in that it simply gives up and returns if it is unable to allocate enough memory. This could have caused the winning entry to fail for larger mazes when given only the amount of memory guaranteed by the problem. However, the size of the mazes that I could practically evaluate was limited by the speed of the other entries, and the memory problem did not show up with those cases, so I elected to ignore it. Under other circumstances, proper handling of low memory conditions would have been required to win.

TOP 20 CONTESTANTS

Here are the Top 20 Contestants for the Programmer’s Challenge. The numbers below include points awarded over the 24 most recent contests, including points earned by this month’s entrants.

RankNamePoints
1.Munter, Ernst193
2.Gregg, Xan92
3.Larsson, Gustav87
4.[Name deleted]60
5.Lengyel, Eric40
6.Lewis, Peter30
7.Boring, Randy27
8.Beith, Gary24
9.Kasparian, Raffi22
10.Vineyard, Jeremy22
11.Cutts, Kevin21
12.Picao, Miguel Cruz21
13.Brown, Jorg20
14.Gundrum, Eric20
15.Karsh, Bill19
16.Stenger, Allen19
17.Cooper, Greg17
18.Mallett, Jeff17
19.Nevard, John17
20.Nicolle, Ludovic14

There are three ways to earn points: (1) scoring in the top 5 of any Challenge, (2) being the first person to find a bug in a published winning solution or, (3) being the first person to suggest a Challenge that I use. The points you can win are:

1st place20 points
2nd place10 points
3rd place7 points
4th place4 points
5th place2 points
finding bug2 points
suggesting Challenge2 points

Here is Randy’s winning solution:

Amazing.cp

Copyright © 1996 Randy Boring

typedef Boolean (*MoveProc) (
 long xMove,long yMove,long zMove,
 long *newXPos,long *newYPos,long *newZPos
 );

// the MoveProc, MakeAMove, is a callback.  It returns true if you 
// have found your way out of the maze.
// You give it (x,y,z) as a delta from your current position,
// each from [-1, 0, 1].  Straight up and straight down (and all zeroes)
// always result in no movement. 

Boolean Maze (long xMove, // these are your initial position
 long yMove,
 long zMove,
 long xSize,// these are the dimensions of the maze
 long ySize,
 long zSize,
 MoveProc MakeAMove, // this is your callback routine
 char *mapStorage// this is your preallocated storage
 );// (one char per position in maze)

Typedefs and Constants
typedef long dirT; // direction enumerator (0-24)

static
const long dir2dx[25]={9,
 -1, 0, 1,  -1, 1,  -1, 0, 1,
 -1, 0, 1,  -1, 1,  -1, 0, 1,
 -1, 0, 1,  -1, 1,  -1, 0, 1,};
static
const long dir2dy[25]={9, 
 1, 1, 1,   0, 0,  -1,-1,-1,
  1, 1, 1,   0, 0,  -1,-1,-1,
  1, 1, 1,   0, 0,  -1,-1,-1,};
static
const long dir2dz[25]={9,
  1, 1, 1,   1, 1,   1, 1, 1,
  0, 0, 0,   0, 0,   0, 0, 0,
 -1,-1,-1,  -1,-1,  -1,-1,-1,};
static
const dirT di[3][3][3] = {
  {{1,2,3},{4,0,5},{6,7,8}},
  {{9,10,11},{12,0,13},{14,15,16}},
  {{17,18,19},{20,0,21},{22,23,24}}};
static const dirT kNoDir = 0;
static const dirT kFirstDir = 1;
static const dirT kLastDir = 24;
static const dirT kNumDirs = 25; // including kNoDir at zero
static const char kUnknown = 0;  // unknown square
    // every type below is known
static const char kWall = 1;// wall
static const char kSpace = 2; // space-above-wall
static const char kFall = 4;// space-above-space
static const char kTriedSpace = 10;// searched space
static const char kTriedFall = 12; // searched fall

static const short kTakeProb = 4;  // 1/4th
static const long kSTNBlockSize = 48;

typedef long squareIndexT;
typedef struct STN {
 struct STN *parent;
 squareIndexT square;// square index of this node
 dirT   direction; // how I got here from my parent
 } SearchTreeNode, *STNPtr, **STNHandle;

Globals
static STNPtr gTreeRoot;
static STNPtr gTreeTop;
static STNPtr gQHead;
static STNPtr gQTail;
static STNPtr gBestMoveList;
static STNHandle gTreeRootH;
static STNHandle gBestMoveListH;
static STNPtr gBestMoveListNextPos;

Defines 
#define myIsUnknown(sq) (kUnknown == (sq))
#define myIsKnown(sq)(kUnknown != (sq))
    // the below should only be used when the square is known
#define myIsWall(sq) (kWall == (sq))
#define myIsUntriedWalkable(sq)  (kSpace == (sq))
#define myIsOpen(sq) (0x00 == (0x01 & (sq)))
#define myIsWalkable(sq) (0x02 == (0x02 & (sq)))
#define myIsUntriedFall(sq) (kFall == (sq))
#define myIsFall(sq) (0x04 == (0x04 & (sq)))
#define myIsTried(sq)(0x08 == (0x08 & (sq)))
#define d2x(d) (dir2dx[d])
#define d2y(d) (dir2dy[d])
#define d2z(d) (dir2dz[d])
#define xvec(d) (d2x(d))
#define yvec(d,xN) (d2y(d) * (xN))
#define zvec(d,xyN) (d2z(d) * (xyN))
#define offsetD(d,xN,xyN) (xvec(d) + yvec(d,xN) + zvec(d,xyN))
#define offsetXYZ(x,y,z,xN,xyN) ((x) + (y) * xN + (z) * (xyN))
#define map(m,x,y,z,xN,xyN) (*(m + (x) + (y) * xN + (z) * (xyN)))

#ifdef powerc
#define BreakToSourceDebugger_()   Debugger()
#else   // 68K
#define BreakToSourceDebugger_()   SysBreak()
#endif  // powerc


isEmptySearchQ
static inline Boolean
isEmptySearchQ() {return (gQHead == gQTail);}

isFullSearchQ
static inline Boolean
isFullSearchQ() {return (gTreeTop <= gQTail);}

DeQ
static inline STNPtr
DeQ(void) {return gQHead++;}

EnQ
static inline STNPtr
EnQ(void) {
 STNPtr p = gQTail++;
    //if (isFullSearchQ())
    //    BreakToSourceDebugger_();
 return (p);
}

EnQNewCandidate
// Add this move (direction to a square index) to the tree at the current node, assumes 
// the square index has not already been visited by this search.
static void
EnQNewCandidate(STNPtr parent, const long sqi, 
 const dirT d)
{
STNPtr newNode = EnQ();
newNode->parent = parent;
newNode->direction = d;
newNode->square = sqi;
}

isEmptySearchTree
static inline Boolean
isEmptySearchTree() {return (gTreeRoot == gQTail);}

PopSearchTree
static inline long
PopSearchTree(void) {return ((-gQTail)->square);}

NewSearchTree
static void
NewSearchTree(void) {
 gQTail = gQHead = gTreeRoot;
 gQTail++;// the only time we’re called,
}//  gTreeRoot is immediately the EnQed elem.

DisposeSearchTree
static void
DisposeSearchTree(char *m) {
while (!isEmptySearchTree()) {
    // remove ‘tried’ mark
 *(m + PopSearchTree()) = kSpace;
 }
}

isEmptyMoveList
static inline Boolean
isEmptyMoveList()
 {return (gBestMoveListNextPos == gBestMoveList);}

PushMoveList
static inline void
PushMoveList(const dirT d) {
 gBestMoveListNextPos->direction = d;
 gBestMoveListNextPos++;}

PopMoveList
static inline dirT
PopMoveList(void) {
 -gBestMoveListNextPos;
 return (gBestMoveListNextPos->direction);}

NewBestMoveList
static void
NewBestMoveList(void) {gBestMoveListNextPos = gBestMoveList;}

SetBestMove
// Copy the sequence of best directions-to-move to the
// best move list.  (The tree is about to be freed.)
static void
SetBestMove(STNPtr node, const dirT d)
{
PushMoveList(d);
while (kNoDir != node->direction) { // the root’s dir
 PushMoveList(node->direction);
 node = node->parent;
 }
}

Initialize
// Initialize everything for the routine
// map is already initialized to zeroes (kUnknown == 0)
static Boolean
Initialize(const long x, const long y, const long z,
 const long xSize, const long ySize, const long zSize,
 char *m)
{
const long xySize = xSize * ySize;
long treeBytesWanted = sizeof(SearchTreeNode)
 * ((xSize - 2) * (ySize - 2) * (zSize - 2) + 4);
long treeBytesNeeded = sizeof(SearchTreeNode)
 * ((xSize - 2) + (ySize - 2) + (zSize - 2) + 1);
Boolean succeed = true;

map(m,x,y,z,xSize,xySize) = kSpace;
map(m,x,y,z-1,xSize,xySize) = kWall;

gBestMoveListH = (STNHandle) NewHandle(sizeof(SearchTreeNode)
 * (zSize + xySize) * 2);
if (!gBestMoveListH) succeed = false;
 // unable to initialize successfully,MEM
HLock((Handle) gBestMoveListH);
gBestMoveList = *gBestMoveListH;

do {
 gTreeRootH = (STNHandle) NewHandle(treeBytesWanted);
 treeBytesWanted *= 0.9;
 }
 while (!gTreeRootH && 
 (treeBytesNeeded <= treeBytesWanted));
if (!gTreeRootH) succeed = false;
 // unable to initialize successfully,MEM
HLock((Handle) gTreeRootH);
gTreeRoot = *gTreeRootH;

NewBestMoveList();
return succeed;
}


DeInitialize
// Free everything we allocated
static void
DeInitialize()
{
HUnlock((Handle) gBestMoveListH);
DisposeHandle((Handle) gBestMoveListH);
HUnlock((Handle) gTreeRootH);
DisposeHandle((Handle) gTreeRootH);
}


AtEdge
// Are we at an edge?
// return true if we are
static Boolean
AtEdge(const long x, const long y, const long z,
 const long xMax, const long yMax, const long zMax)
{
if (0==x || 0==y || 0==z) return true;
if (xMax==x || yMax==y || zMax==z) return true;
return false;
}


MapMove
// Map the move we just made/tried.
// I.e., store our new knowledge of the maze in the map
static void
MapMove(const long oldx, const long oldy, const long oldz,
 const long newx, const long newy, const long newz,
 char *m, const long xN, const long xyN,
 const long dx, const long dy, const long dz)
{
long x = oldx + dx; // where we tried to move
long y = oldy + dy; // (we need these in either case below)
long z = oldz + dz;
long sqi = offsetXYZ(x, y, z, xN, xyN);
if (newx==oldx && newy==oldy && newz==oldz) { // bump!
    // mark as wall the square we bumped into
 *(m + sqi) = kWall;
    // lastTried was not successful
    //gLastWasSuccess = false;
 }
else {
    // mark as clear any squares we fell through
    // actually mark them with as fall because they can’t
    // really be moved ‘to’ only through, downwards.
 if (newz < z) {
 do {
 *(m + sqi) = kFall;
 z-;
 sqi -= xyN;
 } while (newz < z);
 }
    // mark as clear our current space
 *(m + sqi) = kSpace;
    // mark as ‘wall’ the square we are standing on
 sqi -= xyN;
 *(m + sqi) = kWall;
    // lastTried was successful
    //gLastWasSuccess = true;
    // direction of last success
    //gLastSuccess = lastTried;
 }
}


APossibleExit
// return a direction in which
// there may be an adjacent exit.
// A possible exit must be both an edge and untried.
static dirT
APossibleExit(const long x, const long y, const long z,
 const long xSize, const long ySize, const long zSize,
 char *m, dirT lastTried)
{
dirT tryD = lastTried + 1; // start here
/* check for edge conditions first */
if (x > 1 && x < xSize - 2)
 if (y > 1 && y < ySize - 2)
 if (z > 1 && z < zSize - 2)
    /* no exits are possibly nearby */
    /* because no edges are nearby */
 return kNoDir; // zero

/* search the adjacent spaces for a possible exit */
/* this could improve a lot */
while (tryD <= kLastDir) {
 const long dx = d2x(tryD);
 const long dy = d2y(tryD);
 const long dz = d2z(tryD);
 if (AtEdge(x+dx, y+dy, z+dz, xSize-1, ySize-1, zSize-1))
 if (myIsUnknown(map(m, x + dx, y + dy, z + dz,
 xSize, xSize * ySize)))
 return tryD;
 tryD++;
 }
tryD = kFirstDir;
while (tryD <= lastTried) {
 const long dx = d2x(tryD);
 const long dy = d2y(tryD);
 const long dz = d2z(tryD);
 if (AtEdge(x+dx, y+dy, z+dz, xSize-1, ySize-1, zSize-1))
 if (myIsUnknown(map(m, x + dx, y + dy, z + dz,
 xSize, xSize * ySize)))
 return tryD;
 tryD++;
 }
return kNoDir; // zero, no possible exit found nearby
}


TotallyUnknown
// returns true if every move leads to an unknown square
// this is only possible at the beginning and after falling
// at least three below our last position.
static Boolean
TotallyUnknown(const long x, const long y, const long z,
 const long xSize, const long xySize, char *m)
{
dirT dir;
/* loop through the directions until we find a known spot */
for (dir = kFirstDir; dir <= kLastDir; dir++) {
 const long dx = d2x(dir);
 const long dy = d2y(dir);
 const long dz = d2z(dir);
 if (!myIsUnknown(map(m, x + dx, y + dy, z + dz,
 xSize, xySize)))
 return false;
 }
return true;
}


MoveTowardsNearWall
// Pick a move that is towards a near wall
// (NOTE: only used when all directions are unknown)
static dirT
MoveTowardsNearWall(const long x, const long y, const long z,
 const long xSize, const long ySize, const long zSize,
 char *m)
{
#pragma unused (m)
long distx = xSize - x - 1; // distance from x==xMax edge
long disty = ySize - y - 1; // distance from y==yMax edge
long distz = zSize - z - 1; // distance from z==zMax edge
// [xyz] is distance from [xyz]==0 edge
long dx, dy, dz;
long xy = xSize * ySize;

if (x < distx) dx = -1;
else if (x == distx) dx = 0;
else dx = 1;
if (y < disty) dy = -1;
else if (y == disty) dy = 0;
else dy = 1;
if (z < distz) dz = -1;
else if (z == distz) dz = 0;
else dz = 1;

return di[dy+1][dz+1][dx+1];
}


Gravity
// Fall until we find a floor below us
// Only use after mapping new knowledge
// (Don’t use during mapping)
static long
Gravity(long i, const char *m, const long xySize)
{
do {i -= xySize;}
 while (kWall != *(m + i));
 
return i + xySize;
}


SearchOneSquare
// Search from this square for an adjacent unknown square
// queueing up moveable squares (including spaces below
// falls) for further research later,
// marking enqueued squares as ‘tried’ (immediately meaning
// ‘not-to-be-queued-for-trying’, later ‘actually-tried’)
// NOTE: while I could mark falls as tried (upward from
// every space) they are rather unlikely to be in the
// search path, so it’s not worth it.  Just re-enact
// gravity each time, and check last square for ‘tried’.
static Boolean
SearchOneSquare(STNPtr startSTN,
 const long xSize, const long xySize, char *m)
{
dirT d;
const long startSquare = startSTN->square;

for (d = kFirstDir; d <= kLastDir; d++) {
 const long sqi = startSquare + offsetD(d,xSize,xySize);
 const char sq = *(m + sqi);
 if (myIsUnknown(sq)) {
 SetBestMove(startSTN, d);
 return true;
 }
 else if (myIsUntriedWalkable(sq)) {
 EnQNewCandidate(startSTN,sqi,d);
 *(m + sqi) = kTriedSpace;
 }
 else if (myIsUntriedFall(sq)) {
 const long bottomSqi = Gravity(sqi,m,xySize);
 const long bottomSq = *(m + bottomSqi);
 if (myIsUntriedWalkable(sq)) {
 EnQNewCandidate(startSTN,bottomSqi,d);
 *(m + bottomSqi) = kTriedSpace;
 }
    //else if (myIsUnknown(bottomSq)) {
    //    BreakToSourceDebugger_(); // should be impossible
    //    return true; //??
    //    }
 else ; //it’s an already tried space, do nothing
 }
 else ; // it’s a wall or already tried space, do nothing
 }
return false;
}


FindNearestUnknown
// Find a move sequence that will lead to an unknown
// square (preferably an edge square?).
static dirT
FindNearestUnknown(const long x, const long y, const long z,
 const long xSize, const long ySize, const long zSize,
 char *m)
{
#pragma unused (zSize)
const long xySize = xSize * ySize;
Boolean found = false;
// make new search tree with square at x,y,z
NewSearchTree();
NewBestMoveList();
gTreeRoot->parent = nil;
gTreeRoot->square = offsetXYZ(x,y,z,xSize,xySize);
gTreeRoot->direction = kNoDir;
*(m + gTreeRoot->square) = kTriedSpace;
// fan out from this one layer at a time
while (!isEmptySearchQ() && !found) {
 STNPtr tryNode = DeQ();
 found = SearchOneSquare(tryNode, xSize, xySize, m);
    //FreeSTN(tryNode);
 }
DisposeSearchTree(m);
// found move(list) or failed
return found;
}


CalcBestMove
// Calculate the best move to try
// return true if we have found an exit or are at wit’s end
static dirT
CalcBestMove(const long x, const long y, const long z,
 const long xSize, const long ySize, const long zSize,
 char *m)
{
static dirT lastTried = kNoDir;
dirT d;

if (!isEmptyMoveList()) { // we have a pre-made list of moves

 

Community Search:
MacTech Search:

Software Updates via MacUpdate

CrossOver 14.1.3 - Run Windows apps on y...
CrossOver can get your Windows productivity applications and PC games up and running on your Mac quickly and easily. CrossOver runs the Windows software that you need on Mac at home, in the office,... Read more
Little Snitch 3.5.3 - Alerts you about o...
Little Snitch gives you control over your private outgoing data. Track background activity As soon as your computer connects to the Internet, applications often have permission to send any... Read more
OmniGraffle Pro 6.2.3 - Create diagrams,...
OmniGraffle Pro helps you draw beautiful diagrams, family trees, flow charts, org charts, layouts, and (mathematically speaking) any other directed or non-directed graphs. We've had people use... Read more
OmniFocus 2.2 - GTD task manager with iO...
OmniFocus helps you manage your tasks the way that you want, freeing you to focus your attention on the things that matter to you most. Capturing tasks and ideas is always a keyboard shortcut away in... Read more
Cocktail 8.4 - General maintenance and o...
Cocktail is a general purpose utility for OS X that lets you clean, repair and optimize your Mac. It is a powerful digital toolset that helps hundreds of thousands of Mac users around the world get... Read more
PDFKey Pro 4.3 - Edit and print password...
PDFKey Pro can unlock PDF documents protected for printing and copying when you've forgotten your password. It can now also protect your PDF files with a password to prevent unauthorized access and/... Read more
Kodi 15.0.beta1 - Powerful media center...
Kodi (was XBMC) is an award-winning free and open-source (GPL) software media player and entertainment hub that can be installed on Linux, OS X, Windows, iOS, and Android, featuring a 10-foot user... Read more
DiskCatalogMaker 6.4.12 - Catalog your d...
DiskCatalogMaker is a simple disk management tool which catalogs disks. Simple, light-weight, and fast. Finder-like intuitive look and feel. Super-fast search algorithm. Can compress catalog data... Read more
Macs Fan Control 1.3.0.0 - Monitor and c...
Macs Fan Control allows you to monitor and control almost any aspect of your computer's fans, with support for controlling fan speed, temperature sensors pane, menu-bar icon, and autostart with... Read more
Lyn 1.5.11 - Lightweight image browser a...
Lyn is a lightweight and fast image browser and viewer designed for photographers, graphic artists and Web designers. Featuring an extremely versatile and aesthetically pleasing interface, it... Read more

Battle of Gods: Ascension (Games)
Battle of Gods: Ascension 1.0 Device: iOS Universal Category: Games Price: $2.99, Version: 1.0 (iTunes) Description: TURN-BASED TACTICAL COMBATFight tactical battles against the forces of Hades! In Battle of Gods: Ascension you play... | Read more »
Shadowmatic's Latest Update Adds a...
Shadowmatic's shadowy shadow-ness is getting a little shadowy-er thanks to a recent update that adds an Arcade Mode. [Read more] | Read more »
Sunrise Calendar and Slack Have Assimila...
Wunderlist is perhaps one of the most populat and beloved productivity apps on the App Store - and now it's gone and incorporated itself into other useful services like Sunrise Calendar and Slack. [Read more] | Read more »
Crossy Road Devs Hipster Whale are Bring...
Hipster Whale, the minds behind the rather popular (and rather great) Crossy Road, have teamed-up with Bandai Namco to create PAC-MAN 256: an absolutely bonkers looking maze runner chaser thing. | Read more »
Meet the New Spotify Music
Spotify Music  has a lot going on. They're introducing 3 new modes to serve all your musical needs, with the "Now" start page  gives you curated playlists based on your particular tastes. As you listen the app will learn more about your tastes and... | Read more »
What the Apple Watch Gets Right, and Wha...
| Read more »
Celebrate PAC-MAN's 35th Birthday W...
BANDAI NAMCO Entertainment America is celebrating PAC-MAN's 35th anniversary by releasing updates for PAC-MAN and PAC-MAN Lite for iOS. [Read more] | Read more »
Strike Wing Episode 2 has Landed on the...
Strike Wing: Raptor Rising is an exciting space combat simulator by Crescent Moon Games, which was recently updated to continue the story with Episode 2. [Read more] | Read more »
Kiqplan Expands its Interactive Coaching...
The makers of Fitbug have been hard at work on their Kiqplan lineup, and have added four new summer themed plans to help you get the most out of your workout. [Read more] | Read more »
Make a Photobook in Minutes with Pictyea...
What happens when you can't stop taking photos and have an urge to create a photobook? Pictyear saves the day. [Read more] | Read more »

Price Scanner via MacPrices.net

What Would the ideal Apple Productivity Platf...
For the past four years I’ve kept a foot in both the Mac and iPad camps respectively. my daily computing hours divided about 50/50 between the two devices with remarkable consistency. However, there’... Read more
PageMeUp 1.2.1 Ten Dollar Page Layout Applica...
Paris, France-based Softobe, an OS X software development company, has announced that their PageMeUp v. 1.2.1, is available on the Mac App Store for $9.99. The license can be installed on up to 5... Read more
Eight New Products For USB Type-C Application...
Fresco Logic, specialists in advanced connectivity technologies and ICs, has introduced two new product families targeting the Type-C connector recently introduced across a number of consumer... Read more
Scripps National Spelling Bee Launches Buzzwo...
Scripps National Spelling Bee fans can monitor the action at the 2015 Spelling Bee with the new Buzzworthy app for iOS, Android and Windows mobile devices. The free Buzzworthy app provides friendly... Read more
13-inch 2.5GHz MacBook Pro on sale for $120 o...
B&H Photo has the 13″ 2.5GHz MacBook Pro on sale for $979 including free shipping plus NY sales tax only. Their price is $120 off MSRP, and it’s the lowest price for this model (except for Apple’... Read more
27-inch 3.3GHz 5K iMac on sale for $1899, $10...
B&H Photo has the new 27″ 3.3GHz 5K iMac on sale for $1899.99 including free shipping plus NY tax only. Their price is $100 off MSRP. Read more
Save up to $50 on iPad Air 2, NY tax only, fr...
B&H Photo has iPad Air 2s on sale for up to $50 off MSRP including free shipping plus NY sales tax only: - 16GB iPad Air 2 WiFi: $469 $30 off - 64GB iPad Air 2 WiFi: $549.99 $50 off - 128GB iPad... Read more
Updated Mac Price Trackers
We’ve updated our Mac Price Trackers with the latest information on prices, bundles, and availability on systems from Apple’s authorized internet/catalog resellers: - 15″ MacBook Pros - 13″ MacBook... Read more
New 13-inch 2.9GHz Retina MacBook Pro on sale...
B&H Photo has the 13″ 2.9GHz/512GB Retina MacBook Pro on sale for $1699.99 including free shipping plus NY tax only. Their price is $100 off MSRP, and it’s the lowest price for this model from... Read more
Apple refurbished 2014 MacBook Airs available...
The Apple Store has Apple Certified Refurbished 2014 MacBook Airs available starting at $679. An Apple one-year warranty is included with each MacBook, and shipping is free: - 11″ 1.4GHz/128GB... Read more

Jobs Board

Senior Software Engineer - *Apple* SIM - Ap...
Changing the world is all in a day039s work at Apple . If you love innovation, here039s your chance to make a career of it. You039ll work hard. But the job comes with Read more
Lead *Apple* Solutions Consultant - Retail...
**Job Summary** Job Summary The Lead ASC is an Apple employee who serves as the Apple business manager and influencer in a hyper-business critical Reseller's store Read more
Architect / Senior Software Engineer, *Apple...
Changing the world is all in a day039s work at Apple . If you love innovation, here039s your chance to make a career of it. You039ll work hard. But the job comes with Read more
*Apple* Pay Support Readiness Project Manage...
Changing the world is all in a day039s work at Apple . If you love innovation, here039s your chance to make a career of it. You039ll work hard. But the job comes with Read more
Hardware Design Validation Engineer - *Apple...
**Job Summary** The Apple Watch team is looking for a Hardware Design Validation Engineer. This person will be part of the Apple Watch hardware team with Read more
All contents are Copyright 1984-2011 by Xplain Corporation. All rights reserved. Theme designed by Icreon.