replace two half bytes with one byte?

-=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- (c) WidthPadding Industries 1987 0\|42\|0 -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=- -=+=-
Socoder -> C/C++/C#/Other -> replace two half bytes with one byte?
Mon, 09 Jan 2017, 10:23
spinal	That title might not help but... I have 1 byte that I want to overlap 50% over two other bytes, leaving the remaining outer half bytes untouched... myByte = 11011011 byte1 = 10101010 byte2 = 01010101 what I want = ----1101 1011---- myByte 10101101,10110101 byte1,byte2 void drawTileOffset(int16_t x, int16_t y, const uint8_t bitmap) { for (byte t = 0; t < 8; t++) { if(y>1){ byte temp = sBuffer[(x 8) + ((y-1) * 128) + t]; byte myByte = pgm_read_byte(bitmap + t); sBuffer[(x * 8) + ((y-1) * 128) + t] = ((myByte&0x0F)8); } if(y<8){ byte temp = sBuffer[(x 8) + (y * 128) + t]; byte myByte = pgm_read_byte(bitmap + t); sBuffer[(x * 8) + (y * 128) + t] = ((myByte&0xF)/8); } } } Doesn't work right. --v Like always, I seem to be nrealy there, just not quite Any ideas? -=-=- Check out my excellent homepage!
Mon, 09 Jan 2017, 10:58
Jayenkai	Like so..? (Blitz2D/3D/Plus) Global Byte1,Byte2 Global myByte For n=0 To 50 myByte=Rand(0,255) Byte1=Rand(0,255) Byte2=Rand(0,255) DebugLog Right$(Bin(Byte1),8)+" - "+Right$(Bin(Byte2),8) DebugLog " "+Left$(Right$(Bin(MyByte),8),4)+" "+Right$(Bin(MyByte),4) MixyPops() DebugLog Right$(Bin(Byte1),8)+" - "+Right$(Bin(Byte2),8) DebugLog " " Next Function MixyPops() IntA=(myByte And $F0) Shr 4 ; "And" for Blitz2D/3D/Plus IntB=(myByte And $0F) Shl 4 ; -===- Byte1=Byte1 And $F0 ; -===- Byte2=Byte2 And $0F ; Replace And with & for 99% of languages! Byte1=Byte1+IntA Byte2=Byte2+IntB End Function --v -=-=- ''Load, Next List!''
Mon, 09 Jan 2017, 13:09
spinal	something like that, yup.
Tue, 04 Apr 2017, 09:46
spinal	hmmmm, need to update this to use a variable instead of the '4' for example, add the first X bytes to my other byte. So say I want to drop a 5 bit number starting at bit 2 of my 8bit number... ??? -=-=- Check out my excellent homepage!
Tue, 04 Apr 2017, 14:27
shroom_monk	Here's a simple C++ example that I think does what you want. \|edit\| (although on subsequent readings, I'm not quite sure what you're getting at with the second sentence; I think what I've done covers the first) \|edit\| It can be made more concise than this, of course, but I've spread it out to make it easier to follow. #include <cstdio> int main() { // 0) Inputs unsigned char myByte = 0xdb; // 11011011 unsigned char byte1 = 0xaa; // 10101010 unsigned char byte2 = 0x55; // 01010101 unsigned int offset = 5; // Must be in range 0-8 inclusive // 1) Make masks (examples are for offset = 5) unsigned char maskA = 0xff << (8 - offset); // 11111000 unsigned char maskB = 0xff >> offset; // 00000111 unsigned char maskC = 0xff << offset; // 11100000 unsigned char maskD = 0xff >> (8 - offset); // 00011111 // 2) Assemble result in easy to understand pieces // First <offset> bits of byte1 unsigned char resultByte1Left = byte1 & maskA; // First 8 - <offset> bits of myByte, shifted right by <offset> bits unsigned char resultByte1Right = (myByte & maskC) >> offset; // Last <offset> bits of myByte, shifted left by 8 - <offset> bits unsigned char resultByte2Left = (myByte & maskD) << (8 - offset); // Last 8 - <offset> bits of byte2 unsigned char resultByte2Right = byte2 & maskB; // 3) Put it all together unsigned char resultByte1 = resultByte1Left \| resultByte1Right; unsigned char resultByte2 = resultByte2Left \| resultByte2Right; // 4) Output printf("%#04x\n", maskA); printf("%#04x\n", maskB); printf("%#04x\n", maskC); printf("%#04x\n", maskD); printf("\n"); printf("%#04x\n", resultByte1); printf("%#04x\n", resultByte2); return 0; } --v -=-=- A mushroom a day keeps the doctor away... Keep It Simple, Shroom!
Fri, 07 Apr 2017, 02:58
spinal	Thanks for the help, but unfortunately, I changed my data size to 4 bits, so I''l be going with the first option again. Sorry shroom. -=-=- Check out my excellent homepage!