• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: How to implement float min(float x, ...) ?
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: How to implement float min(float x, ...) ?


  • Subject: Re: How to implement float min(float x, ...) ?
  • From: Kirk Kerekes <email@hidden>
  • Date: Mon, 17 Nov 2008 12:21:49 -0600

How to implement float min(float x, ...) ?


I suggest you not use var_arg functions if you can avoid them.

So I propose a different path.

As a jumping-off point, here is a category method on NSData that assumes that "self" (the NSData instance) is an array of floats:

(The -floats category method merely casts NSData's -bytes to return a pointer-to-float, and the -floatlength category method divides -length by sizeof(float).)

Note that this makes the NSData the memory equivalent of a C array/ vector, so you can now pass it to the Accelerate framework.

Any setup overhead is likely compensated for by the hardware accel of vDSP (when available).

- (float) floatMinValueAndIndex:(int *) outIndex
{
float minValue = 0;
vDSP_minvi((void *)[self floats], 1, &minValue, outIndex, [self floatLength]);
return minValue;
}


The following NSData class category method will convert an arbitrary float array to the NSData required for the above:

+ (NSData *) floatsWithBlock: (float *) block count: (unsigned) count // count of floats
{
NSData * result = [[self class] dataWithBytes: block length: (count * sizeof(float))];
return result;
}


- (const float *) floats
{
	return (const float *) [self bytes];
}

- (unsigned) count
{
	return [self floatLength];
}

- (unsigned) floatLength
{
	return [self length]/sizeof(float);
}


If performance is less of an issue than clarity, I would convert the floats to an NSArray of NSNumbers, and then just sort the array using the -sortedArrayUsing... methods. You get more information back -- not just min, but max and next-to-min, and next-to-max, etc.



_______________________________________________

Cocoa-dev mailing list (email@hidden)

Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com

Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden


  • Prev by Date: CoreImage filters question
  • Next by Date: Re: Set user agent for NSURL
  • Previous by thread: Re: How to implement float min(float x, ...) ?
  • Next by thread: Grow Bug ?
  • Index(es):
    • Date
    • Thread