> >How about this: Given a choice between a 24-bit float format >that's made simple for 8-bit MCUs (e.g. add back the first missing >bit-1), and 16-bit standards-based half float (though the main >users are GPUs), which one would you prefer to use?
That is a though one! In general I tend to always prefer things that follow standards (in this case 16 bit floats). But this has very limited precision and dynamic range so all things considerer I think 24 bits would preferable as it fits (many) 8 bit architectures so much better. In larger scope the priority should be to write the 32 bit float libraries in assembler. The real pain with 32 bits floats is the size of the FB libraries because they are in C. I all my imaginary use scenarios I don't see the need to save a byte or two in in storage space or speed but I do see problem with code size. So was not so tough after all! Just my 2 snt worth. br Kusti This e-mail may contain confidential or privileged information. If you are not the intended recipient (or have received this e-mail in error) please notify the sender immediately and destroy this e-mail. Any unauthorized copying, disclosure or distribution of the material in this e-mail is strictly forbidden. We will not be liable for direct, indirect, special or consequential damages arising from alteration of the contents of this message by a third party or as a result of any virus being passed on or as of transmission of this e-mail in general. ------------------------------------------------------------------------------ Find and fix application performance issues faster with Applications Manager Applications Manager provides deep performance insights into multiple tiers of your business applications. It resolves application problems quickly and reduces your MTTR. Get your free trial! https://ad.doubleclick.net/ddm/clk/302982198;130105516;z _______________________________________________ Sdcc-user mailing list Sdcc-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/sdcc-user
