On 4/13/2016 3:11 PM, Kustaa Nyholm wrote: >> >> It would be nice if you can confirm this, especially the actual >> 16-bit float format used. If the elements of the float are not >> byte-aligned, then did you pack and unpack the bit fields for each >> operation? What operations did you implement? > > Are you talking about my densitometer project? > > This was with 24 (8+16) floating points so no packing > was required. > > I was just putting my project forward as an example of > a kind project where floating point was a good choice > but where a 32 bit floating point was not usable (because > of the code space constraints). > > I implemented the four basic ops + log10. > > The source code is here, unfortunately the line breaks > seem to be broken but if you click on 'fp16math' link > in this page it should display ok: > > http://www.sparetimelabs.com/densitometer/sourcesframeset.html > > and it documents the basic math and format decently I think.
Thanks for the above. Actually I wrote that bit in reply to Vaclav's post, which purports to be an actual case of 16 bit float. It would be a good data point if it can be clarified... -- Cheers, Kein-Hong Man (esq.) Kuala Lumpur, Malaysia ------------------------------------------------------------------------------ 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
