Extended-precision arithmetic


In ARM, instead of Saturating technique, they use Extended- precision Arithmetic. But in ADSP2181, they use Saturating Operations.

But, what are the advantages/dis-advantages w.r.t both these techniques and Which is the better technique ?

I searched the internet, but unable to get any link that talks about these designs in detail . Any ideas ?

Thx in advans, Karthik Balaguru

Reply to
Loading thread data ...

Are you sure? Last I looked, the ARMv5e DSP instructions included saturating arithmetic operations.

Neither are necessarily *better*: they do what they do. Which is more appropriate for the task at hand depends rather a lot on the nature of the task at hand. Can you explain what you're attempting to achieve?


Reply to
Andrew Reilly

Appologies. I should have put in this manner - 'ARM discourages saturating arithmetic operations'. ARM supports saturating arthimetic. Infact, there is the popluar 'Q' flag (Saturating Flag).

I find that ARM encourages Extended-Precision arithmetic and discourages the use of Saturating Arithmetic as they cost some additional cycles .

ADSP2181 is using Saturating Arithmetic.

Though i accept that both of those are good in their own territory.

I was just eager to know if we make those independent of processor, That is, Viewing those as just techniques, which one would be efficient ?

Karthik Balaguru

Reply to

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.