[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Discuss-gnuradio] nlog10: do we consider "normalizes log10(x <= 0)
From: |
Cinaed Simson |
Subject: |
Re: [Discuss-gnuradio] nlog10: do we consider "normalizes log10(x <= 0) to -18" as API? |
Date: |
Tue, 21 Nov 2017 00:50:48 -0800 |
User-agent: |
Mozilla/5.0 (X11; Linux i686; rv:52.0) Gecko/20100101 Thunderbird/52.4.0 |
On 11/20/2017 07:52 AM, Marcus Müller wrote:
> Hi Folks,
>
> I was about to commit a PR VOLKizing nlog10 [1], but it struck me that
> my processor (imho, correctly) says that log(0) == nan and log(-10) ==
> nan.
>
> Whereas the old nlog10 test case asserts that it's log_10(x<=0) == -18.
It would have to be complex
y=log10(-.00000000000000000001)
y=-18 + 1.3644i
For x real, then the limit as x approaches 0 from the positive side is
-infinity.
For x real from the negative side it's undefined since for any y, 10**y > 0.
-- Cinaed
>
> So, am I allowed to break that assertion for the greater mathematical
> appropriateness and speed?
>
> Best regards,
> Marcus
>
> [1]
> https://github.com/marcusmueller/gnuradio/tree/blocks_volkized_nlog10
>
>
>
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>