From: Joerg Riechardt <J.Riechardt@gmx.de>
To: Mauro Carvalho Chehab <mchehab@osg.samsung.com>
Cc: linux-media <linux-media@vger.kernel.org>
Subject: Re: [PATCH] stv090x: use lookup tables for carrier/noise ratio
Date: Sat, 29 Aug 2015 17:02:57 +0200 [thread overview]
Message-ID: <55E1C9A1.6040702@gmx.de> (raw)
In-Reply-To: <20141111082838.05369439@recife.lan>
[-- Attachment #1: Type: text/plain, Size: 2909 bytes --]
Am 11.11.2014 um 11:28 schrieb Mauro Carvalho Chehab:
> Em Wed, 18 Sep 2013 01:05:20 +0200
> Joerg Riechardt <J.Riechardt@gmx.de> escreveu:
>
>> The stv090x driver uses the lookup table for signal strength already,
>> with this patch we use the lookup tables for carrier/noise ratio as well.
>> This has the advantage, that values for DVB-S and DVB-S2 are now
>> corresponding, while before they were way off. The values are now
>> proportional to real carrier/noise ratio, while before they were
>> corresponding to register values. So now applications are able to give
>> the user real carrier/noise ratio.
>>
>> Because the output has to be within 0x0000...0xFFFF the three negative
>> values for DVB-S2 are omitted. This is no significant loss, because
>> reception is lost at 7.5 dB already (TT S2-1600, Cine S2), so the
>> negative values are not really important, and also for DVB-S they don´t
>> exist.
>>
>> Signed-off-by: Joerg Riechardt <j.riechardt@gmx.de>
>>
>> Regards,
>> Joerg
>>
>> --- stv090x.c.bak 2013-09-06 20:59:01.132365872 +0200
>> +++ stv090x.c 2013-09-10 03:21:48.884115191 +0200
>> @@ -173,9 +173,9 @@
>>
>> /* DVBS2 C/N Lookup table */
>> static const struct stv090x_tab stv090x_s2cn_tab[] = {
>> - { -30, 13348 }, /* -3.0dB */
>> - { -20, 12640 }, /* -2d.0B */
>> - { -10, 11883 }, /* -1.0dB */
>> +// { -30, 13348 }, /* -3.0dB */
>> +// { -20, 12640 }, /* -2d.0B */
>> +// { -10, 11883 }, /* -1.0dB */
>> { 0, 11101 }, /* -0.0dB */
>> { 5, 10718 }, /* 0.5dB */
>> { 10, 10339 }, /* 1.0dB */
>
> Instead of commenting, just truncate the value at the DVBv3 stats
> function.
Ok, done.
Changed patch attached.
>
>> @@ -3697,9 +3697,10 @@
>> }
>> val /= 16;
>> last = ARRAY_SIZE(stv090x_s2cn_tab) - 1;
>> - div = stv090x_s2cn_tab[0].read -
>> - stv090x_s2cn_tab[last].read;
>> - *cnr = 0xFFFF - ((val * 0xFFFF) / div);
>> + div = stv090x_s2cn_tab[last].real -
>> + stv090x_s2cn_tab[0].real;
>> + val = stv090x_table_lookup(stv090x_s2cn_tab, last, val);
>> + *cnr = val * 0xFFFF / div;
>> }
>> break;
>>
>> @@ -3719,9 +3720,10 @@
>> }
>> val /= 16;
>> last = ARRAY_SIZE(stv090x_s1cn_tab) - 1;
>> - div = stv090x_s1cn_tab[0].read -
>> - stv090x_s1cn_tab[last].read;
>> - *cnr = 0xFFFF - ((val * 0xFFFF) / div);
>> + div = stv090x_s1cn_tab[last].real -
>> + stv090x_s1cn_tab[0].real;
>> + val = stv090x_table_lookup(stv090x_s1cn_tab, last, val);
>> + *cnr = val * 0xFFFF / div;
>> }
>
> As, with this patch, C/N will be a properly scaled value, the best
> is to add support for DVBv5 stats. With DVBv5 stats, the scale can
> be sent to userspace.
Sorry, I have no plans to add DVBv5 stats, although I agree, it would be
nice, if someone did.
Signed-off-by: Joerg Riechardt <j.riechardt@gmx.de>
Regards,
Joerg
>
>> break;
>> default:
>
> Regards,
> Mauro
>
[-- Attachment #2: stv090x-use-lookup-tables-for-carrier-noise-ratio.v2.patch --]
[-- Type: text/plain, Size: 966 bytes --]
diff -Nru A/stv090x.c B/stv090x.c
--- A/stv090x.c 2015-06-25 14:33:58.000000000 +0200
+++ B/stv090x.c 2015-08-29 16:07:44.889861230 +0200
@@ -3726,9 +3726,12 @@
}
val /= 16;
last = ARRAY_SIZE(stv090x_s2cn_tab) - 1;
- div = stv090x_s2cn_tab[0].read -
- stv090x_s2cn_tab[last].read;
- *cnr = 0xFFFF - ((val * 0xFFFF) / div);
+ div = stv090x_s2cn_tab[last].real -
+ stv090x_s2cn_tab[3].real;
+ val = stv090x_table_lookup(stv090x_s2cn_tab, last, val);
+ if (val < 0)
+ val = 0;
+ *cnr = val * 0xFFFF / div;
}
break;
@@ -3748,9 +3751,10 @@
}
val /= 16;
last = ARRAY_SIZE(stv090x_s1cn_tab) - 1;
- div = stv090x_s1cn_tab[0].read -
- stv090x_s1cn_tab[last].read;
- *cnr = 0xFFFF - ((val * 0xFFFF) / div);
+ div = stv090x_s1cn_tab[last].real -
+ stv090x_s1cn_tab[0].real;
+ val = stv090x_table_lookup(stv090x_s1cn_tab, last, val);
+ *cnr = val * 0xFFFF / div;
}
break;
default:
prev parent reply other threads:[~2015-08-29 15:03 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-17 23:05 [PATCH] stv090x: use lookup tables for carrier/noise ratio Joerg Riechardt
2014-11-11 10:28 ` Mauro Carvalho Chehab
2015-08-29 15:02 ` Joerg Riechardt [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=55E1C9A1.6040702@gmx.de \
--to=j.riechardt@gmx.de \
--cc=linux-media@vger.kernel.org \
--cc=mchehab@osg.samsung.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.