DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=yahoo.co.uk;
  h=X-YMail-OSG:Received:X-Rocket-MIMEInfo:X-Mailer:References:Message-ID:Date:From:Reply-To:Subject:To:In-Reply-To:MIME-Version:Content-Type;
  b=6C809hPHrQEgE8Jq+B0IkU4cdHPomluR4NKlgfNvYJcV3fD8Aeaw+uvBMGS/VwYRzG5KJQ8yDEPM4dAOhJZDezY3Uosbkm/CLHqpA74S9GjS8zJfDHu4vkilUnKas2331o3XJHucxIzX08AF4BQVeekrkybL/J320fJayg++ims=;
References: <871180718.807959.1357202851097.JavaMail.open-xchange@email.1und1.de> <8CFB7BD49F0E58C-6E8-60E@webmail-d164.sysops.aol.com> <4D45055C381D4BECA4B1BEF1254E1059@Black> <1751FF6A14644CC7A968E35A554BF7C4@Black> <EDBC773861614A24BAB2B1E77439684C@Black>
Message-ID: <1358002148.84705.YahooMailNeo@web133206.mail.ir2.yahoo.com>
Date: Sat, 12 Jan 2013 14:49:08 +0000 (GMT)
From: M0FMT <m0fmt@yahoo.co.uk>
To: "rsgb_lf_group@blacksheep.org" <rsgb_lf_group@blacksheep.org>
In-Reply-To: <EDBC773861614A24BAB2B1E77439684C@Black>
MIME-Version: 1.0
Subject: Re: LF: Slow Voice - testing on 137.0 kHz
Content-Type: multipart/alternative; boundary="131296447-2112117833-1358002148=:84705"
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


--131296447-2112117833-1358002148=:84705
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable

Well done Marcus and co.=0A=C2=A0=0AThank you for your innovative contribut=
ions to Ham Radio. Proving that even with current technology the limits can=
 be pushed further.=0A=C2=A0=0AIn the UK (and I am sure elsewhere) there is=
 a single channel VHF=C2=A0Parrot type repeater which is operated in a simi=
lar way where you speak, drop carrier, hear your own=C2=A0transmission rebr=
oadcast and then wait for a reply which some times can be heard directly on=
 the channel and is then re-enforced by the Parrot in the following transmi=
ssion. It is slow and laborious but many stations use it.=0A=C2=A0=0AI gues=
s the difference is that you have an added element of delay=C2=A0each way f=
or audio frequency com/decompression which amounts to the Parrot in the abo=
ve case rebroadcasting each transmission only longer by a factor..=0A=C2=A0=
=0ALike to hear an audo file?=0A=0A=0A73 es GL Pete M0FMT IO91UX=0A=0A=0A>_=
_______________________________=0A> From: Markus Vester <markusvester@aol.c=
om>=0A>To: rsgb_lf_group@blacksheep.org =0A>Sent: Saturday, 12 January 2013=
, 14:03=0A>Subject: LF: Slow Voice - testing on 137.0 kHz=0A>  =0A>=0A>=EF=
=BB=BF  =0A>Yesterday morning I have =0Auploaded a new version SlowVoice v0=
.2 tohttp://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip =0A>I have rework=
ed some details of the timing, and =0Aadded simplified scripts for receive-=
only and transmit-only in mono mode. The =0Aserial port PTT control utility=
=C2=A0has also been=C2=A0modified =0Ato=C2=A0function=C2=A0under WinXP.=0A>=
=C2=A0 =0A>This afternoon=C2=A0I=C2=A0am sending=C2=A0a few slow voice test=
s with 3 =0Atimes deceleration in 1-minute frames. The=C2=A0800 Hz wide pas=
sband=C2=A0is =0Acentered on 137.0 kHz, thus 136.6 to 137.4 kHz=C2=A0corres=
ponds to 0.3 to 2.7 =0AkHz audio.=C2=A0Attached screenshot=C2=A0shows=C2=A0=
the narrow SSB signal on the =0ATwente WebSDR tuned=C2=A0to 135.5 kHz USB.=
=C2=A0I used a narrow passband=C2=A0to =0Aimprove=C2=A0rejection for DCF an=
d HGA. This audio=C2=A0around fc=3D1500 =0AHz=C2=A0was fed=C2=A0to the rece=
ive-only utility, which then produced=C2=A0clear =0Aspeech=C2=A0during 20 s=
econds of each=C2=A0minute.=C2=A0 =0A>=0A>I intend to=C2=A0run some more te=
st=C2=A0transmissions this afternoon, =0Astarting now 14 UT. Any=C2=A0repor=
ts=C2=A0would be=C2=A0much appriciated! =0A>=0A>Best 73, =0A>Markus (DF6NM)=
=C2=A0=C2=A0 =0A>=0A>=C2=A0 =0A>=C2=A0 =0A>----- Original Message ----- =0A=
>From: Markus Vester  =0A>To: rsgb_lf_group@blacksheep.org  =0A>Sent: Tuesd=
ay, January 08, 2013 11:31 PM =0A>Subject: LF: Slow Voice utility =0A>=0A> =
=0A>A=C2=A0few minutes ago=C2=A0I have uploaded an =0Aexperimental=C2=A0Slo=
wVoice script to =0A>http://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip =
=0A>=C2=A0 =0A>The main purpose is to be able to convey=C2=A0high =0Aqualit=
y speech=C2=A0in less than 800 Hz on the RF side, as prescribed by German =
=0ALF and MF regulations.=C2=A0By default,=C2=A0an audio band of 300 to 270=
0 Hz is =0Aslowed by a factor of 3 and upconverted to 1500 Hz center freque=
ncy. Thus=C2=A0a =0Arange of 1200 to 2000 Hz=C2=A0is fed to the SSB=C2=A0tr=
ansmitter. As an aside =0Abenefit we get a very much welcome=C2=A05 dB impr=
ovement in SNR. =0A>=C2=A0 =0A>Like JT9-1, the procedure is clock-synchroni=
zed in =0Afixed one-minute time slots. At the beginning of the minute, the =
transmitting =0Astation is asked to speak for 20 seconds.=C2=A0It feels a b=
it like=C2=A0to =0Aspeaking to a telephone answering machine ;-) The=C2=A0s=
low converted output of =0Athe growing audio buffer begins to play immediat=
ely, and will go on=C2=A0during =0Athe whole minute. On the receive side, f=
ast replay begins at second number 40, =0Aand=C2=A0the shrinking buffer wil=
l finally be used up by the end of the =0Aminute. =0A>=C2=A0 =0A>The SlowVo=
ice.bat script is set up for full duplex =0Aoperation. One channel is suppo=
sed to handle the transmit side (eg. mic =0Aconnected to=C2=A0left input,=
=C2=A0transmitter to left output), while the other =0Achannel has the recei=
ver=C2=A0fed to=C2=A0the input and phones on the output. =0AThis is symmetr=
ical, and there is no further need to switch audio pathways =0Abetween TX a=
nd RX. It is also possible to monitor=C2=A0the result of one's own =0Atrans=
mission, eg. by=C2=A0physically connecting left line out to right in. If =
=0Ayou want you can use=C2=A0SpecLab=C2=A0in conjunction with=C2=A0Virtual =
Audio =0ACable to preprocess both the=C2=A0mic and receiver audio inputs. I=
n this case =0AVAC will have to be selected as the standard audio input dev=
ice before the =0Arunning the script. =0A>=C2=A0 =0A>Whether a=C2=A0slots i=
s used for=C2=A0TX or=C2=A0RX =0Ais not controlled by the software but only=
 by the user's PTT. I have included =0Atwo extra scripts which are supposed=
 to activate the serial=C2=A0handshake lines =0A(RTS and DTR) on either=C2=
=A0even or odd minutes. Unfortunately the=C2=A0port =0Acontrol does not see=
m to work on my=C2=A0XP machine with an external USB to =0Aserial converter=
, so for the time being you=C2=A0may have to do=C2=A0with =0Aeither=C2=A0vo=
x control or manual=C2=A0switching. =0A>=C2=A0 =0A>I haven't gotten around =
writing=C2=A0more detailed =0Ainstructions, but some basic hints have been =
included in the SlowVoice script. Similar to the first slow WSPR experiment=
s, the =0Awhole deceleration / acceleration process is based on=C2=A0Wolf's=
 sound =0Autilities=C2=A0which you will need to download from=C2=A0the DL4Y=
HF=C2=A0web =0Asite.=C2=A0You will have to run the included patch =0Autilit=
y once to duplicate SndInput and =0ASndOutpt=C2=A0instances. =0A>=C2=A0 =0A=
>Have fun! And please report bugs and other =0Atroubles. =0A>=C2=A0 =0A>73,=
 Markus (DF6NM) =0A>=C2=A0 =0A>=C2=A0 =0A>=C2=A0 =0A>=C2=A0 =0A>----- Origi=
nal Message ----- =0A>From: Markus Vester  =0A>To: rsgb_lf_group@blacksheep=
.org  =0A>Sent: Sunday, January 06, 2013 1:50 PM =0A>Subject: LF: Bandlimit=
ed SSB test on 136 kHz =0A>=0A> =0A>... =0A>=C2=A0 =0A>Slow voice transmiss=
ion=C2=A0(ie the=C2=A0audio =0Adeceleration/acceleration method originally =
used by DK8KW and myself) would =0Abe=C2=A0nicer as it can fit=C2=A0a full =
SSB channel into 800 =0AHz.=C2=A0I=C2=A0have been working on a=C2=A0semi-au=
tomatic control, with=C2=A0a =0Afixed one minute raster similar to JT9-1. T=
his will hopefully allow us to =0Acomfortably exchange one 20 second voice =
message per time slot =0A(speak=C2=A0during seconds 0 to 20, concurrent tra=
nsmit=C2=A0and=C2=A0receive at =0A1/3 speed from 0 to 60, replay=C2=A0start=
ing 40 to 60). Anyone =0Ainterested? =0A>=C2=A0 =0A>... =0A>=0A>=C2=A0carr =
  
--131296447-2112117833-1358002148=:84705
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable

<html><body><div style=3D"color:#000; background-color:#fff; font-family:ar=
ial, helvetica, sans-serif;font-size:14pt"><div><span>Well done Marcus and =
co.</span></div><div><span></span>&nbsp;</div><div><span>Thank you for your=
 <span id=3D"misspell-0"><span>innovative</span></span> contributions to Ha=
m Radio. Proving that even with current <span id=3D"misspell-1"><span>techn=
ology</span></span> the limits can be pushed further.</span></div><div><spa=
n></span>&nbsp;</div><div><span>In the UK (and I am sure elsewhere) there i=
s a single channel VHF&nbsp;Parrot type repeater which is operated in a sim=
ilar way where you speak, drop carrier, hear your own&nbsp;transmission reb=
roadcast and then wait for a reply which some times can be heard directly o=
n the channel and is then re-enforced by the Parrot in the following transm=
ission. It is slow and laborious but many stations use it.</span></div><div=
><span></span>&nbsp;</div><div><span>I guess the difference is that you
 have an added element of delay&nbsp;each way for audio frequency com/decom=
pression which amounts to the Parrot in the above case <var id=3D"yui-ie-cu=
rsor"></var>rebroadcasting each transmission only longer by a factor..</spa=
n></div><div><span></span>&nbsp;</div><div><span>Like to hear an audo file?=
</span><br><br></div><div>73 es GL Pete M0FMT IO91UX<br> <blockquote style=
=3D"padding-left: 5px; margin-top: 5px; margin-left: 5px; border-left-color=
: rgb(16, 16, 255); border-left-width: 2px; border-left-style: solid;">  <d=
iv style=3D"font-family: arial, helvetica, sans-serif; font-size: 14pt;"> <=
div style=3D"font-family: times new roman, new york, times, serif; font-siz=
e: 12pt;"> <div dir=3D"ltr"> <font size=3D"2" face=3D"Arial"> <div style=3D=
"margin: 5px 0px; padding: 0px; border: 1px solid rgb(204, 204, 204); heigh=
t: 0px; line-height: 0; font-size: 0px;" class=3D"hr" contentEditable=3D"fa=
lse" readonly=3D"true"></div>  <b><span style=3D"font-weight: bold;">From:<=
/span></b> Markus
 Vester &lt;markusvester@aol.com&gt;<br> <b><span style=3D"font-weight: bol=
d;">To:</span></b> rsgb_lf_group@blacksheep.org <br> <b><span style=3D"font=
-weight: bold;">Sent:</span></b> Saturday, 12 January 2013, 14:03<br> <b><s=
pan style=3D"font-weight: bold;">Subject:</span></b> LF: Slow Voice - testi=
ng on 137.0 kHz<br> </font> </div> <br><meta content=3D"off" http-equiv=3D"=
x-dns-prefetch-control"><div id=3D"yiv617873430">=EF=BB=BF=0A=0A =0A =0A<st=
yle></style>=0A=0A<div>=0A<div><font size=3D"2" face=3D"Arial"><font size=
=3D"2" face=3D"Arial">Yesterday morning I have =0Auploaded a new version Sl=
owVoice v0.2 to</font></font></div><font size=3D"2" face=3D"Arial">=0A<div>=
<font size=3D"2" face=3D"Arial">=0A<div><font size=3D"2" face=3D"Arial"><a =
href=3D"http://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip" rel=3D"nofoll=
ow" target=3D"_blank">http://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip<=
/a></font></div></font></div>=0A<div><font size=3D"2" face=3D"Arial">I have=
 reworked some details of the timing, and =0Aadded simplified scripts for r=
eceive-only and transmit-only in mono mode. The =0Aserial port PTT control =
utility&nbsp;has also been&nbsp;modified =0Ato&nbsp;function&nbsp;under Win=
XP.</font></div><font size=3D"2" face=3D"Arial">=0A<div></div></font><div><=
font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div>This afternoon&nb=
sp;I&nbsp;am sending&nbsp;a few slow voice tests with 3 =0Atimes decelerati=
on in 1-minute frames. The&nbsp;800 Hz wide passband&nbsp;is =0Acentered on=
 137.0 kHz, thus 136.6 to 137.4 kHz&nbsp;corresponds to 0.3 to 2.7 =0AkHz a=
udio.&nbsp;Attached screenshot&nbsp;shows&nbsp;the narrow SSB signal on the=
 =0ATwente WebSDR tuned&nbsp;to 135.5 kHz USB.&nbsp;I used a narrow passban=
d&nbsp;to =0Aimprove&nbsp;rejection for DCF and HGA. This audio&nbsp;around=
 fc=3D1500 =0AHz&nbsp;was fed&nbsp;to the receive-only utility, which then =
produced&nbsp;clear =0Aspeech&nbsp;during 20 seconds of each&nbsp;minute.&n=
bsp;</div>=0A<div>&nbsp;</div>=0A<div>I intend to&nbsp;run some more test&n=
bsp;transmissions this afternoon, =0Astarting now 14 UT. Any&nbsp;reports&n=
bsp;would be&nbsp;much appriciated!</div>=0A<div>&nbsp;</div>=0A<div>Best 7=
3,</div>=0A<div>Markus (DF6NM)&nbsp;&nbsp;</div>=0A<div></div></font><div>&=
nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<d=
iv><font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div style=3D"font=
: 10pt/normal arial; font-size-adjust: none; font-stretch: normal;">----- O=
riginal Message ----- =0A<div style=3D"background: rgb(228, 228, 228);"><b>=
From:</b> <a title=3D"markusvester@aol.com" href=3D"mailto:markusvester@aol=
.com" rel=3D"nofollow" target=3D"_blank" ymailto=3D"mailto:markusvester@aol=
.com">Markus Vester</a> =0A</div>=0A<div><b>To:</b> <a title=3D"rsgb_lf_gro=
up@blacksheep.org" href=3D"mailto:rsgb_lf_group@blacksheep.org" rel=3D"nofo=
llow" target=3D"_blank" ymailto=3D"mailto:rsgb_lf_group@blacksheep.org">rsg=
b_lf_group@blacksheep.org</a> =0A</div>=0A<div><b>Sent:</b> Tuesday, Januar=
y 08, 2013 11:31 PM</div>=0A<div><b>Subject:</b> LF: Slow Voice utility</di=
v></div>=0A<div><font size=3D"2" face=3D"Arial"></font><font size=3D"2" fac=
e=3D"Arial"></font><font size=3D"2" face=3D"Arial"></font><font size=3D"2" =
face=3D"Arial"></font><br></div>=0A<div><font size=3D"2" face=3D"Arial">A&n=
bsp;few minutes ago&nbsp;I have uploaded an =0Aexperimental&nbsp;SlowVoice =
script to</font></div>=0A<div><font size=3D"2" face=3D"Arial"><a href=3D"ht=
tp://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip" rel=3D"nofollow" target=
=3D"_blank">http://www.df6nm.bplaced.net/slowvoice/SlowVoice.zip</a></font>=
</div>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div><f=
ont size=3D"2" face=3D"Arial">The main purpose is to be able to convey&nbsp=
;high =0Aquality speech&nbsp;in less than 800 Hz on the RF side, as prescri=
bed by German =0ALF and MF regulations.&nbsp;By default,&nbsp;an audio band=
 of 300 to 2700 Hz is =0Aslowed by a factor of 3 and upconverted to 1500 Hz=
 center frequency. Thus&nbsp;a =0Arange of 1200 to 2000 Hz&nbsp;is fed to t=
he SSB&nbsp;transmitter. As an aside =0Abenefit we get a very much welcome&=
nbsp;5 dB improvement in SNR.</font></div>=0A<div><font size=3D"2" face=3D"=
Arial"></font>&nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial">Like JT9-=
1, the procedure is clock-synchronized in =0Afixed one-minute time slots. A=
t the beginning of the minute, the transmitting =0Astation is asked to spea=
k for 20 seconds.&nbsp;It feels a bit like&nbsp;to =0Aspeaking to a telepho=
ne answering machine ;-) The&nbsp;slow converted output of =0Athe growing a=
udio buffer begins to play immediately, and will go on&nbsp;during =0Athe w=
hole minute. On the receive side, fast replay begins at second number 40, =
=0Aand&nbsp;the shrinking buffer will finally be used up by the end of the =
=0Aminute.</font></div>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp=
;</div>=0A<div><font size=3D"2" face=3D"Arial">The SlowVoice.bat script is =
set up for full duplex =0Aoperation. One channel is supposed to handle the =
transmit side (eg. mic =0Aconnected to&nbsp;left input,&nbsp;transmitter to=
 left output), while the other =0Achannel has the receiver&nbsp;fed to&nbsp=
;the input and phones on the output. =0AThis is symmetrical, and there is n=
o further need to switch audio pathways =0Abetween TX and RX. It is also po=
ssible to monitor&nbsp;the result of one's own =0Atransmission, eg. by&nbsp=
;physically connecting left line out to right in. If =0Ayou want you can us=
e&nbsp;SpecLab&nbsp;in conjunction with&nbsp;Virtual Audio =0ACable to prep=
rocess both the&nbsp;mic and receiver audio inputs. In this case =0AVAC wil=
l have to be selected as the standard audio input device before the =0Arunn=
ing the script.</font></div>=0A<div><font size=3D"2" face=3D"Arial"></font>=
&nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial">Whether a&nbsp;slots is=
 used for&nbsp;TX or&nbsp;RX =0Ais not controlled by the software but only =
by the user's PTT. I have included =0Atwo extra scripts which are supposed =
to activate the serial&nbsp;handshake lines =0A(RTS and DTR) on either&nbsp=
;even or odd minutes. Unfortunately the&nbsp;port =0Acontrol does not seem =
to work on my&nbsp;XP machine with an external USB to =0Aserial converter, =
so for the time being you&nbsp;may have to do&nbsp;with =0Aeither&nbsp;vox =
control or manual&nbsp;switching.</font></div>=0A<div><font size=3D"2" face=
=3D"Arial"></font>&nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial">=0A<d=
iv><font size=3D"2" face=3D"Arial">I haven't gotten around writing&nbsp;mor=
e detailed =0Ainstructions, but some basic hints have been included in the =
SlowVoice script. =0A</font><font size=3D"2" face=3D"Arial">Similar to the =
first slow WSPR experiments, the =0Awhole deceleration / acceleration proce=
ss is based on&nbsp;Wolf's sound =0Autilities&nbsp;which you will need to d=
ownload from&nbsp;the DL4YHF&nbsp;web =0Asite.&nbsp;Y<font size=3D"2" face=
=3D"Arial">ou will have to run the included patch =0Autility once to duplic=
ate SndInput and =0ASndOutpt&nbsp;instances.</font></font></div></font></di=
v>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div><font =
size=3D"2" face=3D"Arial">Have fun! And please report bugs and other =0Atro=
ubles.</font></div>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</d=
iv>=0A<div><font size=3D"2" face=3D"Arial">73, Markus (DF6NM)</font></div>=
=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div><font si=
ze=3D"2" face=3D"Arial"></font>&nbsp;</div>=0A<div><font size=3D"2" face=3D=
"Arial"></font>&nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial"></font>&=
nbsp;</div>=0A<div style=3D"font: 10pt/normal arial; font-size-adjust: none=
; font-stretch: normal;">----- Original Message ----- =0A<div style=3D"back=
ground: rgb(228, 228, 228);"><b>From:</b> <a title=3D"markusvester@aol.com"=
 href=3D"mailto:markusvester@aol.com" rel=3D"nofollow" target=3D"_blank" ym=
ailto=3D"mailto:markusvester@aol.com">Markus Vester</a> =0A</div>=0A<div><b=
>To:</b> <a title=3D"rsgb_lf_group@blacksheep.org" href=3D"mailto:rsgb_lf_g=
roup@blacksheep.org" rel=3D"nofollow" target=3D"_blank" ymailto=3D"mailto:r=
sgb_lf_group@blacksheep.org">rsgb_lf_group@blacksheep.org</a> =0A</div>=0A<=
div><b>Sent:</b> Sunday, January 06, 2013 1:50 PM</div>=0A<div><b>Subject:<=
/b> LF: Bandlimited SSB test on 136 kHz</div></div>=0A<div><font size=3D"2"=
 face=3D"Arial"></font><font size=3D"2" face=3D"Arial"></font><font size=3D=
"2" face=3D"Arial"></font><br></div>=0A<div><font size=3D"2" face=3D"Arial"=
>...</font></div>=0A<div><font size=3D"2" face=3D"Arial"></font>&nbsp;</div=
>=0A<div><font size=3D"2" face=3D"Arial">Slow voice transmission&nbsp;(ie t=
he&nbsp;audio =0Adeceleration/acceleration method originally used by DK8KW =
and myself) would =0Abe&nbsp;nicer as it can fit&nbsp;a full SSB channel in=
to 800 =0AHz.&nbsp;I&nbsp;have been working on a&nbsp;semi-automatic contro=
l, with&nbsp;a =0Afixed one minute raster similar to JT9-1. This will hopef=
ully allow us to =0Acomfortably exchange one 20 second voice message per ti=
me slot =0A(speak&nbsp;during seconds 0 to 20, concurrent transmit&nbsp;and=
&nbsp;receive at =0A1/3 speed from 0 to 60, replay&nbsp;starting 40 to 60).=
 Anyone =0Ainterested?</font></div>=0A<div><font size=3D"2" face=3D"Arial">=
</font>&nbsp;</div>=0A<div><font size=3D"2" face=3D"Arial">...</font></div>=
</div>=0A</div><meta content=3D"on" http-equiv=3D"x-dns-prefetch-control"><=
br><br>&nbsp;carr</div> </div> </blockquote>  </div></div></body></html>
--131296447-2112117833-1358002148=:84705--