| 文章內容 |
- From nrothschild@nmctech.com Fri May 17 11:31:36 2002
Return-Path: <nrothschild@nmctech.com>
X-Sender: nrothschild@nmctech.com
X-Apparently-To: colonial-coins@yahoogroups.com
Received: (EGP: mail-8_0_3_2); 17 May 2002 18:31:36 -0000
Received: (qmail 83482 invoked from network); 17 May 2002 18:31:36 -0000
Received: from unknown (66.218.66.216)
by m15.grp.scd.yahoo.com with QMQP; 17 May 2002 18:31:36 -0000
Received: from unknown (HELO mail01d.rapidsite.net) (207.158.192.52)
by mta1.grp.scd.yahoo.com with SMTP; 17 May 2002 18:31:36 -0000
Received: from www.nmctech.com (209.130.70.223)
by mail01d.rapidsite.net (RS ver 1.0.63s) with SMTP id 038121
for <colonial-coins@yahoogroups.com>; Fri, 17 May 2002 14:31:18 -0400 (EDT)
To: <colonial-coins@yahoogroups.com>
Subject: [Colonial Numismatics] Re: Ct Copper help (sorting by rarity)
Date: Fri, 17 May 2002 14:33:18 -0400
Message-ID: <MKEDLPECPDBKFIBMDEOHCEOMCFAA.nrothschild@nmctech.com>
MIME-Version: 1.0
Content-Type: text/plain;
charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000
X-Loop-Detect: 1
From: "Neil Rothschild" <nrothschild@nmctech.com>
Reply-To: <nrothschild@nmctech.com>
X-Yahoo-Group-Post: member; u=113398776
X-Yahoo-Profile: nrothschild1
-----Original Message-----
From: Neil Rothschild [mailto:nrothschild@dellepro.com]
Sent: Friday, May 17, 2002 1:20 PM
To: colonial-coins@yahoogroups.com
Subject: RE: [Colonial Numismatics] Re: Ct Copper help (sorting by
rarity)
Morris,
Glad you liked the tips....
I have a unique setup since I wrote software to help attribute Connecticuts.
I scan/photo a coin, and it displays next to a reference photo. I can
filter the list of reference photos (all 1787's, just draped bust lefts, all
obverse 33 and/or reverse Z or specific obverse/reverse, etc.)
After I set up a filter (example- all 33-Z, which is around 67 varieties), I
can flip through the reference coins until I find a match.
It is harder to do in practice than in theory, mainly because of the need
for good photos of each variety. A good coin is not necessarily a good
photo. Some varieties are all problematic strikes, etc. I can flag a
particular photo as the "reference" photo for a variety and that displays
first. I can then flip through all the photos in the database for that
variety if the reference photo is not suitable. So it is an evolving
project to get the right reference photos to do this quickly. Most of my
photos are scans of auction catalog half tones and many leave much to be
desired.
I have been working on the mechanics of the attribution tool for years,
tweaking it little by little. The last thing I did was to add the known
population (in my database) as a column in the filter list. By sorting on
that column in descending sequence, I can flip through the most common to
least common for a given list. 95% of unattributed 33-Z's will fall into
the first 10-20 of the 67 total varieties. That saves a lot of time. By
the time I get done flipping through 67 possible matches, my eyes are glazed
over. Even if I can't nail the variety right away, I get a sense of the
potential rarity if I can eliminate the more common varieties.
One of the problems with attributing draped bust left CT's is that although
most obverses and reverses are common within at least one die combination,
there are many that have no good photos (at least in my database).
Eventually I hope to build a "hypothetical" collection of photos, where each
variety is represented with the best obverse and reverse photo known to me,
even if the photo is not of a coin in that die combination. The downside is
that certain combinations have certain strike, centering and die break
characteristics, and the resulting images would be mis-leading. The photos
would only be used in the context of die diagnostics. This would actually
be a set of obverses, one per die, and a set of reverses, one per die, and
my software would put them together according to the variety. This would
minimize the number of photos required, but would require a lot of software
work to integrate into my existing displays, which are based on a single
photo of obverse and reverse. Plus selecting and cropping down the 300 odd
die photos. Plus maintenance as better photos come along, etc.
Just one more good idea that may never get done because of the time involved
<s>
Back to your question...
As I was writing my tip list, it occurred to me that a collector could
somehow cut and paste a set of photos in rarity sequence, rather than
variety sequence. For this purpose, published rarities would work well.
I'm not really sure how I would approach it, but it is an idea worth
exploring for anyone who does a lot of Ct attributions (or any other wide
series). Obviously, some catalog would have to be sacrificed (actually TWO
catalogs) or good quality and cheap reproductions made. A "best of" set
would require many more sacrifices or repros. If you have two spare
Taylors, that would make a good start, I guess. It would be easier, faster
and cheaper than all the scanning and software work that I did.
If anyone wants a list of the most common varieties, in true(er) rarity
sequence, I could put that together and post it. Although my populations
are probably not exact (I believe the more common varieties are more common
than the numbers indicate, for example) I was very careful to record even
common varieties in my database qwhen enterig various auctions and other
sources and I believe the sequence (common to less common) would be very
close to reality.
Another good way to organize photos is by legend devices. Cinquefoils,
scrolls, etc. I'd like to filter my reference list to include only stars or
cinquefoils, or scrolls, or whatever. Large letters, small letters, etc.
I've thought about it but never actually done it. Then within a group, by
rarity. I guess the EAC75 sale is more or less organized that way. Most
modern collectors don't like EAC75 because it is not in variety sequence,
but the way it is grouped is very useful as an attribution tool. It is
probably the most UNDER utilized Connecticut reference.
Regards,
Neil
-----Original Message-----
From: Morris Hankins [mailto:joshalso2000@yahoo.com]
Sent: Thursday, May 16, 2002 6:42 PM
To: colonial-coins@yahoogroups.com
Subject: Re: [Colonial Numismatics] Re: Ct Copper help
Neil - superb set of tips even for the "experts".
Would love for your to expand on Tip #10 specifically
the thirty varieties that make up the 98%.
I am sure that a lot of people are giving thanks for
these tips.
Morris
--- nrothschild1 <nrothschild@comcast.net> wrote:
> Ray,
>
> What threw you off? Your reverse has a clear hyphen
> between ET and
> LIB.
>
> The punctuation (hyphens/dots/etc) between ET and
> LIB are critical
> for CT attributions. More problematic for worn
> specimens (unlike
> yours, which is very nice).
>
> Here are a few tips that help me:
>
> 1) If the obverse is 33, start with the reverse. I
> usually do the
> reverse first anyway, but that is a personal
> preference.
>
> 2) If worn, assume punctuation may be missing, so
> make a list of
> anything that COULD be your coin- if your coin has
> one cinquefoil
> after INDE, the actual die might have two, with one
> missing. Same
> with the ET-LIB punctuation and ETLIB/ETLIR. With
> a little practice
> you can tell if the coin is worn enough to miss a
> device. This may
> not eliminate much, but it's better than looking at
> everything.
>
> 3) I also always start with the Miller type list,
> as you did.
>
> 4) As mentioned, die breaks are critical. s.1, for
> instance is
> always broken to the left of the face to the rim.
> It's a dead ringer
> if you spend a few thousand hours with he series <s>
>
> 5) Study the space between ET and LIB VERY
> CAREFULLY. If there is
> ANYTHING disturbing the field, assume it is
> punctuation and try to
> find something that fits. Some of the hyphens will
> look like dots,
> so don't take Miller too literally. There is not
> much difference
> between a short hyphen and a dot, especially if
> there is a weak
> strike or wear or other disturbances in the field.
>
> 6) Learn the difference between Jarvis letters and
> original Company
> of Coining Coppers letters. The Jarvis letters are
> smaller and
> distinctive. That immediately narrows the search if
> you know which
> types are from each mint. Obverses 16-29 are
> generally CCC, 32, 33
> are Jarvis. Look at Taylor and you should see the
> difference.
> Jarvis letters are the same as Fugios. Of course,
> Jarvis used
> cinquefoils and CCC did not....
>
> 7) There are ETLIB and ETLIR legends. Each ETLIR
> usually has a
> sister type with ETLIB, all else is the same. There
> is a continuum
> between a fully formed B and a fully formed R. A
> fully formed B and
> a fully formed R is usually obvious, but there are
> some dies that I
> think (IMHO) could be classed as ETLIB or ETLIR. So
> if still
> stumped, check the ETLIRs if you assumed ETLIB, etc.
> In the case of
> certain ETLIBs, the bottom crossbar of the B is very
> delicate and in
> VG or lower grade, can be easily missed or
> completely missing. So if
> you think it is an ETLIR, and it is low grade,
> always check the
> sister ETLIB type if you can't find an ETLIR to
> match. I have seen
> ETLIR's that I think should have been classed as
> ETLIB, but I suspect
> Miller might not have seen a high enough grade to
> see the delicate
> crossbar.
>
> 8) For 1786, forget it. These were all hubbed and
> it's like doing
> Morgan VAM varieties. Very diffcult unless the coin
> is high grade
> and well struck.
>
> 9) For 1787 reverses the BRANCH is the key,
> especially for Z
> reverses. It is almost always well struck and wears
> well. Every
> branch on every die is unique, as they were all
> constructed with tiny
> punches. Same for 1786, by the way. Same for every
> CT die, for that
> matter.
>
> 10) 98% of unattributed 1787 CT's are one of maybe
> 30 varieties.
> And most unattributed Ct's are 1787. When I am
> doing tough types,
> like 33 and/or Z, I search in rarity sequence (but I
> do it with a
> computer.. tougher by hand).
>
> 11) If you want to spend more time than you really
> want to put into
> this (since it is not one of your specialties) you
> could do this
>
> a. Get your Taylor, EAC 75, Perkins, Oechsner,
> Hessberg, Stacks
> 6/94, maybe a few others.... together.
>
> b) Decide which has the best overall photos
> for attibution-
> Taylor or Perkins. Most people prefer Taylor since
> he had more high
> cencus pieces. That is your "master" guide.
>
> c) Look at each variety in (Taylor/Perkins,
> the master catalog)
> and where the variety is weak or difficult to use,
> check the other
> catalogs. If there is a better specimen in one of
> the other sales,
> make a note in your master catalog (i.e. "EAC 75:999
> is best pix").
>
> 12. Re-read tip #9. It is THE KEY.
>
> 13) Upload photo to this group. It worked the last
> time you tried!
>
> Hope this helps a little....
>
> Neil
>
>
>
>
>
>
>
> --- In colonial-coins@y..., Ray Williams
> <njraywms@o...> wrote:
> > Could any of you CT guys out there help me
> with attribution? I
> was thinking this is a 33 obverse and a T reverse
> but I'm having
> difficulty matching it. I just got this off ebay
> from Don V as a Red
> Book type coin.
> >
> > Thanks,
> > Ray
>
>
> ------------------------ Yahoo! Groups Sponsor
>
> To unsubscribe from this group, send an email to:
> colonial-coins-unsubscribe@egroups.com
>
>
>
> Your use of Yahoo! Groups is subject to
> http://docs.yahoo.com/info/terms/
>
>
__________________________________________________
Do You Yahoo!?
LAUNCH - Your Yahoo! Music Experience
http://launch.yahoo.com
To unsubscribe from this group, send an email to:
colonial-coins-unsubscribe@egroups.com
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
|
|---|
