In message <upk2va$2quv7$1@dont-email.me> at Sat, 3 Feb 2024 00:56:10,
Java Jive <java@evij.com.invalid> writes
>On 02/02/2024 23:21, J. P. Gilliver wrote:
>> In message <upjs8u$2q0s8$1@dont-email.me> at Fri, 2 Feb 2024
>>23:01:50, Java Jive <java@evij.com.invalid> writes
>>> On 02/02/2024 21:38, J. P. Gilliver wrote:
>>>>
>>>> In message <upjk2g$2om7c$1@dont-email.me> at Fri, 2 Feb 2024
>>>>20:41:52,  Java Jive <java@evij.com.invalid> writes
>> []
>>>>> In my experience, as long as both are in equal condition (in other
>>>>>words without either being degraded markedly by time), you'll get
>>>>>better results from the negative.  There's the cropping that you
>>>>>mention, but also the fact that in many prints the images were
>>>>>constructed from a dot pattern that interacts with the resolution
[]
>>>>  Yes, I'm familiar with the dot patterns in (almost) any sort of
>> []
>>>> developing it, i. e.  no dot _pattern_ - just the grain of the
>>>>negative and paper.
>>>
>>> I have enlargements from the 60s or 70s which show this effect.
>
>Actually, TBF, those show a similar but slightly different problem, as
>described below ...
>
>> Interesting! What do you think is the cause? I remember seeing -
>>though not from which decade - a sort of hexagon patterning in the
>>emulsion on some prints (deliberate - I think it was some deliberate
>>effect - maybe trying to look like canvas?), but I think that was a
>>lot coarser than the pattern you illustrated, which looks more like a
>>very fine screen matrix as used in printing.
>
>I think it's particularly noticeable on those that were done on matt as
>opposed to glossy paper. I preferred matt paper because with glossy
>paper you tended to just see reflections of windows in the room, etc.

Me too - not sure if for that reason, or that glossy ones seemed to show
fingerprints much more; also felt like the glossy surface was an extra
something between you and the image (though the opposite may have been
the case, see below).

>However, that preference proved a disadvantage when I came to scan the
>photos many years later. While for most of them I still had the negs
[]
>paper in the resulting scan. Here's an example, the photo was taken in
>the early '60s, but I think the enlargement that was scanned was most
>probably made around 1970 when I was working at the photo-lab. The
>effect is nothing like as noticeable as in the last example, but if you
>download the image and zoom in to it far enough, you can see quite
>clearly the dot pattern of the matt surface of the paper. This image

Yes, I can - a definite hexagonal grid. Are you saying that all matt
prints exhibit this - i. e., the natural finish of prints is glossy, and
to get matt they had to be (presumably a mechanical process) embossed
with a pattern like that? Interesting, if so.
[]
>www.macfh.co.uk/Temp/Pap_Of_Glencoe_From_Across_Loch_Leven.png
>
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

They'd never heard of me; they didn't like me; they didn't like my speech;
they tutted and clucked and looked at their watches and eventually I sat down
to a thunderous lack of applause. - Barry Norman (on preceding Douglas Bader),
in RT 6-12 July 2013

On 03/02/2024 07:27, J. P. Gilliver wrote:
>
> In message <upk2va$2quv7$1@dont-email.me> at Sat, 3 Feb 2024 00:56:10,
> Java Jive <java@evij.com.invalid> writes
>>
>> Here's an example, the photo was taken
>> in the early '60s, but I think the enlargement that was scanned was
>> most probably made around 1970 when I was working at the photo-lab.
>> The effect is nothing like as noticeable as in the last example, but
>> if you download the image and zoom in to it far enough, you can see
>> quite clearly the dot pattern of the matt surface of the paper.
>
> Yes, I can - a definite hexagonal grid. Are you saying that all matt
> prints exhibit this - i. e., the natural finish of prints is glossy, and
> to get matt they had to be (presumably a mechanical process) embossed
> with a pattern like that? Interesting, if so.

That I can't answer, I just know that wherever I had to use a print
rather than a neg, for reasons already given including but not only the
above, the results weren't as good.

>> www.macfh.co.uk/Temp/Pap_Of_Glencoe_From_Across_Loch_Leven.png

--

Fake news kills!

I may be contacted via the contact address given on my website:
www.macfh.co.uk

J. P. Gilliver <G6JPG@255soft.uk> wrote:
[...]
> In message <Ne9vN.56351$5Hnd.38815@fx03.iad> at Fri, 2 Feb 2024
> 11:02:37, Mark Lloyd <not.email@all.invalid> writes
> >[snip]
> >
> >> The bottom mostly fell out of the standalone flatbed scanner market,
> >>I suspect, when printer manufacturers started incorporating them. And
> >>printers last even less long than OSs these days.
> >
> >I usually don't like combination devices, although a printer/scanner
> >combination does make it easy to make copies.
> >
> Me neither, because if one part fails, you're stuck with either having
> to replace the lot, or having to keep an only-partly-working device.
> (And "fails" can include "doesn't work with ..." as well as actually
> failing.) And I _believe_ in the case of _some_ printer/scanners, they
> won't scan if there's no ink in one of the cartridges.
>
> OK, the copying ability is handy. But not on the whole at the expense of
> having to use an ink printer. (There _are_ a few laser/scanner models,
> but very few.)

Not that I would recommend a printer/scanner - for the reasons
mentioned -, but there are quite a lot of laser/scanner models, at least
from HP. A quick check shows some 15 models.

FWIW, I have an HP LaserJet Pro M1132 MFP, bought in January 2015 for
EUR 99.99. But I bought it for the printing part and just use the
scanner for the odd quick 'photo copy'.

My scanners are an EPSON PERFECTION V30 and a Canon CanoScan 9000F
MarkII slide scanner. The Epson started its life on Vista, then 8.1 and
now 11.

gfretwell@aol.com wrote:
> On Fri, 2 Feb 2024 15:10:27 +0000, "J. P. Gilliver" <G6JPG@255soft.uk>
> wrote:
[...]

> I feel your pain. I have a few scanners and they all have limited
> driver support. It seems like they want you to buy a new scanner every
> time you get a different OS. The hardware doesn't seem to change, just
> the drivers.
> That Vue Scan does support lots of old scanners but it might just be
> cheaper to get a new scanner.

John's current issue is the other way around, a new scanner on an old
OS (Windows 7, 32-bit).

As I just posted, I've been 'lucky' with the direction you mention, an
old scanner on a new OS. My EPSON PERFECTION V30 flatbed scanner started
its life on Vista, then 8.1 and now 11.

On 2/3/2024 11:34 AM, Frank Slootweg wrote:
> J. P. Gilliver <G6JPG@255soft.uk> wrote:
> [...]
>> In message <Ne9vN.56351$5Hnd.38815@fx03.iad> at Fri, 2 Feb 2024
>> 11:02:37, Mark Lloyd <not.email@all.invalid> writes
>>> [snip]
>>>
>>>> The bottom mostly fell out of the standalone flatbed scanner market,
>>>> I suspect, when printer manufacturers started incorporating them. And
>>>> printers last even less long than OSs these days.
>>>
>>> I usually don't like combination devices, although a printer/scanner
>>> combination does make it easy to make copies.
>>>
>> Me neither, because if one part fails, you're stuck with either having
>> to replace the lot, or having to keep an only-partly-working device.
>> (And "fails" can include "doesn't work with ..." as well as actually
>> failing.) And I _believe_ in the case of _some_ printer/scanners, they
>> won't scan if there's no ink in one of the cartridges.
>>
>> OK, the copying ability is handy. But not on the whole at the expense of
>> having to use an ink printer. (There _are_ a few laser/scanner models,
>> but very few.)
>
> Not that I would recommend a printer/scanner - for the reasons
> mentioned -, but there are quite a lot of laser/scanner models, at least
> from HP. A quick check shows some 15 models.
>
> FWIW, I have an HP LaserJet Pro M1132 MFP, bought in January 2015 for
> EUR 99.99. But I bought it for the printing part and just use the
> scanner for the odd quick 'photo copy'.
>
> My scanners are an EPSON PERFECTION V30 and a Canon CanoScan 9000F
> MarkII slide scanner. The Epson started its life on Vista, then 8.1 and
> now 11.
>

https://www.filmscanner.info/en/CanonCanoScan9000FMark2.html

"Resolution of the CAnon CAnoScan 9000F Mark II

The resolution test using the USAF-testchart yields an actual average
resolution of approximately 1700 ppi according to our resolution chart,
the same value as the predecessor CanoScan 9000F. This comes up to
about 17% of the rated value of 9600ppi. The attained resolution value
is hence still too low to scan 35mm material for an output size larger
than 13x18cm without degradation."

So that's a statement about the properties of the optical path, versus
how many elements the CCD has in it. If set to 9600 DPI, you still get
the 9600 dots, but the sharpness value is not the same.

The same thing happens with analog television. The "effective resolution"
is different than the scan lines used by the TV standard. And I believe
at least some analog TV "test patterns", used to include materials
suitable for the resolution test described above.

https://www.drumscanning.co.uk/about/resolution/

"Manufacturers say they can achieve up to 11,000dpi for the very high end models
(but we’ve tested most and the limit seems to be about 6000dpi for real film).
However, with higher resolutions we get diminishing returns. We’ve tested 35mm,
medium format and large format (5×4 and 10×8), and have found that there is almost
nothing gained above 5000dpi (6000dpi for some 35mm) apart from a cleaner resolving
of grain and for large format film shot at typical apertures, there is usually
little benefit of scanning past 3000dpi...
"

So even with a drum scanner, that gives you some idea where some of the film poops out.

There is actually some film that did benefit from more than 3000, but the
person operating the camera went to a lot of trouble to do that. He was
getting useful results at 4800 optical limits. This was someone resolving
blades of grass in his photographs ("nerd alert!"). But it's a hobby.
and the grass didn't mind.

Paul

"J. P. Gilliver" <G6JPG@255soft.uk> wrote:

> VanguardLH <V@nguard.LH> *WROTE*:
>
>> Are you planning on buying a Canon scanner that has already been
>> discontinued?
>
> I'm not worried about whether it's been discontinued - I'm using Windows
> 7, after all (look where we're discussing this!); I just want to be sure
> drivers are available to make it work with my laptop.

While Canon provides drivers for old models to use on old Windows
versions, getting a discontinued model could mean no warranty. Scanners
are mechanical devices. The more you use, the sooner the death. Even
if you don't much use a scanner, it could die after a few uses if it had
a manufacturing defect. Using an old OS should dictate contraints on
buying new hardware except for driver availability which doesn't seem a
problem with Canon.

In message <uplrqd$376fh$1@dont-email.me> at Sat, 3 Feb 2024 12:06:20,
Paul <nospam@needed.invalid> writes
>On 2/3/2024 11:34 AM, Frank Slootweg wrote:
[]
>> Not that I would recommend a printer/scanner - for the reasons
>> mentioned -, but there are quite a lot of laser/scanner models, at least
>> from HP. A quick check shows some 15 models.

Interesting. Good. (Last time I looked - quite a few years ago - there
were some printer/scanner models where the printer was laser, but not
colour. The scanner was colour.)
>>
>> FWIW, I have an HP LaserJet Pro M1132 MFP, bought in January 2015 for
>> EUR 99.99. But I bought it for the printing part and just use the
>> scanner for the odd quick 'photo copy'.

Mine's a Samsung CLP775 - bought for 25 pounds, and I knew each
cartridge would cost me more than that when it ran out! (Which it did,
but not much more.) It's a huge thing - about a two foot cube, and
weighs - well, I'm not sure I could lift it now. But it's a nice machine
- does double sided, and (although I know that's the driver) booklet
printing. And I know it works with at least Windows 10, as it did with
the council laptop I had.
>>
>> My scanners are an EPSON PERFECTION V30 and a Canon CanoScan 9000F
>> MarkII slide scanner. The Epson started its life on Vista, then 8.1 and

Very nice. I have some film/slide scanners, and did have the 656 until
it played up.

>> now 11.
>>
That's lucky!
>
>https://www.filmscanner.info/en/CanonCanoScan9000FMark2.html
>
> "Resolution of the CAnon CAnoScan 9000F Mark II
>
> The resolution test using the USAF-testchart yields an actual average
> resolution of approximately 1700 ppi according to our resolution chart,
> the same value as the predecessor CanoScan 9000F. This comes up to
> about 17% of the rated value of 9600ppi. The attained resolution value

That's what the review I came across said - that some scanners, even if
they have a high sensor resolution (we're not talking silly
interpolation here), don't have optics good enough to use it.

> is hence still too low to scan 35mm material for an output size larger
> than 13x18cm without degradation."
>
>So that's a statement about the properties of the optical path, versus
>how many elements the CCD has in it. If set to 9600 DPI, you still get
>the 9600 dots, but the sharpness value is not the same.

Indeed - little point, just makes a bigger file (and slower scanning,
too) - not that different from interpolation, in fact.
>
>The same thing happens with analog television. The "effective resolution"
>is different than the scan lines used by the TV standard. And I believe
>at least some analog TV "test patterns", used to include materials
>suitable for the resolution test described above.

Well, most test cards had resolution grids - usually vertical, but
sometimes diagonal. The situation was somewhat different: vertical
resolution _was_ the number of lines (less a few for flyback: here in
625-land, I think vertical was 576i; I'm not sure what it was in
525-land - probably 480, as I presume that's where the old computer
monitor number came from). But horizontal depended on the bandwidth of
the allocated channel or recording equipment; in 625-land, UK had about
6 MHz broadcast channels (rest of EU 5.5), and home video recorders
about 3 MHz (V2000 and Betamax) or 2.5 (VHS).
>
>https://www.drumscanning.co.uk/about/resolution/
>
> "Manufacturers say they can achieve up to 11,000dpi for the very
>high end models
> (but we’ve tested most and the limit seems to be about 6000dpi
>for real film).
> However, with higher resolutions we get diminishing returns.
>We’ve tested 35mm,
> medium format and large format (5×4 and 10×8), and have found
>that there is almost
> nothing gained above 5000dpi (6000dpi for some 35mm) apart from a
>cleaner resolving
> of grain and for large format film shot at typical apertures, there
>is usually
> little benefit of scanning past 3000dpi...
> "
Yes, above a certain point, you just get a better picture of the film
grains!
>
>So even with a drum scanner, that gives you some idea where some of the
>film poops out.
>
>There is actually some film that did benefit from more than 3000, but the
>person operating the camera went to a lot of trouble to do that. He was
>getting useful results at 4800 optical limits. This was someone resolving
>blades of grass in his photographs ("nerd alert!"). But it's a hobby.
>and the grass didn't mind.
>
> Paul

Depends on the "speed" of the film; in general, high speed (i. e. works
with low light) had larger grains. About the lowest speed generally
available in "35mm" (135 format) was 25ASA, with 50 or 64 being
commoner, and 100, 200, and 400 widely available, especially in B/W. (I
once got hold of a roll of 1000 ASA; it was good for some long shots I
took of an indoor ice show, but definitely had a "sparkly" appearance.)
For home movies - 8mm - 40ASA (super 8) and 25 (standard 8) were the
norm - the tiny size of the film frames meant it had to be fine-grain
film; this was compensated for by not needing a huge lens to get a
relatively large amount of light onto the small area, and also moving
images make grain less noticeable. (The old camera I started with - the
same or a very similar model to the one Mr. Zapruder used - had
autoexposure for 10 ASA film, and when I was using it - 1970s - there
was one make still available at that speed [Perutz].) There were slower
(finer grain) films for specialised applications, such as document
copying where exposure time wasn't a problem, and I presume for
microfilm and microfiche use.

But for films used in "normal" cameras - either "35mm" [or Instamatic]
or 120 format - what they said, the grain size normally used was such
that 3000-4800 DPI resolution is probably the most you'd need. Certainly
for prints.
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

one can't go from `supposed crackpot ideas have been right before' to `we
should
take this latest crackpot idea onboard without making it fight for acceptance
like all the previous ones'. - Richard Caley, 2002 February 11 00:02:28

In message <1gybky8x2xhk6.dlg@v.nguard.lh> at Sat, 3 Feb 2024 12:50:31,
VanguardLH <V@nguard.LH> writes
>"J. P. Gilliver" <G6JPG@255soft.uk> wrote:
>
>> VanguardLH <V@nguard.LH> *WROTE*:
>>
>>> Are you planning on buying a Canon scanner that has already been
>>> discontinued?
>>
>> I'm not worried about whether it's been discontinued - I'm using Windows
>> 7, after all (look where we're discussing this!); I just want to be sure
>> drivers are available to make it work with my laptop.
>
>While Canon provides drivers for old models to use on old Windows
>versions, getting a discontinued model could mean no warranty. Scanners

It's more a matter of ensuring it will work with (has drivers for) my
OS.

>are mechanical devices. The more you use, the sooner the death. Even
>if you don't much use a scanner, it could die after a few uses if it had
>a manufacturing defect. Using an old OS should dictate contraints on
>buying new hardware except for driver availability which doesn't seem a
>problem with Canon.

I'll be buying second-hand, at a price level where I won't expect any
warranty, other than that it is working at the point I get it! (My 656 -
which I did I think buy new [I have the box etc. anyway] - was long out
of warranty by the time it died.) I nearly got a 5200F - far better than
I was thinking of, and with film/slide holder - for a ridiculously low
price, but someone beat me to it. As you say, Canon seem good at having
drivers (for 7-32) for their older models, back to LiDE
20/25/35/60/90/100/200/210/220/400, though for some reason not LiDE 30
and 50. EPSON some, hp some, Mustek not really (maybe via VueScan, but I
don't really want to go that route).
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

"When _I_ saw him, he was dead." "uh, he looked exactly the same when he was
alive, except he was vertical." (The Trouble with Harry)