Sign in with
Sign up | Sign in
Your question

Nikon Coolscan V ED JPG Compression Quality / or TIFF

Last response: in Digital Camera
Share
Anonymous
January 31, 2005 1:20:04 PM

Archived from groups: rec.photo.digital (More info?)

I recently acquired this scanner to digitize my 35mm and APS collection
of negatives. At 4000dpi with JPG compression set to highest quality
my files were averaging 38MB per neg. I then read on some NGs that the
best format was TIFF since it is lossless. While I agree with that
statement I honestly could not see a difference in quality on my
Samsung 1200NF 22" AG monitor. I decided to scan a neg and save it
both as a JPG (highest quality) and TIFF. The JPG again was
approximately 38MB while the TIFF was approximately 68MB; a big
difference but understandable since the TIFF was not compressed.
However, I then opened the JPG in Photoshop CS and saved it under a
different filename as a JPG (highest quality,12); with NO changes. The
new file size was 16MB. My issue then is I will eventually edit these
files and resave them in Photoshop CS. When doing so, the TIFF will be
the same size and quality while the JPG will be signicantly smaller and
possibly not of the same quality as the Coolscan JPG and definitely not
the same as a TIFF. My question then is:

1. If capacity is somewhat of an issue and none of the negs are
professional, will the JPG quality of the NIKON Coolscan save be a high
enough quality for future editing as long as I do not resave to the
same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?

2. Why is the JPG compression of Photoshop CS so much more than
Nikon's Coolscan software?

3. Should I forget JPG all together and just use TIFF?

50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
have a server which stores my data and I guess I could always add more
drives. It would then be a matter of backing-up the data which I do
daily to tape.

Thank you
Anonymous
January 31, 2005 5:30:53 PM

Archived from groups: rec.photo.digital (More info?)

Steph,

Yes, I considered DVD and have a DVD writer which I use to archive and
store offsite. One thing I failed to mention is that I use Photoshop
Album for organization and a DLink DSM320 media server for quick
viewing of pics and playing of MP3s. I realized after getting my first
digital camera that to enjoy all of the pictures I had to print them or
view via a computer. The media server allows me to create quick slide
shows for the family/friends and display them with music on my TV.
After all is said, the files would have to be readily available.

Nevertheless, I definitely archive to external media (DVD). The tape
solution is simply my IT administrative side coming through. The tape
solution is a DLT robot which is automated with daily backups, 3 months
retention, lossless compression, and allows quick recovery. The DVD
backup would require repetative manual execution to achieve the same
level of retention period.

Jason


Steph wrote:
> My solution would be to use a DVD writer - you'd have enough room on
2
> DVD-Rs for 130 TIFF files at a cost of less than £1, or roughly $1.
Backing
> up to tape is much slower and more costly.
Anonymous
January 31, 2005 5:38:28 PM

Archived from groups: rec.photo.digital (More info?)

I would only save in a lossless format and backup to external hard
drives that are offline and stored in a safe location. I would not
trust tape nor optical media for primary backup of important/master
files, but *maybe* as 2nd level backup would be OK. Personally, I
backup to 3 different external hard drives that are kept offline and
in different locations.
Related resources
Can't find your answer ? Ask !
Anonymous
January 31, 2005 5:48:35 PM

Archived from groups: rec.photo.digital (More info?)

Exactly... the tape solution is a 2nd level backup that is performed
automatically on a daily basis. I archive to CD or DVD. I guess I
would consider my external hard drive to be my server. The drives are
mirrored so if one fails I still have the other. Along with tape and
DVD as backups I think I am covered.
If storage was not an issue I guess TIFF would be the only way to go.
Anonymous
January 31, 2005 6:06:09 PM

Archived from groups: rec.photo.digital (More info?)

I am ready to send the email but can not confirm your address.
Anonymous
January 31, 2005 7:58:29 PM

Archived from groups: rec.photo.digital (More info?)

If you cannot tell the difference, then there is no point in wasting
storage. I have been scanning at 4000 dpi but saving the output as 2000 dpi
JPG. In case of something that are truly outstanding, then I rescan and
save as 48bit TIFF.

<istman@bellsouth.net> wrote in message
news:1107195604.932549.307950@f14g2000cwb.googlegroups.com...
>I recently acquired this scanner to digitize my 35mm and APS collection
> of negatives. At 4000dpi with JPG compression set to highest quality
> my files were averaging 38MB per neg. I then read on some NGs that the
> best format was TIFF since it is lossless. While I agree with that
> statement I honestly could not see a difference in quality on my
> Samsung 1200NF 22" AG monitor. I decided to scan a neg and save it
> both as a JPG (highest quality) and TIFF. The JPG again was
> approximately 38MB while the TIFF was approximately 68MB; a big
> difference but understandable since the TIFF was not compressed.
> However, I then opened the JPG in Photoshop CS and saved it under a
> different filename as a JPG (highest quality,12); with NO changes. The
> new file size was 16MB. My issue then is I will eventually edit these
> files and resave them in Photoshop CS. When doing so, the TIFF will be
> the same size and quality while the JPG will be signicantly smaller and
> possibly not of the same quality as the Coolscan JPG and definitely not
> the same as a TIFF. My question then is:
>
> 1. If capacity is somewhat of an issue and none of the negs are
> professional, will the JPG quality of the NIKON Coolscan save be a high
> enough quality for future editing as long as I do not resave to the
> same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?
>
> 2. Why is the JPG compression of Photoshop CS so much more than
> Nikon's Coolscan software?
>
> 3. Should I forget JPG all together and just use TIFF?
>
> 50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
> have a server which stores my data and I guess I could always add more
> drives. It would then be a matter of backing-up the data which I do
> daily to tape.
>
> Thank you
>
January 31, 2005 11:09:10 PM

Archived from groups: rec.photo.digital (More info?)

My solution would be to use a DVD writer - you'd have enough room on 2
DVD-Rs for 130 TIFF files at a cost of less than £1, or roughly $1. Backing
up to tape is much slower and more costly.
Anonymous
February 1, 2005 12:58:46 AM

Archived from groups: rec.photo.digital (More info?)

istman@bellsouth.net wrote:

> I recently acquired this scanner to digitize my 35mm and APS collection
> of negatives. At 4000dpi with JPG compression set to highest quality
> my files were averaging 38MB per neg. I then read on some NGs that the
> best format was TIFF since it is lossless.

Although this is true. In many circumstances the JPEG file may well be
perfectly adequate. You really only need TIFF (or PNG) when the image
will need contrast stretching or other vigorous processing.

> both as a JPG (highest quality) and TIFF. The JPG again was
> approximately 38MB while the TIFF was approximately 68MB; a big
> difference but understandable since the TIFF was not compressed.
> However, I then opened the JPG in Photoshop CS and saved it under a
> different filename as a JPG (highest quality,12); with NO changes. The
> new file size was 16MB.

> 1. If capacity is somewhat of an issue and none of the negs are
> professional, will the JPG quality of the NIKON Coolscan save be a high
> enough quality for future editing as long as I do not resave to the
> same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?

For most images it will probably be adequate, but only you can decide.
>
> 2. Why is the JPG compression of Photoshop CS so much more than
> Nikon's Coolscan software?

In part because Photoshop uses optimised Huffman tables and probably
different quantisation. It is a bad idea to save any work in progress as
a JPEG file. Each save will incur a slight additional degradation.

If you want a definitive answer please scan a *tiny* piece of an image
(filesize <2MB) at max JPEG quality and send it to me privately. I am
curious as to what the latest Nikon Coolscan is doing here.

The difference in filesize is surprising since Photoshop highest quality
makes pretty large files.

> 3. Should I forget JPG all together and just use TIFF?

Only if you have shares in a disk drive manufacturer.

> 50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
> have a server which stores my data and I guess I could always add more
> drives. It would then be a matter of backing-up the data which I do
> daily to tape.

It would be worth looking at using a transcoder that offers lossless
re-encoding of the JPEG stream with optimised Huffman tables. It might
save you a significant chunk of disk space. But always check that the
new files decode correctly and keep redundant copies of all data.

Regards,
Martin Brown
Anonymous
February 1, 2005 7:48:09 PM

Archived from groups: rec.photo.digital (More info?)

Martin,

Thank you very much. I have sent the two pics and anxiously await your
input.

Jason

"Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message
news:ctm9mn$68o$1@news6.svr.pol.co.uk...
> istman@bellsouth.net wrote:
>
>> I recently acquired this scanner to digitize my 35mm and APS collection
>> of negatives. At 4000dpi with JPG compression set to highest quality
>> my files were averaging 38MB per neg. I then read on some NGs that the
>> best format was TIFF since it is lossless.
>
> Although this is true. In many circumstances the JPEG file may well be
> perfectly adequate. You really only need TIFF (or PNG) when the image will
> need contrast stretching or other vigorous processing.
>
>> both as a JPG (highest quality) and TIFF. The JPG again was
>> approximately 38MB while the TIFF was approximately 68MB; a big
>> difference but understandable since the TIFF was not compressed.
>> However, I then opened the JPG in Photoshop CS and saved it under a
>> different filename as a JPG (highest quality,12); with NO changes. The
>> new file size was 16MB.
>
>> 1. If capacity is somewhat of an issue and none of the negs are
>> professional, will the JPG quality of the NIKON Coolscan save be a high
>> enough quality for future editing as long as I do not resave to the
>> same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?
>
> For most images it will probably be adequate, but only you can decide.
>>
>> 2. Why is the JPG compression of Photoshop CS so much more than
>> Nikon's Coolscan software?
>
> In part because Photoshop uses optimised Huffman tables and probably
> different quantisation. It is a bad idea to save any work in progress as a
> JPEG file. Each save will incur a slight additional degradation.
>
> If you want a definitive answer please scan a *tiny* piece of an image
> (filesize <2MB) at max JPEG quality and send it to me privately. I am
> curious as to what the latest Nikon Coolscan is doing here.
>
> The difference in filesize is surprising since Photoshop highest quality
> makes pretty large files.
>
>> 3. Should I forget JPG all together and just use TIFF?
>
> Only if you have shares in a disk drive manufacturer.
>
>> 50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
>> have a server which stores my data and I guess I could always add more
>> drives. It would then be a matter of backing-up the data which I do
>> daily to tape.
>
> It would be worth looking at using a transcoder that offers lossless
> re-encoding of the JPEG stream with optimised Huffman tables. It might
> save you a significant chunk of disk space. But always check that the new
> files decode correctly and keep redundant copies of all data.
>
> Regards,
> Martin Brown
Anonymous
February 1, 2005 9:27:49 PM

Archived from groups: rec.photo.digital (More info?)

istman@bellsouth.net wrote:
>
> I recently acquired this scanner to digitize my 35mm and APS collection
> of negatives. At 4000dpi with JPG compression set to highest quality
> my files were averaging 38MB per neg. I then read on some NGs that the
> best format was TIFF since it is lossless. While I agree with that
> statement I honestly could not see a difference in quality on my
> Samsung 1200NF 22" AG monitor. I decided to scan a neg and save it
> both as a JPG (highest quality) and TIFF. The JPG again was
> approximately 38MB while the TIFF was approximately 68MB; a big
> difference but understandable since the TIFF was not compressed.
> However, I then opened the JPG in Photoshop CS and saved it under a
> different filename as a JPG (highest quality,12); with NO changes. The
> new file size was 16MB. My issue then is I will eventually edit these
> files and resave them in Photoshop CS. When doing so, the TIFF will be
> the same size and quality while the JPG will be signicantly smaller and
> possibly not of the same quality as the Coolscan JPG and definitely not
> the same as a TIFF. My question then is:
>
> 1. If capacity is somewhat of an issue and none of the negs are
> professional, will the JPG quality of the NIKON Coolscan save be a high
> enough quality for future editing as long as I do not resave to the
> same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?
>
> 2. Why is the JPG compression of Photoshop CS so much more than
> Nikon's Coolscan software?
>
> 3. Should I forget JPG all together and just use TIFF?
>
> 50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
> have a server which stores my data and I guess I could always add more
> drives. It would then be a matter of backing-up the data which I do
> daily to tape.
>
> Thank you


Hello

Use LZW tiff compression in Photoshop, but at 8 bit.
This is also lossless. 16 bit tiffs do not compress well.

Mike Engles
Anonymous
February 1, 2005 11:58:40 PM

Archived from groups: rec.photo.digital (More info?)

I also sent an inquiry regarding the JPEG file size comparison to NIKON.
Here is there response:

<NIKON>
Thanks for your inquiry. TIFF is significantly superior to JPEG, as a TIFF
file is uncompressed and stores 8 bits per color channel (RGB) while JPEG is
compressed and stored 8 bits in total. The reason you are likely getting
different results from Nikon Scan as opposed to Photoshop is that the two
pieces of software have varying algorithm's for how they compress their
JPEG's. I tried it here, and was able to achieve a 35mb JPEG in Nikon Scan
as well. In Photoshop, setting the JPEG quality to 12 (the highest it will
go) I could only achieve 27mb... so it's simply a difference in the way
Photoshop saves it's JPEG.
<end NIKON>

Not really an answer but they too see the file size difference. I think I
will probably save to 8-bit TIFFs until my drives are full then get some
more. It is probably best since I do not want to have to scan any negatives
twice. Better safe than sorry.

I would still like to hear from Martin regarding the JPEG comparison.

Thanks to all for the input. I really like this group.

Jason

<istman@bellsouth.net> wrote in message
news:1107195604.932549.307950@f14g2000cwb.googlegroups.com...
>I recently acquired this scanner to digitize my 35mm and APS collection
> of negatives. At 4000dpi with JPG compression set to highest quality
> my files were averaging 38MB per neg. I then read on some NGs that the
> best format was TIFF since it is lossless. While I agree with that
> statement I honestly could not see a difference in quality on my
> Samsung 1200NF 22" AG monitor. I decided to scan a neg and save it
> both as a JPG (highest quality) and TIFF. The JPG again was
> approximately 38MB while the TIFF was approximately 68MB; a big
> difference but understandable since the TIFF was not compressed.
> However, I then opened the JPG in Photoshop CS and saved it under a
> different filename as a JPG (highest quality,12); with NO changes. The
> new file size was 16MB. My issue then is I will eventually edit these
> files and resave them in Photoshop CS. When doing so, the TIFF will be
> the same size and quality while the JPG will be signicantly smaller and
> possibly not of the same quality as the Coolscan JPG and definitely not
> the same as a TIFF. My question then is:
>
> 1. If capacity is somewhat of an issue and none of the negs are
> professional, will the JPG quality of the NIKON Coolscan save be a high
> enough quality for future editing as long as I do not resave to the
> same file? And, will the Photoshop CS JPG as good as a Coolscan JPG?
>
> 2. Why is the JPG compression of Photoshop CS so much more than
> Nikon's Coolscan software?
>
> 3. Should I forget JPG all together and just use TIFF?
>
> 50 rolls with 24 exposures will result in 84GB TIFFs or 43GB JPGs. I
> have a server which stores my data and I guess I could always add more
> drives. It would then be a matter of backing-up the data which I do
> daily to tape.
>
> Thank you
>
Anonymous
February 2, 2005 8:14:04 PM

Archived from groups: rec.photo.digital (More info?)

ISTMAN wrote:

> Thank you very much. I have sent the two pics and anxiously await your
> input.
>
> Jason
>
> "Martin Brown" <|||newspam|||@nezumi.demon.co.uk> wrote in message
> news:ctm9mn$68o$1@news6.svr.pol.co.uk...
>
>>istman@bellsouth.net wrote:
>>
>>>I recently acquired this scanner to digitize my 35mm and APS collection
>>>of negatives. At 4000dpi with JPG compression set to highest quality
>>>my files were averaging 38MB per neg. I then read on some NGs that the
>>>best format was TIFF since it is lossless.
>>
>>>both as a JPG (highest quality) and TIFF. The JPG again was
>>>approximately 38MB while the TIFF was approximately 68MB; a big
>>>difference but understandable since the TIFF was not compressed.
>>>However, I then opened the JPG in Photoshop CS and saved it under a
>>>different filename as a JPG (highest quality,12); with NO changes. The
>>>new file size was 16MB.

>>>2. Why is the JPG compression of Photoshop CS so much more than
>>>Nikon's Coolscan software?
>>
>>In part because Photoshop uses optimised Huffman tables and probably
>>different quantisation. It is a bad idea to save any work in progress as a
>>JPEG file. Each save will incur a slight additional degradation.

I guessed wrong. Both packages are using optimised Huffman tables with
full chroma subsampling to preserve colours better.

The difference is that the Nikon scan includes two copies of the image
thumbnail in the header and has an unusually fine quantisation table
that is identically "1" in every cell. Pushing the envelope of sensible
tables right to the limit! This trades a big increase in filesize for a
miniscule improvement in image quality (mainly it preserves high
frequency noise more effectively). How much of a difference it makes
will depend on the source material.

>>The difference in filesize is surprising since Photoshop highest quality
>>makes pretty large files.

Photoshop JPEG at max quality is generally good enough. The Nikon format
should be marginally better, and TIFF will beat both. Only you can
decide whether or not the additional increase in file size is
worthwhile. Avoid recompressing a JPEG if at all possible.

I keep PNG or TIFFs of difficult scans and JPEGs for the rest. YMMV

Regards,
Martin Brown
February 2, 2005 8:14:05 PM

Archived from groups: rec.photo.digital (More info?)

> ISTMAN wrote:
>
>
>>> The difference in filesize is surprising since Photoshop highest
>>> quality makes pretty large files.


Quality of 11 in PS is equal to Nikon's 'Fine' jpeg setting.
Quality of 12 is about twice the size.
!