SHQ, HQ, or SQ?

[ME-GM]

I took several shots of the same subjecst in three different modes. To me, I have to blow it up real big (600 to 800 percent) to see some differences but nothing real major.
Just wondering if anybody else did the comparison and give me some input.
Thanks.

 

[eileen68597]

To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.

And you won't see anything major at the outset. SHQ, HQ and SQ refer, obliquely, to the compression ratios of the generated .jpg files. Files shot at SHQ are recorded with a compression ratio of 1:2.7. HQ are recorded with a compression ratio of 1:4. SQ are recorded with a compression ratio of 1:8. Every time you open a .jpg file, it decompresses. Every time you save a .jpg file, it recompresses. This process, if done repeatedly, results in artifacting, aka "jaggies". SHQ files tend to remain stable longer due to their minimal compression; however, they will artifact at some point. HQ files are less stable, but OK if you're not planning on saving over them too often. SQ, well, they're unstable from the first click of the "save" button.

IMHO, it's best to shoot in the highest quality you possibly can (SHQ .jpg at minimum, lossless .Tiff or RAW optimally). Then, upon opening the image in your image editor of choice, do a "save as" and ensure the copy is saved in .tiff format.

'Hope this is helpful!

eileen

 

[ME-GM]

First, thanks for your comment.

I realized what you wrote. If I was to work on a file, I open it and save it as something that's not being compressed (i.e. .bmp or .tiff file) and go on from there. I was just curious about the quality of the pictures came straight from the camera. SHQ is almost 3X the size of SQ while I can't notice much differences even in blow way up mode.

To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.

And you won't see anything major at the outset. SHQ, HQ and SQ
refer, obliquely, to the compression ratios of the generated .jpg
files. Files shot at SHQ are recorded with a compression ratio of
1:2.7. HQ are recorded with a compression ratio of 1:4. SQ are
recorded with a compression ratio of 1:8. Every time you open a .jpg file, it decompresses. Every time you save a .jpg file, it
recompresses. This process, if done repeatedly, results in
artifacting, aka "jaggies". SHQ files tend to remain stable longer
due to their minimal compression; however, they will artifact at
some point. HQ files are less stable, but OK if you're not
planning on saving over them too often. SQ, well, they're unstable
from the first click of the "save" button.

IMHO, it's best to shoot in the highest quality you possibly can
(SHQ .jpg at minimum, lossless .Tiff or RAW optimally). Then,
upon opening the image in your image editor of choice, do a "save
as" and ensure the copy is saved in .tiff format.

'Hope this is helpful!

eileen

 

[Joe Peoples]

ME-GM wrote:
I was just curious about the

quality of the pictures came straight from the camera. SHQ is
almost 3X the size of SQ while I can't notice much differences even
in blow way up mode.>

Joe Peoples writes:

Did you print the different files or view them on your monitor?

 

[Jerry Monroe]

Eileen,

Thanks for the succinct explanation. I was aware of most of what you wrote but would like your opinion concerning the following: suppose I were to shoot at SQ (1/8) but (a) never do a save until the final product is completed and ready for printing or (b) not even save for the final print but instead just alter the image and then print it? I believe that the latter is what happens (or can happen if you handle it correctly) in Qimage with its filter concept that leaves the original image intact.

I would think that there would be little image deterioration in example (a) and no image deterioration in example (b). Am I correct in this?

Would it make a difference in example (b) if I were to shoot in HQ or SHQ?

To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.

And you won't see anything major at the outset. SHQ, HQ and SQ
refer, obliquely, to the compression ratios of the generated .jpg
files. Files shot at SHQ are recorded with a compression ratio of
1:2.7. HQ are recorded with a compression ratio of 1:4. SQ are
recorded with a compression ratio of 1:8. Every time you open a .jpg file, it decompresses. Every time you save a .jpg file, it
recompresses. This process, if done repeatedly, results in
artifacting, aka "jaggies". SHQ files tend to remain stable longer
due to their minimal compression; however, they will artifact at
some point. HQ files are less stable, but OK if you're not
planning on saving over them too often. SQ, well, they're unstable
from the first click of the "save" button.

IMHO, it's best to shoot in the highest quality you possibly can
(SHQ .jpg at minimum, lossless .Tiff or RAW optimally). Then,
upon opening the image in your image editor of choice, do a "save
as" and ensure the copy is saved in .tiff format.

'Hope this is helpful!

eileen

 

[Vance]

Howdy ME-GM,

First you can set those to anything you want.

Personally, when I got the 340mb MD, I set all mine to the maximum resolution, and then each is set to a different compression, starting at 1/2.7

Think of the SHQ/HQ/SQ as how large you want the image file to be.

Trust me. You will see a difference between say a SHQ/max res/1-2.7 and a RAW file. Their subtle, but can be seen.

Hope that helps, and doesn't confuse you more. I'm at work, and can loose myself and what I'm trying to say sometimes. )

--Vance.

 

[Vance]

Jerry,

You always want to make a copy of your original, and work on the copy. NEVER work on the original file. If something goes haywire, and you can't go back to reshoot, your screwed.

--Vance.

 

[Lawrence Keeney]

I now follow a procedure that someone else on this forum suggested and it has worked for me. I keep all my photographs in a folder with the year as the name of the folder. Under this folder, I have two folders, one named Original Photos, and the other named Working Photos. I download all photos from the camera into the Original Photos folder, and then change them to Read Only.

This way, the original files can never be overwritten.

I can't remember who suggested this originally, but I think it was a good suggestion.

Lawrence

Jerry,

You always want to make a copy of your original, and work on the
copy. NEVER work on the original file. If something goes haywire,
and you can't go back to reshoot, your screwed.

--
Vance.

http://users.ev1.net/~txcowboy

 

[ME-GM]

My question was simply mean that I can't tell any difference between the three modes without blowing up the picture. And by doing this on the monitor (I have a 21" and resolution set at 12--by 10-- with 32 bit color depth and calibrated with photoshop). I was just trying to compare the differences in clarity and details lost after the file is compressed in three different modes. I never work on the original and never use jpg format while working on a file.

I am just trying to clarify that do I need to bring a stack of compact flash with me to a shooting or just simply let the camera compress the file a little more and carry less cards.

 

[Vance]

I guess we can get carried away sometimes. )

I would suggest that if you are trying to put a product out, that you would want it to be the best. Therefore, use the best your camera can do. Save the other settings for when you're just out having fun. )

I use the E10, and the 340mb Microdrive works fine with nimh batteries or the lipo grip. I also carry 2 Kingston 128mb CF cards, and the 3gig Digital wallet. (just incase ;o)
--Vance.

 

[eileen68597]

In theory, I agree with Vance; however, I just have to interject a few items:

1. When I shoot in any mode other than RAW (and since most of my end results are not intended to be "photorealistic" but, rather "photosurrealistic") I shoot in SHQ .jpg;

2. If I open the SHQ file in Photoshop and determine it's, definitely, something I want to work with further, I save a copy thereof as a lossless .tiff and do all my additional work thereon. I keep the original file as my "negative" and never, ever open it in an editor again;

3. If I open the SHQ file in Photoshop and determine it's something I MIGHT like to work with in the future I, merely, close the file. That way, I've only done one true save [capture to record on card] and a total of two compressions and have, thus, reduced the artifacting down to minimum;

4. If I open the SHQ file in Photoshop and determine it's not something I will ever work with, I delete the file to save drive space.

Now...a caveat: The smallest real print I create is 8 X 10. (I may do smaller proofs just to get a read on the color values - but those prints are just for my viewing only.) Imaging experts will tell you that it's safe to record in SQ IF you know, for a fact, you will never, ever print the resulting image any larger than 8 X 10. This might, in fact, be true if the images you're recording are meant to be snapshots capturing a moment in time; however, how can you KNOW - straight up - you'll never print that image any larger than 8 X 10? How can you know that image - or some part thereof - will be so not worth printing in a massive size - either directly "as is" or as a result of maniputlation - that you'd, intentionally, not capture it in the best manner possible? Granted, artifacting CAN be reparied - but unless you happen to like sitting mere inches from a computer screen adding additional data to an image manually, why would you want to go through all that trouble?

Again, I hope this is helpful to you

eileen

 

[leohoward]

Eileen,

Im kind of confused on your 3rd point, maybe you could clarify for me, the part where you say " a total of two compressions " has me a bit confused, or maybe Im reading it wrong, but if you capture an image in any mode, it does its initial compression, when you open it to view it and decide not to do anything with the file now, and you close it without making any changes, where does the " second compression" happen?

Thanks

Leo Howard

3. If I open the SHQ file in Photoshop and determine it's something
I MIGHT like to work with in the future I, merely, close the file.
That way, I've only done one true save [capture to record on
card] and a total of two compressions and have, thus, reduced the
artifacting down to minimum;

 

[Ferenc]

...8:1 compression. It depleats info on the Luminance channel (excessive, for my comfort).

Therefore, if you choose to shoot with .JPGs, you can safely baffle between 2.7-1 and 4:1. I use 4:1 as it is the has the best performance vs. quality compromise. That's it 5.25 secs of write time per shot, on the CF port, @ 2240x1680 pixels, and minimum compression artifacts in either Luminance and Chrominance channels.

For quality, I do not like the idea of TIFF since it is a waste of camera processing and writing time. Quiet unpractical. Therefore, RAW is probably the way to go, even at ISO 160 and 320, were it really shines (relative to JPEG or TIFF at same ISO speeds). Again, RAW boasts the best performance vs. quality ratio: 11.5 secs, 620 KBytes/sec, of write speed on the CF port, at full resolution. You later need to import in PS, which is the second chapter of this book.

Happy shooting,

Ferenc

P.S.: be careful with fast shooting in RAW, since your life will depend on that modest 32 Mbytes RAM buffer, ok?

I took several shots of the same subjecst in three different modes.
To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.
Just wondering if anybody else did the comparison and give me some
input.
Thanks.

 

[Jerry Monroe]

Eileen, Vance and Lawrence,

Thank you so much for taking the time to contribute to my education. My ardor for photography was rekindled about three weeks ago with the acquisition of an E-20 and I am hungry to learn anything I can that will help me make the best possible use of it. The subject matter of this thread, obviously, is very important.

Let me try to summarize what you have just taught me.

1. Low compression does, in fact, result in better images so
always shoot in the best mode that you can.

2. Never save over the original in an editor if it is something you
may use.

3. If you are going to use an image, first save it as a TFF and work
on that.

4. Protect your originals

5. Discard the images that you will never use

I am most grateful to you for sharing your expertise with me. My 256K CF just arrived to day so I will be able to follow rule one and I also plan to follow the other three rules. Good advice.

Jerry

 

[Jerry Monroe]

And thanks to you, Ferenc. RAW will be my next project.

Therefore, if you choose to shoot with .JPGs, you can safely baffle
between 2.7-1 and 4:1. I use 4:1 as it is the has the best
performance vs. quality compromise. That's it 5.25 secs of write
time per shot, on the CF port, @ 2240x1680 pixels, and minimum
compression artifacts in either Luminance and Chrominance channels.

For quality, I do not like the idea of TIFF since it is a waste of
camera processing and writing time. Quiet unpractical. Therefore,
RAW is probably the way to go, even at ISO 160 and 320, were it
really shines (relative to JPEG or TIFF at same ISO speeds).
Again, RAW boasts the best performance vs. quality ratio: 11.5
secs, 620 KBytes/sec, of write speed on the CF port, at full
resolution. You later need to import in PS, which is the second
chapter of this book.

Happy shooting,

Ferenc
P.S.: be careful with fast shooting in RAW, since your life will
depend on that modest 32 Mbytes RAM buffer, ok?

I took several shots of the same subjecst in three different modes.
To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.
Just wondering if anybody else did the comparison and give me some
input.
Thanks.

 

[eileen68597]

When shooting in .jpg record format, you capture the camera's full amount of set data and write it to the card. As a part of the write process this initial information is compressed into whichever .jpg format you've chosen to save it in. This is the first compression. When you open the image in an editor, the data decompresses. When you close the file directly or save it and then close it - regardless of whether or not you've made any alterations to the file - the data compresses again. This is the second compression.

Hope this makes sense.

eileen

 

[Joe-TN]

When shooting in .jpg record format, you capture the camera's full
amount of set data and write it to the card. As a part of the
write process this initial information is compressed into whichever .jpg format you've chosen to save it in. This is the first
compression. When you open the image in an editor, the data
decompresses. When you close the file directly or save it and
then close it - regardless of whether or not you've made any
alterations to the file - the data compresses again. This is the
second compression.

Hope this makes sense.

eileen

Eileen,

You may want to refigure that. If you close the file without saving it nothing is changed in the original file. I generally write-protect my originals immediately after downloding them from the camera. You can open and close them all day, no problem. But if you try a save from your image editor you'll be told you can't because the file is write protected.

Joe--Joe-TN http://www.pbase.com/joe_tn/shared_photos

 

[Don Daugherty]

I did a comparison. I used the same subject matter and the same size, only changing the compression. I was trying to determine what was going to be the best trade off for me in terms of storage. If I can shoot 150 pictures when I am out and about, I will. But, if I can shoot 300, then all the better, I will.

I used a Remy Martin box (bright red box, green bottle on the label with a large gold label on it with black lettering), an Eveready flashlight (odd blue hue with bright yellow foot on it, very smooth texture) and a few other miscelaneous small items in front of those two things. This was all set on a particleboard workbench.

My video is set at 1600x1200@32bit. I did a side by side comparison at 50% zoom with all three, as well as a 200% and 400%. Then I printed each as a 4x6 using QImage Pro on the same printer (HP 932C).

1:2.7, looked very good. Looked very true to the original.

1:4, very slight amount of abberations appearing in really complex patterns. Very difficult to see at all but the higher zoom. Didn't appear in the prints and they looked as good as the 1:2.7.

1:8, noticable noise in complex patterns (the particleboard workbench had kind of an orange dot pattern that formed on it, it was visible at all zoom levels), deterioration of colors was dramatic (the gold label on the box began to look dirty, the bright yellow on the flashlight also looked mottled and dirty). Problems were very apparent in the prints.

Those are my observations from a little ad-hoc test. It was very informative for me, hope it helps. I decided to go with 1:4 after my test. I am very happy with that compramise so far. I guess that's what it's all about, compramise. Good luck!

As always, just my two cents.

I took several shots of the same subjecst in three different modes.
To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.
Just wondering if anybody else did the comparison and give me some
input.
Thanks.

 

[excal]

Many moons ago I got my E10 and played with different settings. 1:8 as everyone can see but the totally blind, does suffer the 8x8 blocks and compression artifacts, but it's the quickest the E10 can offer and uses little space, so I guess it has it's place (Jaja uses it, so there's some proof to start with).

TIFF vs. RAW - besides the inability to zoom in and check stuff on the camera, and the hassle of importing it later, RAW is smaller and better than TIFF, so that's clearly no. 1 quality/speed etc.

So, with TIFF, RAW and HQ taken out, it boils down to SHQ and HQ (1:2.7 and 1:4 respectively) Now, when I did my tests some time ago, I settled on 1:4 as I really couldn't see the difference. But experience changes the eyes. Until recently when Ferenc bought it to my attention, I didn't realise in my early tests that RAW was soooo much cleaner than JPG. (is that also true if I said cleaner than TIF?)

Anyway, I got to thinking. It's all very well comparing compressions for artifacts and blockiness, but what about the Z dimension? ie. dynamic range. JPEG throws much colour information away to get the compression it does (the human eye isn't so sensitive to colour as luminance apparently). So, Ferenc, do you know... does 1:2.7 give any significant post processing dynamic range advantage? In other words, latitude for incorrect exposure.

--Excal

 

[ME-GM]

Exactly what I am looking for: some simple test like that for me to compare.
Thanks a million. Your two cents worth a whole lot more for me.

I used a Remy Martin box (bright red box, green bottle on the label
with a large gold label on it with black lettering), an Eveready
flashlight (odd blue hue with bright yellow foot on it, very smooth
texture) and a few other miscelaneous small items in front of those
two things. This was all set on a particleboard workbench.

My video is set at 1600x1200@32bit. I did a side by side
comparison at 50% zoom with all three, as well as a 200% and 400%.
Then I printed each as a 4x6 using QImage Pro on the same printer
(HP 932C).

1:2.7, looked very good. Looked very true to the original.

1:4, very slight amount of abberations appearing in really complex
patterns. Very difficult to see at all but the higher zoom.
Didn't appear in the prints and they looked as good as the 1:2.7.

1:8, noticable noise in complex patterns (the particleboard
workbench had kind of an orange dot pattern that formed on it, it
was visible at all zoom levels), deterioration of colors was
dramatic (the gold label on the box began to look dirty, the bright
yellow on the flashlight also looked mottled and dirty). Problems
were very apparent in the prints.

Those are my observations from a little ad-hoc test. It was very
informative for me, hope it helps. I decided to go with 1:4 after
my test. I am very happy with that compramise so far. I guess
that's what it's all about, compramise. Good luck!

As always, just my two cents.

I took several shots of the same subjecst in three different modes.
To me, I have to blow it up real big (600 to 800 percent) to see
some differences but nothing real major.
Just wondering if anybody else did the comparison and give me some
input.
Thanks.