Testing which to use: DNG or TIFF?

[Flycaster]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

 

[ishwanu]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit)

16-bit would be a lot better for further editing.

over to PSE,

Which specific things do you do in PSE rather than in PL?

and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL.

The answer probably depends on what you normally do in PSE. As stated in another concurrent thread here, DNG offers more flexibility if you want or need it, Quoted from DxO documentation:

'Depending on its intended purpose, a Linear DNG might also complete other parts of the RAW conversion process such as denoising and lens corrections. You can then take this optimized file into your photo-editing software of choice to complete the creative process — everything from white balance and exposure through to tone curves and color grading — with the knowledge that your final image will be the highest possible quality.'

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways from PhotoLab and try your most extreme PSE modifications on both file types. See which you prefer.

 

[NAwlins Contrarian]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

I would export from PL as a 16-bit TIFF in ProPhoto RGB, which provides as much quality as there is to be had. Whether the extra quality matters visually depends on your image and what further processing you do in PSE--but you say, "Storage is not a problem," so I don't see a downside.

Any 8-bit file, TIFF or other format, can lose some ability to edit without risk of problems. Often it doesn't matter, but sometimes it does.

If your final use is web posting or uploading to commercial sites for sharing or printing, then you may want to convert from ProPhoto RGB to sRGB, but presumably you can do that as a last step with exporting from PSE.

FWIW, Adobe intended DNG to be a universal format, but it hasn't truly been accepted as such. TIFF really is an almost-universal format. Also, DNG is merely a wrapper; it can contain almost-raw, quasi-raw, or non-raw image data, depending on how it is created.

 

[Flycaster]

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways and try your most extreme PSE modifications on both file types. See which you prefer.

That was what I was thinking. But, as I started doing it, I started to get confused on just how to keep things equal.

1. Start out with a good RAW, do just basic editing with PL (nothing fancy.)

2. Export to PSE

a. Export as 8 bit TIFF, which goes directly into PSE's Expert mode

b. Export as a DNG, which goes directly into PSE's ARC

Confusion: Should I do anything in ARC before sending to PSE? I think I should do two things:

1) Not do anything and click Done

2) Adjust histogram to two black triangles

3. Send 1) and 2) to PSE and save as JPG

4. View all

 

[ishwanu]

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways and try your most extreme PSE modifications on both file types. See which you prefer.

That was what I was thinking. But, as I started doing it, I started to get confused on just how to keep things equal.

1. Start out with a good RAW, do just basic editing with PL (nothing fancy.)

2. Export to PSE

a. Export as 8 bit TIFF, which goes directly into PSE's Expert mode

Oh, I forgot that PSE doesn't natively support 16 bits. That's unfortunate.

b. Export as a DNG, which goes directly into PSE's ARC

That's ACR.

Confusion: Should I do anything in ARC ACR before sending to PSE?

Depends on what changes you intend to make in PSE.

I think I should do two things:

1) Not do anything and click Done

2) Adjust histogram to two black triangles

3. Send 1) and 2) to PSE and save as JPG

4. View all

For testing, I think I would pass the DNG through ACR with few if any changes - mainly white balance adjustments to match the TIFF if they don't already match. After white balance is established, there isn't a lot of difference between the two types.

On the other hand ... if ACR works in 16-bit mode rather than PSE's 8-bit mode, it could be advantageous to do as much as possible there first.

 

[Digital Nigel]

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways and try your most extreme PSE modifications on both file types. See which you prefer.

That was what I was thinking. But, as I started doing it, I started to get confused on just how to keep things equal.

1. Start out with a good RAW, do just basic editing with PL (nothing fancy.)

2. Export to PSE

a. Export as 8 bit TIFF, which goes directly into PSE's Expert mode

b. Export as a DNG, which goes directly into PSE's ARC

Confusion: Should I do anything in ARC before sending to PSE? I think I should do two things:

1) Not do anything and click Done

2) Adjust histogram to two black triangles

3. Send 1) and 2) to PSE and save as JPG

4. View all

Just curious: what do you do in PSE that you couldn’t have done in PL? Most PL users make very limited, if any, use of a pixel editor after PL. After all, the more that you can do in the non-destructive editing world of PL the better.

 

[Flycaster]

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways and try your most extreme PSE modifications on both file types. See which you prefer.

That was what I was thinking. But, as I started doing it, I started to get confused on just how to keep things equal.

1. Start out with a good RAW, do just basic editing with PL (nothing fancy.)

2. Export to PSE

a. Export as 8 bit TIFF, which goes directly into PSE's Expert mode

Oh, I forgot that PSE doesn't natively support 16 bits. That's unfortunate.

b. Export as a DNG, which goes directly into PSE's ARC

That's ACR.

Confusion: Should I do anything in ARC ACR before sending to PSE?

Depends on what changes you intend to make in PSE.

I think I should do two things:

1) Not do anything and click Done

2) Adjust histogram to two black triangles

3. Send 1) and 2) to PSE and save as JPG

4. View all

For testing, I think I would pass the DNG through ACR with few if any changes - mainly white balance adjustments to match the TIFF if they don't already match. After white balance is established, there isn't a lot of difference between the two types.

I don't think that having the dng match the tiff makes for a test. But, I do think that dng may be better in that when coming from PL into ACR, the histogram needs some balancing. Whereas, the tiff goes directly into PSE. Could be that the histogram balancing may make the dng a little better? Or, should I try to put PL's tiff into ACR? To do this, I think that it will take extra steps: Save tiff in a folder and the choose to edit in ACR by selecting image from PSE File?

On the other hand ... if ACR works in 16-bit mode rather than PSE's 8-bit mode, it could be advantageous to do as much as possible there first.

Yeah, but as you already know, PSE can't handle 16-bit.

 

[Flycaster]

Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Export both ways and try your most extreme PSE modifications on both file types. See which you prefer.

That was what I was thinking. But, as I started doing it, I started to get confused on just how to keep things equal.

1. Start out with a good RAW, do just basic editing with PL (nothing fancy.)

2. Export to PSE

a. Export as 8 bit TIFF, which goes directly into PSE's Expert mode

b. Export as a DNG, which goes directly into PSE's ARC

Confusion: Should I do anything in ARC before sending to PSE? I think I should do two things:

1) Not do anything and click Done

2) Adjust histogram to two black triangles

3. Send 1) and 2) to PSE and save as JPG

4. View all

Just curious: what do you do in PSE that you couldn’t have done in PL? Most PL users make very limited, if any, use of a pixel editor after PL. After all, the more that you can do in the non-destructive editing world of PL the better.

For me, I find that the healing brush in PSE is better, DeFine seems to add a little more controllable noise reduction (at times) and using cropping in PSE easier. Also, I'll use NIK while in both apps.

 

[Flycaster]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

I would export from PL as a 16-bit TIFF in ProPhoto RGB, which provides as much quality as there is to be had. Whether the extra quality matters visually depends on your image and what further processing you do in PSE--but you say, "Storage is not a problem," so I don't see a downside.

Unfortunately, PSE can't handle 16-bit.

Any 8-bit file, TIFF or other format, can lose some ability to edit without risk of problems. Often it doesn't matter, but sometimes it does.

Not being a professional, this won't matter to me.

If your final use is web posting or uploading to commercial sites for sharing or printing, then you may want to convert from ProPhoto RGB to sRGB, but presumably you can do that as a last step with exporting from PSE.

FWIW, Adobe intended DNG to be a universal format, but it hasn't truly been accepted as such. TIFF really is an almost-universal format. Also, DNG is merely a wrapper; it can contain almost-raw, quasi-raw, or non-raw image data, depending on how it is created.

 

[NAwlins Contrarian]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

I would export from PL as a 16-bit TIFF in ProPhoto RGB, which provides as much quality as there is to be had. Whether the extra quality matters visually depends on your image and what further processing you do in PSE--but you say, "Storage is not a problem," so I don't see a downside.

Unfortunately, PSE can't handle 16-bit.

Hmm. I haven't had a version of PSE since version 14 or 2014, whichever it was called*, and it's never been a regular part of my workflow. But I was under the impression that some PSE operations could be performed at 16-bit precision, and over time that group had expanded. Maybe I'm wrong, but you might want to double-check.

Any 8-bit file, TIFF or other format, can lose some ability to edit without risk of problems. Often it doesn't matter, but sometimes it does.

Not being a professional, this won't matter to me.

IME / IMO the dividing line about whether this matters isn't anywhere near / has nothing to do with "professional" versus "amateur". It mostly concerns making more than slight chances to lightness, color, and/or contrast, especially in the shadows.
*It was a special Costco deal, I think like $39.99, and I was curious enough to try it.

 

[Flycaster]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

I would export from PL as a 16-bit TIFF in ProPhoto RGB, which provides as much quality as there is to be had. Whether the extra quality matters visually depends on your image and what further processing you do in PSE--but you say, "Storage is not a problem," so I don't see a downside.

Unfortunately, PSE can't handle 16-bit.

Hmm. I haven't had a version of PSE since version 14 or 2014, whichever it was called*, and it's never been a regular part of my workflow. But I was under the impression that some PSE operations could be performed at 16-bit precision, and over time that group had expanded. Maybe I'm wrong, but you might want to double-check.

2x checked, no 16-bit.

Any 8-bit file, TIFF or other format, can lose some ability to edit without risk of problems. Often it doesn't matter, but sometimes it does.

Not being a professional, this won't matter to me.

IME / IMO the dividing line about whether this matters isn't anywhere near / has nothing to do with "professional" versus "amateur". It mostly concerns making more than slight chances to lightness, color, and/or contrast, especially in the shadows.

*It was a special Costco deal, I think like $39.99, and I was curious enough to try it.

 

[Flycaster]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

 

[Flycaster]

I shoot RAW and use PhotoLab7 Elite, Photoshop Elements 2020 and the Google NIK. Viewed on Asus PA278QV.

My general workflow has been PL first, then send a tiff (8-bit) over to PSE, and finally save as a jpg. I have done a little reading on dng vs tiff, and have been mulling whether I should change over to dng from PL. Storage is not a problem and my calibrated monitor is almost 100% on sRGB. Perhaps, you could suggest a way that I could test the outcome (jpg) differences, if any, on my equipment? Thanks.

Having nothing to do today, I figured that I would run a visual outcome test of whether or not I should export from PL to PSE using DNG or TIF. I opened a single RAW image in PL, did some slight editing and made three exports of it to PSE (with all corrections.) One export was as an 8-bit TIF (PSE doesn't do 16-bit), which went directly into PSE; and two exports of DNG because DNG is exported directly into PSE's ACR (no editing in ACR as the histo looked good, two black triangles,) as I wasn't sure whether saving from ACR with Done or Open would make a differences. As it turned out, there was no difference in their PSE images, so I continued with using the Open image for the test. So, I now had images of the same subject, one DNG and one TIF. Each one was saved twice as JPGs. Once without doing any editing, and once with Enhancement (Auto Smart Fix, which I use with minor adjustments, but no adjustment to ASF here.) All in all, I now had four images to compare.

Assuming that my testing methodology has some legitimacy, there really was hardly any visual difference between the images, although I think I prefer the DNG derived images. To my eye they appear to have very slightly more "oomph." However, for those of you who might like to make some judgment/comment on the test, here are the images:

TIF/JPG, Not enhanced, 7.59MB

TIF/JPG, Enhanced, 8.60MB

DNG/JPG, Not enhanced, 7.82MB

DNG/JPG, Enhanced, 8.72MB

 

[Flycaster]

Having nothing to do today, I figured that I would run a visual outcome test of whether or not I should export from PL to PSE using DNG or TIF. I opened a single RAW image in PL, did some slight editing and made three exports of it to PSE (with all corrections.) One export was as an 8-bit TIF (PSE doesn't do 16-bit), which went directly into PSE; and two exports of DNG because DNG is exported directly into PSE's ACR (no editing in ACR as the histo looked good, two black triangles,) as I wasn't sure whether saving from ACR with Done or Open would make a differences. As it turned out, there was no difference in their PSE images, so I continued with using the Open image for the test. So, I now had images of the same subject, one DNG and one TIF. Each one was saved twice as JPGs. Once without doing any editing, and once with Enhancement (Auto Smart Fix, which I use with minor adjustments, but no adjustment to ASF here.) All in all, I now had four images to compare.

Assuming that my testing methodology has some legitimacy, there really was hardly any visual difference between the images, although I think I prefer the DNG derived images. To my eye they appear to have very slightly more "oomph." However, for those of you who might like to make some judgment/comment on the test, here are the images:

TIF/JPG, Not enhanced, 7.59MB

View: original size

TIF/JPG, Enhanced, 8.60MB

View: original size

DNG/JPG, Not enhanced, 7.82MB