Topics

moderated Case Study Announcement: Hotel Lobby


Dan Margulis
 

We launch our series of ten case studies, with an eleventh optional one. This one, sort of a warmup, is one of the two easiest of the 11, IMHO.
 
Dan
***********
 
*This is one of 5,000 images taken from an MIT study. for which permission is given for educational use and which contributes four challenges to our series of ten. The sponsors solicited contributions of images that were supposed to represent the range that a professional retoucher might receive. They then hired five intermediate retouchers to correct each. The 25,000 resulting files were made public, as were the .dng files they started with. The best of these efforts will be posted along with those of our group.
 
*In the study, no instructions were given as to what the client wanted. Here, let's assume it is promotional for a certain hotel, showing how tasteful their interior design is.
 
*You can use whatever methods you like to improve the picture. In previous case studies we were not supposed to use Unsharp Mask, because the MIT retouchers weren't permitted to. But inasmuch as we demonstrated that their work was not generally competitive with ours, there's no longer a need for any such limitation.
 
*Please keep clear records of what you did for discussion. List members find these very valuable.
 
*In the Photos section, Case Study: Hotel Lobby, 
https://groups.io/g/colortheory/album?id=259709
I have uploaded a version opened with Camera Raw defaults, and another where the settings were much flatter. You may use either, or fetch the .dng as below.
 
*groups.io does not allow .dng format in the Photos section. If you want the .dng, you must download a zipped file from the Files section, https://groups.io/g/colortheory/files. NOTE: the zipped file contains the two default images as well, you don't need to download them separately. DO NOT USE THE THUMBNAIL IMAGE ATTACHED TO THIS POST.
Filename=2021_Hotel-Lobby_case-study_source.zip 
 
*The designated size of this exercise is 2000 x 3006 pixels. If you use the .dng image be sure to open into the correct size. Do not crop, rotate, or alter the sizing, and don't delete any objects, because any of these things will make it impossible to use your version as part of a par assembly. Also, we have found that Lightroom was not acquiring these .dngs in exactly the same aspect as other modules due to a lens correction. Therefore, we recommend that as soon as you acquire, you apply one of the default versions to it to make sure that all pixels line up.
 
*Your final file is to be sRGB with a proper tag. If you work in a different colorspace you must Edit: Convert to Profile>sRGB before submitting the file.
 
*When finished, save in JPEG form, quality level 9. E-mail it to me, dmargulis (at) aol.com, with your brief explanation of how you produced it. DO NOT POST IMAGES TO THE LIST.
 
*Remember that some e-mail clients automatically downsize image attachments. Make sure you’re sending it to me at the original size.
 
*Entries close Monday morning, 1 February, at 06:00 Eastern/1100Z/12:00 ora italiana.
 
*Rather than confirm every entrant I've received, I will periodically post the initials of everyone whose file I have.
 
*As soon as convenient after the deadline, I'll post all the entrants in a random order. Names will not be revealed. Once discussion begins, those who wish to identify themselves publicly may do so.
 
*If the filename of your version contains your name or initials, they will be deleted before posting. If you want to be able to identify your version by name once posted, throw in a few extra letters; I won't delete them.
 
*I will leave discussion of the results to the group for the first two days after posting, and will then weigh in with my own opinion.
 
Dan Margulis



Dan Margulis
 


On Jan 24, 2021, at 7:23 AM, Dan Margulis via groups.io <dmargulis@...> wrote:

 
*The designated size of this exercise is 2000 x 3006 pixels. If you use the .dng image be sure to open into the correct size. Do not crop, rotate, or alter the sizing, and don't delete any objects, because any of these things will make it impossible to use your version as part of a par assembly.

Having received an inquiry offline, I respond here in case other members of the group have the same question.

When I say “don’t delete any objects” I mean tangible objects, not the specular highlights on the wall that my correspondent was asking about.

If the object is intangible then one interpretation can be blended seamlessly with another. I imagine there will be various treatments of these specular, ranging from ignoring them to erasing them outright. If they are blended into a par version nobody will know that anything’s wrong.

A tangible object, like the surplus leaves apparently coming in from another plant on the left, cannot be deleted even if you think it would be helpful, because then if your version were blended with that of somebody else who had not deleted them, the result would look ridiculous.

Dan


Paco
 

On Sun, Jan 24, 2021 at 08:23 AM, Dan Margulis wrote:
https://groups.io/g/colortheory/files
Hi! The DNG file measured 2832 x 4248 when I opened it. 


Jim Sanderson
 

Paco, change your opening settings from your raw converter to the correct dimensions.  That way it'll open to those dimensions in PS.  By default, the opening settings in Bridge and Lightroom is the current document size.  You can change that to any dimension you want.  

Jim Sanderson 


-----Original Message-----
From: Paco <paco@...>
To: colortheory@groups.io
Sent: Mon, Jan 25, 2021 8:00 am
Subject: Re: [colortheory] Case Study Announcement: Hotel Lobby

On Sun, Jan 24, 2021 at 08:23 AM, Dan Margulis wrote:
https://groups.io/g/colortheory/files
Hi! The DNG file measured 2832 x 4248 when I opened it. 


Kenneth Harris
 

I've tried ON1, Iridient, Affinity, and Camera Raw, and they all seem to think it's 12mpx, 2832x4235.   It's shot on a Nikon D700, that's the spec.
Ken Harris


Kenneth Harris
 

Pardon, 2832x4256.  FileViewer also agrees.  
Ken Harris


Dan Margulis
 



On Jan 25, 2021, at 1:48 PM, Kenneth Harris <reg@...> wrote:

Pardon, 2832x4256.  FileViewer also agrees.  
Ken Harris

OK, that solves half the mystery, but there’s more.

When this was originally given to the MIT group the size was indeed 2832x4256. I know this for a fact because I participated and opened the same .dng that they did.

For the purpose of this group, I wanted a smaller filesize, so I specified 2000x3006, which is the same aspect. When I re-opened the .dng in Camera Raw to make the default versions, I just typed in 2000 px short direction and got the same result.

Anyone who has a size of 2832x4256, all you have to do is Image: Image Size, choose Resample and Constrain Proportions, and type in 2000.

Last time, we may recall, certain people were acquiring the image in the wrong dimensions because Lightroom (I think it was Lightroom) was applying some kind of lens correction that distorted the size. I forget the solution, but there was one. Therefore, since that time I’ve put a notice in the case study announcement that people need to check their acquisition size against the default if there is any doubt.

Since Ken has 2832x4256 he’s out of the woods. BUT:

Paco reports 2832x4248, without saying in what module.

Then, in view of Ken saying he opened it at a different size in Affinity, I moved the .dng over to where I have that application. And before I tried to open it, from the Finder I did a File Info. Hallelujah! 2832x4256 just like I hoped.

But, as the devil would have it, five seconds later Affinity opened it as 2844 x 4284. That would reduce to 2000x3012.6. When I forced it anamorphically to 2000x3006, it was pretty close, but not close enough for blending into a par.

So apparently there is some hocus-pocus going on by default that needs to be corrected. The suspect is a lens correction, but everybody’s system is likely different. I don’t know what else to say except be careful.

Thanks for these reports.

Dan





Gerald Bakker
 

On Mon, Jan 25, 2021 at 08:51 PM, Dan Margulis wrote:
Last time, we may recall, certain people were acquiring the image in the wrong dimensions because Lightroom (I think it was Lightroom) was applying some kind of lens correction that distorted the size. I forget the solution, but there was one.
I bet the solution is to switch off any lens corrections that may have been applied automatically (as configured in a lens profile), or manually.
For an automatic correction, switching off the "Enable Profile Corrections" checkbutton in the Lens Correction panel may do the trick.
--
Gerald Bakker
http://geraldbakker.nl


Ronny Light
 

In Bridge—I don’t use Lightroom—you can open an image in Photoshop at its size or choose to resize.

 

 

Ronny

www.RonnyLightPhoto.com

5010 B Wilkerson Dr., Nashville, TN 37211

 

 

 

From: colortheory@groups.io <colortheory@groups.io> On Behalf Of Gerald Bakker
Sent: Monday, 25 January, 2021 2:20 PM
To: colortheory@groups.io
Subject: Re: [colortheory] Case Study Announcement: Hotel Lobby

 

On Mon, Jan 25, 2021 at 08:51 PM, Dan Margulis wrote:

Last time, we may recall, certain people were acquiring the image in the wrong dimensions because Lightroom (I think it was Lightroom) was applying some kind of lens correction that distorted the size. I forget the solution, but there was one.

I bet the solution is to switch off any lens corrections that may have been applied automatically (as configured in a lens profile), or manually.
For an automatic correction, switching off the "Enable Profile Corrections" checkbutton in the Lens Correction panel may do the trick.
--
Gerald Bakker
http://geraldbakker.nl

_._,_._,_

 


Ronny Light
 

I opened the file, Hotel Lobby a3500-default.jpg, into Photoshop, through Dropbox, without resizing, and the file in Photoshop is 2000 x 3006 pixels, just as Dan said.

 

 

Ronny

www.RonnyLightPhoto.com

5010 B Wilkerson Dr., Nashville, TN 37211

 

 

 

From: colortheory@groups.io <colortheory@groups.io> On Behalf Of Kenneth Harris
Sent: Monday, 25 January, 2021 12:36 PM
To: colortheory@groups.io
Subject: Re: [colortheory] Case Study Announcement: Hotel Lobby

 

I've tried ON1, Iridient, Affinity, and Camera Raw, and they all seem to think it's 12mpx, 2832x4235.   It's shot on a Nikon D700, that's the spec.
Ken Harris

_._,_._,_

 


Paco
 

Posted just as a heads up. In PS it can be downsized without any problem an it matches the jpegs perfectly.

Paco


Dan Margulis
 



On Jan 24, 2021, at 7:23 AM, Dan Margulis via groups.io <dmargulis@...> wrote:

We launch our series of ten case studies, with an eleventh optional one. This one, sort of a warmup, is one of the two easiest of the 11, IMHO.
 
During these case studies entries have been known to go astray, or to be submitted at the wrong size etc. Therefore my practice has been to post a herd confirmation roughly 48 hours ahead of time, so that if there is a problem with your entry, or it wasn’t received by me, there’s time to make amends. Since there are more than a dozen entries already, I’ll post now, and then again Sunday morning.

A reminder that entries are due in this case study in about 72 hours, at 06:00 eastern time Monday/1100Z/12:00 ora italiana

I confirm receipt of entries from the following individuals as of 06:00 eastern time this morning:

BB
GB
RB
KC
MC
HD
JaG
RG*
PM
JP
DS
KS
LV

*indicates that a corrected version was submitted

Entries from the following were at an incorrect size/cropping and would have to be resubmitted:

None!


Dan Margulis


Dan Margulis
 



On Jan 29, 2021, at 6:29 AM, Dan Margulis via groups.io <dmargulis@...> wrote:

During these case studies entries have been known to go astray, or to be submitted at the wrong size etc. Therefore my practice has been to post a herd confirmation roughly 48 hours ahead of time, so that if there is a problem with your entry, or it wasn’t received by me, there’s time to make amends. Since there are more than a dozen entries already, I’ll post now, and then again Sunday morning.

Make that Saturday morning, because an unreasonable number (four) of entries have come in with an incorrect format. I have fixed three and the fourth has been resubmitted correctly, so no further action is required. I request, however, that more attention be paid to these details. The issues so far:

1) file size reduced by e-mail client, requiring that I contact the sender to get a full-size copy.

2) file size at 2832x4256 rather than 2000x3006.

3) file submitted in Adobe RGB rather than sRGB.

4) file submitted in ProPhoto RGB and saved as JPEG level 12, resulting in a filesize three times as large as anyone else's.

A reminder that entries are due in this case study in about 48 hours, at 06:00 eastern time Monday/1100Z/12:00 ora italiana

I confirm receipt of entries from the following individuals:

BB
EB
GB
ReB
RoB
KC
MC
HD
JaG
RG*
HH
KH
BI
PM
JP
DS
JS
KS
LV
FY

*indicates that a corrected version was submitted

Entries from the following were at an incorrect size/cropping and would have to be resubmitted:

None

Dan Margulis


Doug Schafer
 

Dan,

Can you comment on Hotel Lobby submissions: what would be your max. file size expected .jpg, so we can assess our submissions to not exceed file size submitted?...and to avoid too large files in future?

When next entries (next weeks, different images) are due, can you estimate max. file size to submit; so we can check before we submit?

Doug Schafer


Dan Margulis
 



On Jan 30, 2021, at 9:52 AM, k_d@... wrote:

Dan,

Can you comment on Hotel Lobby submissions: what would be your max. file size expected .jpg, so we can assess our submissions to not exceed file size submitted?...and to avoid too large files in future?

When next entries (next weeks, different images) are due, can you estimate max. file size to submit; so we can check before we submit?

There’s no way to give exact file sizes because, among other things everybody uses different levels of sharpening, which is difficult to compress in JPEG methodology and results in larger sizes.  Therefore it wouldn’t do to examine the files sizes of the unsharpened default versions because it wouldn’t be surprising if our own work took up two or even three times as much space.

And that is the case here. The defaults are around 850k. All of our versions are larger, typically around 1.2mb with one as high as 2.5.

File storage is cheap. Nevertheless we do have certain limitations, which is why we work on files that are smaller than the MIT retouchers had. I don’t see why we need higher resolutions for the purpose of this group and I don’t see the need to save files that take up ten times as much space as the defaults.

Dan


Dan Margulis
 


On Jan 29, 2021, at 6:29 AM, Dan Margulis via groups.io <dmargulis@...> wrote:

Since there are more than a dozen entries already, I’ll post now, and then again Sunday morning.

OK, this is the final confirmation before posting. If you submitted an entry before 06:00 eastern time today, you should find your initials here. The deadline for receipt of others is 06:00 easters/1100Z/12:00 ora italiana tomorrow.

I confirm receipt of entries from the following individuals:

BB
EB
GB
ReB
RoB
KC
MC
HD
JaG
RG*
HH
KH
BI
SJ
PM
SN
JP
DR
DS
JS
KS
RT
LV
FY
CZ

*indicates that a corrected version was submitted

Entries from the following were at an incorrect size/cropping and would have to be resubmitted:

JC

Dan Margulis


John Furnes
 

Hi everybody,

This was bit more difficult than it seemed.

Mine is #119.

As several others have commented, finding the correct colour value was difficult. There are several light sources – the yellow spots reflected on the wall, and the light emanating from under the ventilation, and then there is the reflection from a TV and perhaps a window, also reflected. Finally there is white light coming from above.

All this put me out, and I struggled with the greenery and took a reading from the ventilation shaft to make some sort of colour correction.  In the greenery I wound up with 45;(31);(34) in the petals of the flower pot– a tad too dark, I can now see. Having adjusted the light a bit to +63 (Ctrl+L adjusting midtones), it looks much better.

 

I did not try to hide/ correct the reflections on the wall, as I perceived this to be about colour.

I think #109 is better than #142 as the par-version to me looks a bit too red.

 

Kind regards

John Furnes

 

tel. +45 2917 9927

Sitz ApS – Kildegade 26 - 8700 Horsens – Denmark - www.sitz.dk

 

NOTICE: This e-mail and its associated attachments are confidential and privileged and intended

solely for the use of the addressee. Any use, retransmission, storage, or publication by persons 

other than the addressee is prohibited. If you received this transmission and are not the intended

recipient, please notify the sender immediately, then discard and erase all copies of the 

transmission and attachments.