Rob de Loe

Rob de Loe

Lives in Canada Canada
Joined on Jul 11, 2019

Comments

Total: 50, showing: 1 – 20
« First‹ Previous123Next ›Last »
In reply to:

pannumon: That's not a car. That's a Lada.

And you could start them with a crank! You had to drill a hole through the bumper, but with that small mod you never had to fear a dead battery again. ;)

Link | Posted on May 15, 2021 at 02:48 UTC
On article Why are modern 50mm lenses so damned complicated? (875 comments in total)
In reply to:

pollup: #1 you don't seem to understand how DOF works ... if you're close like 1m away from the subject, then yes you will have a 3cm DOF, but if you're 2m from the subject, the DOF becomes 11cm, so suddenly much deeper, and if you're 6m away, then you have a 1m DOF, so that's not shallow at all!

Wait, @pollup, did you just tell Roger Cicacla that he doesn't understand how depth of field works? That's cute.

Link | Posted on May 10, 2021 at 21:03 UTC
On article Why are modern 50mm lenses so damned complicated? (875 comments in total)

Well, that explains that!

That was both clear and useful. As others have said, more please.

Link | Posted on May 9, 2021 at 14:05 UTC as 199th comment
In reply to:

JE River: I've found that gear does matter, very much. But, I see it mattering in the other direction than the typical "full frame, high MP, crazy sharp lenses" trope that seems to always be applied to landscape photography.

I've dumped all of my full frame gear, costing thousands, in favor of APS-C mirrorless, action cameras, and my smartphone. I shoot 100% landscape photos. What I have now is a very compact and portable camera system that can cover every type of shot I encounter (besides ultra-long tele shots, which I don't like much anyway) in both daylight and night. An APS-C camera, with a $200 sharp prime lens, can blow the doors off any MF camera ever made if one is inclined to use stitching of multiple frames of the APS-C to match or exceed single exposures from MF. Many smartphones are effectively water/dust proof and are an only option in adverse conditions that would destroy an ILC in minutes or seconds. Same with action cameras, but utilizing wider FOV than the smartphone.

My previous post clearly didn't come across well because I was agreeing with you! And I agree with everything you wrote in your latest point. If we were in a pub, we'd have had this sorted ages ago and be on to more interesting conversations about photography. Alas... forums.

I don't have a strong desire to print more than 17" wide, and carefully made APS-C files are already more than good enough for that size. For a couple years I did all my work with a Fuji X-T2, a tilt-shift adapter and a set of Olympus OM lenses. It was an excellent, light outfit. I didn't switch to a GFX camera because image quality wasn't acceptable. Rather, I wanted more movements than tilt-shift adapters allow. A digital view camera and a GFX 50R gives me that. MF 33x44 just happens to be in the sweet spot for sensor size versus image circle of available wide angle lenses; a 50R is the only camera I can mount easily on my Toyo. This approach is a good fit for how I want to work.

Link | Posted on May 6, 2021 at 14:15 UTC
In reply to:

JE River: I've found that gear does matter, very much. But, I see it mattering in the other direction than the typical "full frame, high MP, crazy sharp lenses" trope that seems to always be applied to landscape photography.

I've dumped all of my full frame gear, costing thousands, in favor of APS-C mirrorless, action cameras, and my smartphone. I shoot 100% landscape photos. What I have now is a very compact and portable camera system that can cover every type of shot I encounter (besides ultra-long tele shots, which I don't like much anyway) in both daylight and night. An APS-C camera, with a $200 sharp prime lens, can blow the doors off any MF camera ever made if one is inclined to use stitching of multiple frames of the APS-C to match or exceed single exposures from MF. Many smartphones are effectively water/dust proof and are an only option in adverse conditions that would destroy an ILC in minutes or seconds. Same with action cameras, but utilizing wider FOV than the smartphone.

In your original post you referred to "matching" a single exposure from a MF camera. To me, that means stitching several APS-C frames to produce an image equivalent to one made with a MF sensor.

But from your follow-up it sounds like you're just saying people can shoot a lot of frames and stitch them together to make a composite that has more pixels than what a single 33mm x 44mm frame can produce. Sure. With a GigapixelCam X80 you can make 80,000 MP images. https://www.earthcam.net/projects/empirestatebuilding/gigapixelpanorama/2021/

Sensor size doesn't particularly matter in that scenario. With enough frames you can manage noise, dynamic range, and apparent depth of field. You're not getting around subject movement, but maybe that's not a concern in your work.

Thanks, but that's not my cup of tea. I'd rather see what I'm going to get in the viewfinder, and get what I want with camera movements -- in one frame. Each to their own.

Link | Posted on May 6, 2021 at 12:24 UTC
In reply to:

JE River: I've found that gear does matter, very much. But, I see it mattering in the other direction than the typical "full frame, high MP, crazy sharp lenses" trope that seems to always be applied to landscape photography.

I've dumped all of my full frame gear, costing thousands, in favor of APS-C mirrorless, action cameras, and my smartphone. I shoot 100% landscape photos. What I have now is a very compact and portable camera system that can cover every type of shot I encounter (besides ultra-long tele shots, which I don't like much anyway) in both daylight and night. An APS-C camera, with a $200 sharp prime lens, can blow the doors off any MF camera ever made if one is inclined to use stitching of multiple frames of the APS-C to match or exceed single exposures from MF. Many smartphones are effectively water/dust proof and are an only option in adverse conditions that would destroy an ILC in minutes or seconds. Same with action cameras, but utilizing wider FOV than the smartphone.

"An APS-C camera, with a $200 sharp prime lens, can blow the doors off any MF camera ever made if one is inclined to use stitching of multiple frames of the APS-C to match or exceed single exposures from MF."

Now you're just saying silly things.

I use both systems (Fuji APS-C and Fuji GFX), and I've actually done the testing. With flat-stitching, I can create a file using a an APS-C camera that matches the resolution of a GFX 50R file, and I can even do it with the same lens for both cameras. You need to be able to combine rise/fall and shift, and to do that you need a technical camera. It's not simple. The resulting file is indistinguishable -- but it's also a huge pain to create, and certainly not something anyone serious about working in environments where things move (e.g., landscape...) would do.

You can do amazing landscape photography with APS-C equipment. It's not a contest, and you don't need to make outlandish claims.

Link | Posted on May 5, 2021 at 19:03 UTC
On article Fujifilm GFX 100S review (927 comments in total)
In reply to:

slhimager: I am pretty much sold on the camera for my uses, but I never see any mention of how much image post processing is affected. My guess is that you start out with a 200 megapixel raw file before anything is adjusted in post, so a computer like a late model iMac may be heavily burdened.

@slhimager, you could download some RAW samples and see for yourself. My machine is old(er) now, and does fine with 50R files. I tried a few 100S files and haven't noticed a significant slow down.

Link | Posted on Apr 29, 2021 at 23:40 UTC
On article Fujifilm GFX 100S review (927 comments in total)
In reply to:

RJGee: Do you guys know ANYONE who doesn't want the best color and sharpness?
I've never met a serious shooter who said, oh, bad color and muddy sharpness?
I'll take it!
So, what your review says is, other than better Image quality, great sharpness the Fuji isn't better than it's competition.
I guess Fuji doesn't bribe enough for reviews.

A bizarre and offensive post RJGee. Kudos to Richard Butler for responding like a professional.

Link | Posted on Apr 29, 2021 at 23:34 UTC

A little shout-out for Darin Boville's blog, which has this interesting piece on Ansel Adams. From the story,

"Throughout his life Ansel made money by shooting commercial work, always worried about the bills. It was only in the last years of his life when his fame—and print prices—reached the point where he could rely solely on his artwork for an income."

https://www.abiggercamera.com/2020/10/12/was-ansel-adams-a-landscape-photographer/

Link | Posted on Apr 27, 2021 at 23:17 UTC as 24th comment
In reply to:

Chris Crevasse: Re the default noise reduction and sharpening settings: do either ACR or CO change the default settings depending on the ISO of the image?

I don't know about C1, but Lightroom lets you apply settings on import based on camera, ISO, etc. You just have to set it up.

Link | Posted on Apr 5, 2021 at 18:26 UTC
In reply to:

ChelseaPhotographer: My burning question: is there banding in the deep shadows like the GFX 100 or has the issue been solved?

Check out Jim Kasson's website. He tested and concluded it's fixed.

Link | Posted on Mar 26, 2021 at 19:00 UTC
In reply to:

Harold66: The ONE things that does not get mentioned about the FP cameras is the fact that you can have many more ratios than the 3.2 ratio . Credits to Sigma for allowing more creativity by offering ratios which usually are only available on medium format cameras . I think this camera in a way competes with the Fuji GX50r

@peterwr -- Cropping in post is easy enough, that's true. However, I prefer to compose to the whole frame, so that approach isn't a good fit for me,

Link | Posted on Mar 26, 2021 at 12:53 UTC
In reply to:

Harold66: The ONE things that does not get mentioned about the FP cameras is the fact that you can have many more ratios than the 3.2 ratio . Credits to Sigma for allowing more creativity by offering ratios which usually are only available on medium format cameras . I think this camera in a way competes with the Fuji GX50r

Almost the first thing I did when I saw this was check if it can do 4:3. Since switching from an X-T2 to a GFX 50R, I've really come to prefer 4:3. It bothers me to no end that Fuji won't just add 4:3 to its older X cameras.

I really like my lunchbox by the way! The 50R is a superb camera (for how I use it).

Link | Posted on Mar 25, 2021 at 22:51 UTC
In reply to:

Richard Murdey: Me: "This seems like a very welcome announcement, I can't imagine anyone would find a reason to knock it."

[starts to read dpreview comments]

Oh...

It seems to depend on the forum the DPReview Medium Format thread is moderated, so bad behaviour is nipped in the bud. More importantly, it tends not to attract the toxic people. Anything brand-specific naturally trends to unpleasant

Link | Posted on Mar 24, 2021 at 16:48 UTC
In reply to:

Eugene A: How would such a tool handle out of focus areas in case of a very shallow dept of field - would it preserve the intended creativity when a super-fast lens was used ?

That would be the next real challenge to machine learning aided enhancement tools.

The way it handles lettering is not necessarily a deal breaker. I would encourage you to see for yourself if you have the software. I just made a wide-open frame of my messy desk, with lots of papers on it that have lettering. I focused in the centre so there would be increasingly out of focus writing in foreground and background. I then compared to the original.

When the letter is clearly defocused, it leaves it alone. Clearly focused lettering gets the edges tidied up a bit (like sharpening does). It's that transition zone where I noticed some lettering had been enhanced. The result is a pushing out (if I can describe it that way) of the zone of in-focus.

Whether or not someone finds this objectionable will be case-by-case. I can say that in this experiment, I would not have noticed without comparing to the original.

To be honest I have problems with tree branches. In some test images they got unrealistically sharp; they were clearly in the out-of-focus zone, but sharp.

Link | Posted on Mar 20, 2021 at 17:47 UTC
In reply to:

Eugene A: How would such a tool handle out of focus areas in case of a very shallow dept of field - would it preserve the intended creativity when a super-fast lens was used ?

That would be the next real challenge to machine learning aided enhancement tools.

Eugene A, another area where things can get strange with this tool is text. If there's a sign in your picture that isn't sharp, it can enhance each letter to different degrees based on whether or not it can figure out what the letter is. That makes sense because it's easy enough to supply a huge amount of text during training.

You can end up with a mixture of blurry and sharp letters on the same sign, which looks strange. I posted some examples in this thread (scroll down about 1/3rd): https://www.dpreview.com/forums/thread/4560860

Link | Posted on Mar 20, 2021 at 15:42 UTC
In reply to:

Eugene A: How would such a tool handle out of focus areas in case of a very shallow dept of field - would it preserve the intended creativity when a super-fast lens was used ?

That would be the next real challenge to machine learning aided enhancement tools.

It's mixed. Something that is really out of focus stays out of focus. It's the transition zones where it can go to work on things you might want left alone.

There are some sample images here: https://www.dpreview.com/forums/thread/4561112 IMPORTANT: When I posted this thread Adobe had not yet fixed the bug where ACR wrote the original file resolution into the DNG it created rather than the actual x2 in X and Y resolution of the file, so it looked like the DNG had enhancements but not increased pixels. That was a bug that has since been fixed. The pictures in that sample are still relevant though.

Link | Posted on Mar 20, 2021 at 15:20 UTC
In reply to:

tinternaut: Has Lightroom CC been updated to make use of ACRs new Super Res feature?

A minor friendly amendment to Charles Hull's approach (for Windows anyway): In your Lightroom catalogue, right click the file you want to process and choose "Find in Explorer". Right click on the file in Explorer and Open with Photoshop. Run Enhance, and then click "Done" without saving. ACR dumps the enhanced file in the same folder on creating it so you don't have to save. Then do the Synchronize Folders trick to bring the new DNG into your catalogue.

If you want to apply Super Resolution after you developed a file, you have to copy or sync settings manually. Don't forget to check whether or not you need as much Texture, Sharpening, Detail, etc. You might not.

Link | Posted on Mar 17, 2021 at 23:31 UTC
In reply to:

Edenhurst: Lightroom is garbage. I HATE that I have to pay for this crap to get Photoshop and STILL PAY FOR CAPTURE ONE—because crappy Lightroom can’t do everything I need it to, and Lightroom Classic is an old, buggy, piece of trash that was coded about ninety years ago. How can they charge for a tool that can’t even show you two images SIDE BY SIDE. You can’t do that in the new version of Lightroom!!!! You can’t compare two images!!!

You can compare two images side-by-side in Lightroom. It's called Compare view. You can develop one image and have a different image beside it.

What are you talking about?

Link | Posted on Mar 17, 2021 at 23:22 UTC

So fun fact, but it doesn't quite work as described here.

While trying out the new Enhance tool in ACR I noticed it dumps a new DNG into the folder where my source file was. This DNG has the resolution of the source file, but it has been enhanced with the machine learning secret sauce. I can import the file into Lightroom as a DNG. If I export the file from LR with x2 X and Y resolution and create a TIFF file, that TIFF file is exactly the same as the TIFF that is created by ACR if you save after using the Enhance tool. In other words, the enhancements seem to be in the DNG and the x2 X and Y resolution TIFF is completely unnecessary.

To be sure I wasn't imagining things, I compared the TIFF I made from the DNG ACR created, and the one created by ACR. They're the same. I also checked to see if the "enhanced with Super Resolution" DNG ACR created was the same as what I'd get by using the existing Enhance tool in Lightroom on the source RAW file. They're not the same.

Link | Posted on Mar 14, 2021 at 03:06 UTC as 9th comment
Total: 50, showing: 1 – 20
« First‹ Previous123Next ›Last »