Hacker News Reader

Comment by piva00 | Hacker News Reader

Not sure what you were using, I don't see that at all with my Fuji kits.

I have both a X-Pro 3 and a X-T30 for street photography and both shoot much, much superior pictures than any smartphone is capable of...

APS-C is still pretty relevant, your old DSLR might not be up to par to latest smartphone cameras though. And the image processing done by smartphones using AI tend to create weird and ugly artefacts depending on conditions, that doesn't happen with my mirrorless cameras, for example.

Have you tried printing smartphone pictures and compare them to your DSLR shots?

View on HN

Disclaimer, I'm a retired pro photographer that sold his full-frame to focus on software engineering.

  - The X-Pro 3 is $2k, not what I call middle-end.
  - I agree on the aggressive AI processing. Fortunately I could disable it.
  - It was a Nikon D5500. I used the 18-55 kit lens, but f/1.8 prime lenses can do better indeed, at the cost of switching lenses all day.
  - I compared on my 27" screen, no difference, even in low-light scenarios and at different ISOs

The X-T30 is US$ 800-900 and uses the same sensor and processor of the X-Pro3 so they're pretty equivalent on picture quality. Don't stick to the X-Pro3 mention as that's missing the point.

The D5300 is pretty old, I had one in 2013-2014, coming from a D3200.

> - I compared on my 27" screen, no difference, even in low-light scenarios and at different ISOs

This might be the main difference between us, I usually do prints in A3+ sizes and the differences in picture quality between a smartphone and my cameras are very noticeable.

The point of my comparison was price. You can find better smartphones too.

The A72 and D5500 have the same used price.

On the price point I agree with you but then the comparison becomes not so level by comparing semi-conductors technology from 2014 to the ones from 2021, that's 7 years of evolution on sensor technology packed into the phone's sensor, plus all the image processing advances since then.

Again, I understand the price point but it's an oddball comparison. Perhaps a comparison between the A72 and a Fujifilm X-E2 could tell us more but I don't have either devices to directly compare myself :/

I'd love to see a graph of sharpness/$ for both categories ! I think they're equal until the $400 range, after Mirrorless obviously wins

Well, you won't find anyone saying their 2022 smartphone outperforms their 2022 DSLR

Because people who find their 2022 smartphone outperforming their 2015 DSLR don't upgrade to a 2022 DSLR.

I sold my DSLR gear in 2015 including my absolute favorite 35/f2 lens, and I have an X100V on backorder in 2022 for its 35mm equivalent f/2 prime lens: seven years of AI missteps and absent bokeh in my preferred framing has finally gotten to me. I know that my phone will take better telephotos, and I know my phone has RAW mode and three lenses and takes amazingly great pictures. So I'm specializing my camera to exactly where I love it most, and will let my phone handle everything else, and I’m content that each has their strengths.

> The X-Pro 3 is $2k, not what I call middle-end.

Fuji uses the same sensors on many cameras, you can get an xt-2 or xe-3 for much cheaper, with the same sensor

D5300 is APS-C. I know because I have its descendent, the D5600. I wish it were full frame, especially this time of year. I should probably sell it since I never use it now that I have a good phone camera, but I would eventually miss my 70-300.

Huh. When I compare RAW output from my D5300 (using the default 18-55 mm lens) and a Pixel 6, the difference is staggering. Granted, the JPG output from the Pixel is usually on par with the JPG from the DSLR, high dynamic range is something where the Pixel is even a bit better most of the time.

But once you take RAW photos and hit the Auto button in Lightroom, the Pixel doesn't hold a chance against the D5300.

Sure, prosumers like us can squeeze extra juice out of the DSLR. But our mothers cannot.

That was a bit sexist. But most people can't frame, compose or level either. So doing some automated post-processing inside a phone won't help.

It doesn't even have to be a prosumer, average kids in my city know raw and post process. Kids are very familiar with editing, in fact, gen z is also blowing gen y out of water when it comes to editing video.

Even if you have been a professional photographer for a significant length of time, you shouldn't use it to try to appeal to authority. However, a cursory glance at your profile tells me you aren't even 30, come on mate. Unless you were a professional photographer before your 10th birthday I really don't think your experience is better than any other enthusiast.

Did you do a real image diff on the same setup? I doubt it. Phone cameras have come a long, long way but a lot of the advances are through "smoothing" things out through software.

What's wrong with my age?

I've done 6 years of professional photography to pay for college. Portrait shoots, weddings, even produced videos, ads, festivals, wildlife documentaries. I worked on Nikon D4S fullframes. How is my age relevant ?

My point is, professionals squeeze extra juice of the hardware, but the average consumer does not.

The image your phone generates isn’t real. It’s a medium quality photo enhanced by “AI.” See all the cases of iPhone pictures adding faces where people aren’t there.

You can use third party camera apps to avoid Deep Fusion.

Maybe show some comparisons so we can judge rather than just assessing "facts"

I just felt it's such an outrageous claim that I really hadn't to show that cameras with much larger sensors and better optics would shoot better pictures than a smartphone.

For me it's the opposite, show how smartphones are better than the current crop of mirrorless APS-C as this is the extraordinary claim requiring evidence.

When I get some time I might shoot some comparison pictures, but if I don't: remember that I'm not here to serve your demands, I'm sorry.

> I just felt it's such an outrageous claim

I have used an iPhone SE and a mirrorless M4/3 camera to photograph a sheet of paper containing barcodes of varying sizes (including some with bars less than 1 pixel wide). I then checked which barcodes were readable in the resulting image.

The light levels were the same, both cameras were positioned and zoomed so the target took up the entire image, and both cameras were supported on a tripod.

I expected the M4/3 camera would blow the iPhone out of the water with its much larger lens, bigger sensor, and higher price. But no, the iPhone's image had marginally more readable barcodes.

Modern smartphone camera performance is just crazy, for the sensor size.

You didn't specify if the lens was up to the task in terms of sharpness on that MFT.

A fair point but unfair in that you are not also asking this of the OP. The status quo is that that a dedicated camera will be better than a smartphone (see other comments)

Sure, here it is ! https://imgur.com/a/VyHLYqP

Protocol: handheld at 10 PM, 10 shots each, at different ISOs, picked the best one

The bottlenecks are different, but the sharpness is comparable.

  - The DSLR was limited by optics, it's blurrier with some chromatic aberration
  - The Phone has strong AI processing, I wish I disabled it

> The Phone has strong AI processing, I wish I disabled it

It's still over-sharpened and probably used multiple shots to get high dynamic range, it's much more noisy too, and shows less resolution

Also the d5500 is a lower tier camera from 2015, the phone was released in 2021

They have the same used price, which is the point of my comparison.

Yes there's HDR bracketing, but we only care about the result.

One is blurry (optics), the other has artifacts (AI), but overall sharpness is similar.

TBH both photos look horrible compared to smartphones nowadays or is this just 100% crop? I'd like to see whole photo.

Yes this is a 100% crop to highlight the sharpness is handheld low-light, the worst scenario there is

Dynamic range is better on the phone, but otherwise the DSLR has sharper edges, less noise, nicer colors, and is less mushy (but that is possibly due to the "AI processing", so ditto about the "real" image). That said, noise reduction is usually more advanced on phones, and handheld with a kit lens at night with high-contrast zones is kind of the worst scenario for DSLRs (hopefully it was a stabilized kit lens at least).

You can read the words "Hotel Platinum" on the phone photo. And it's blurred and "mushy" on D5500. And the phone had additional glare from an oncoming train, and it still pulled out things out of the dark.

Depends on what you need, of course, but for most people the photo from the phone is superior.

Reply on news.ycombinator.com