New X-E1 owner joins the forum

Home · Register · Search · Software · Hosting · Software · Join Upload & Sell

FM Forums \| Alternative Gear & Lenses \| Join Upload & Sell
1 2 3 4 end

carstenw
Offline
• • • • • •
Upload & Sell: Off

p.2 #1 · New X-E1 owner joins the forum

In this discussion, I guess I know who sets the toaster on high

Jan 02, 2013 at 09:50 PM

mirkoc
Offline
• • •
Upload & Sell: Off

p.2 #2 · New X-E1 owner joins the forum

I don't like crunchy photos either. But, as someone said, the turkey...

Jan 03, 2013 at 12:29 AM

Jochenb
Offline
• • •
Upload & Sell: Off

p.2 #3 · New X-E1 owner joins the forum

Hmm... maybe I'm less critical.
Here's the same example I used earlier this week to show the difference between an OOC Fuji jpeg and the RAW file converted in the C1 beta. In harsh light.
100% corner crop:

Looking at the RAW file on the right. I'm perfectly fine with such results. What crunchy mess?

Jan 03, 2013 at 12:46 AM

mortyb
Offline
• • •
Upload & Sell: Off

p.2 #4 · New X-E1 owner joins the forum

Good example. Sometimes I get a feeling some people are more focused on theory than on real, practical results. I trust my eyes and what I see in the final output.

Jan 03, 2013 at 12:54 AM

alba63
Offline
• •
Upload & Sell: Off

p.2 #5 · New X-E1 owner joins the forum

Toothwalker wrote:
The knowledge that aliasing is irreducible and occurs in any image with high-frequency contents (sharp edges in the scene combined with high-quality optics at their best apertures) is more important to me than the question whether it is easily recognized.

I have seen examples of rough files, mainly by the Sigma Merrill, but your statement sounds as if theoretical knowledge goes over what your eyes see. If you don't recognize it, it shouldn't be a problem.

Bernie

Jan 03, 2013 at 01:09 AM

DavidWEGS
Offline
• • • •
Upload & Sell: On

p.2 #6 · New X-E1 owner joins the forum

rattymouse wrote:
I wonder if those who make this claim ever even shot film, or have evaluated film images next to digital. The difference is clear as day.

I've shot thousands of rolls of film and can spot the difference with almost unerring accuracy.

Well, I too have shot film for years (since the mid 80's). Sill have half my old film cams,(though they mostly collect dust these days). I might even have a few remaining packs of fuji in the freezer.

I must say, you are correct about digital, no matter how clever they get, can never replicate an organic emulsion. Just not possible.

However, this sensor steps closer to this goal, and I think the thing from the POV of those who have souped lots of their own handiwork and printed with great care those results, that is something to enjoy progress toward. Incremental, yes, but closer in look and feel than the D1 I have sitting here, and I dare say they will find another way to "randomize" the look of those pixels further in the future.

Jan 03, 2013 at 02:58 AM

Ripolini
Offline
• •
Upload & Sell: Off

p.2 #7 · New X-E1 owner joins the forum

alba63 wrote:
... your statement sounds as if theoretical knowledge goes over what your eyes see.

mortyb wrote:
Sometimes I get a feeling some people are more focused on theory than on real, practical results. I trust my eyes and what I see in the final output.

First, one example/picture doesn't tell much about the behaviour of a sensor. In particular, that picture shows a 'corner' crop and no lens has its best performance in the corners.
Secondly, as you probably know, besides the disadvantages of an AA-less sensor, the 6x6 matrix of the X-Trans sensor seems to suffer from fractal-like artifacts (see here).
Last but not least, there is nothing more practical than a good theory ...

Jan 03, 2013 at 07:51 AM

alba63
Offline
• •
Upload & Sell: Off

p.2 #8 · New X-E1 owner joins the forum

Ripolini wrote:
besides the disadvantages of an AA-less sensor, the 6x6 matrix of the X-Trans sensor seems to suffer from fractal-like artifacts (see here).

As far as I can tell from all the reviews and samples I have seen, the problem with those artefacts is not the sensor in itself, but the demosaicing algorithms of the RAW converter. RPP (on the Mac) seems to mostly avoid them, Capture One support is on it's way to be published. Lloyd Chambers used Adobe converter which currently is definately not the best option for the X-Trans sensor.

All in all, not sure whether the very specific sensor layout and it's predictable problems with RAW software support were the most clever decision, but at least Fujifilm shows creative ideas in terms of sensor design.
Lack of AA- less sensor have disadvantages, just like the AA filters themselves. I have been shooting with AA- less cameras for quite a while, and the advantages of such designs are visible day in day out. It is a question of personal balance or whether one does want to live with the downsides like moiré. The latter is theoretically present in almost every image, but in practice irrelevant in most of the prints I ever did with those cameras.

Bernhard

Jan 03, 2013 at 11:46 AM

Ripolini
Offline
• •
Upload & Sell: Off

p.2 #9 · New X-E1 owner joins the forum

alba63 wrote:
As far as I can tell from all the reviews and samples I have seen, the problem with those artefacts is not the sensor in itself, but the demosaicing algorithms of the RAW converter.
Well, sensor and demosaicing algorithms are strictly related

alba63 wrote:
RPP (on the Mac) seems to mostly avoid them, Capture One support is on it's way to be published. Lloyd Chambers used Adobe converter which currently is definately not the best option for the X-Trans sensor.
Unfortunately I don't use a Mac so I'm a little bit worried about what I've read about fractal-like artifacts. It seems that the supplied raw converter too doesn't perform flawlessly.
I would love buying the E-X1 so I'm trying to get as much info as possible. Therefore I'm grateful to you for sharing your experience.

alba63 wrote:
Lack of AA- less sensor have disadvantages, just like the AA filters themselves. I have been shooting with AA- less cameras for quite a while, and the advantages of such designs are visible day in day out. It is a question of personal balance or whether one does want to live with the downsides like moiré.
I have taken pictures with my D700 where moiré is visible ...

alba63 wrote:The latter is theoretically present in almost every image, but in practice irrelevant in most of the prints I ever did with those cameras.

I would like knowing how large your prints are, and if "irrelevant moiré" is less visible in small prints, or if it is negligible in large prints as well.
Thanks.

Riccardo

Jan 03, 2013 at 12:27 PM

Stuart
Offline
• •
Upload & Sell: Off

p.2 #10 · New X-E1 owner joins the forum

Jochenb wrote:
Hmm... maybe I'm less critical.
Here's the same example I used earlier this week to show the difference between an OOC Fuji jpeg and the RAW file converted in the C1 beta. In harsh light.
100% corner crop:

Looking at the RAW file on the right. I'm perfectly fine with such results. What crunchy mess?

Can you share the recipe you used in C1 to produce that output on the right? I'd like to do some experimenting in C1 Beta on a few of my RAW images. Thanks!

Jan 03, 2013 at 07:18 PM

Toothwalker
Offline
• • •
Upload & Sell: Off

p.2 #11 · New X-E1 owner joins the forum

Jochenb wrote:
Looking at the RAW file on the right. I'm perfectly fine with such results. What crunchy mess?

I did not say that every picture of vegetation looks like a crunchy mess, but an example from an image corner with possible wind blur may not the best test case.

Just look at the first image on this page and imagine the hair is grass or hay. The last example on that page is directly relevant.

mortyb wrote:
Good example. Sometimes I get a feeling some people are more focused on theory than on real, practical results. I trust my eyes and what I see in the final output.

Theory is important, as are real practical results. When the theory says there is aliasing, there is aliasing. If simultaneously the final output looks marvelous, it looks marvelous. One should just be careful with claims about "original" detail.

alba63 wrote:
As far as I can tell from all the reviews and samples I have seen, the problem with those artefacts is not the sensor in itself, but the demosaicing algorithms of the RAW converter.

That is unlikely, but could easily be verified with tests at smaller apertures (where diffraction acts as a natural AA-filter). I don't know whether these reviews do that.

Jan 03, 2013 at 07:30 PM

Jochenb
Offline
• • •
Upload & Sell: Off

p.2 #12 · New X-E1 owner joins the forum

You guys are always referring to diglloyd. He's using the flawed Adobe camera RAW conversion. He also heavily sharpens in ACR. This leads to totally messed up results. Not very fair IMHO.
The C1 beta is massively better.

Stuart wrote:
Can you share the recipe you used in C1 to produce that output on the right? I'd like to do some experimenting in C1 Beta on a few of my RAW images. Thanks!

I didn't do much to it. Just played around with the sharpnening sliders.
Can't remember the exact settings, it was just a quick random test.

Jan 03, 2013 at 07:44 PM

U.C.
Offline
• •
Upload & Sell: Off

p.2 #13 · New X-E1 owner joins the forum

Brody LeBlanc wrote:
+1. I owned the 5DmkII before the X-Pro1 and I'm much happier with the X-Trans output.
Honestly, the focus is just as good and the ISO performance is so much better than the 5DmkIl. It's hard to believe considering a 1.5x sensor beats out a full frame sensor in almost every way, but X-Trans marks a huge leap forward in sensor design though it's not perfect, it's incredibly innovative.

http://www.dpreview.com/previews/fujifilm-x-e1/10 tells me a different story. All I see is much more noise reduction and chroma noise at high iso.

Jan 03, 2013 at 09:15 PM

Toothwalker
Offline
• • •
Upload & Sell: Off

p.2 #14 · New X-E1 owner joins the forum

Jochenb wrote:
You guys are always referring to diglloyd. He's using the flawed Adobe camera RAW conversion. He also heavily sharpens in ACR. This leads to totally messed up results. Not very fair IMHO.
The C1 beta is massively better.

If memory serves me well this was the first time in my life that I referred to diglloyd ...

It is good practice to show examples to prove your points. You show examples and I say they are no good. I show examples and you say they are no good.

Case closed. Let's continue with photography.

Jan 03, 2013 at 09:23 PM

sebboh
Offline
• • • • •
Upload & Sell: Off

p.2 #15 · New X-E1 owner joins the forum

Toothwalker wrote:
mortyb wrote:
Good example. Sometimes I get a feeling some people are more focused on theory than on real, practical results. I trust my eyes and what I see in the final output.

Theory is important, as are real practical results. When the theory says there is aliasing, there is aliasing. If simultaneously the final output looks marvelous, it looks marvelous. One should just be careful with claims about "original" detail.

+1, though i could care less about "original" detail, i just care that it looks good to me.

Jan 03, 2013 at 09:49 PM

mortyb
Offline
• • •
Upload & Sell: Off

p.2 #16 · New X-E1 owner joins the forum

Who has made claims about "original" detail? I'm not following. Anyways, I agree with sebboh, as long as it looks good, I'm happy.

Jan 03, 2013 at 10:17 PM

Mescalamba
Offline
• • • •
Upload & Sell: Off

p.2 #17 · New X-E1 owner joins the forum

There were ways (and still are) to get decent files from X-Pro 1. Im glad C1 will have support. They could fix S5 Pro when they are at it..

Judging camera from Adobe output is pretty stupid.

Jan 03, 2013 at 10:47 PM

Jochenb
Offline
• • •
Upload & Sell: Off

p.2 #18 · New X-E1 owner joins the forum

Toothwalker wrote:
Jochenb wrote:
You guys are always referring to diglloyd. He's using the flawed Adobe camera RAW conversion. He also heavily sharpens in ACR. This leads to totally messed up results. Not very fair IMHO.
The C1 beta is massively better.

If memory serves me well this was the first time in my life that I referred to diglloyd ...

It is good practice to show examples to prove your points. You show examples and I say they are no good. I show examples and you say they are no good.

Case closed. Let's continue with photography.

Don't feel offended. I wasn't just talking about you, but in general. Ever since Lloyd Chambers posted his flawed examples and the word "unusable" the forums seem to be overflowing with talk about the artifacts.
I agree that my example maybe wasn't the best, but here's another one I found on the web:

Fuji RAW file converted with C1 Beta

Look how much better the result is than those ACR ones from Lloyd Chambers.
To me, this performance is acceptable. I like what I'm seeing on my screen when I edit my Fuji files. I don't mind if people don't agree, everybody has their own preferences.

Jan 03, 2013 at 11:34 PM

aleksanderpolo
Offline
• • •
Upload & Sell: On

p.2 #19 · New X-E1 owner joins the forum

Thanks for the link, I did a quick comparison of this file with a file from K5IIs (also 16MP without AA filter), you can see the difference in fine detail:

This image is copyrighted by the owner

The bottom one is from the horrible first gen m43 sensor, upsampled to 16MP to roughly match the size, just for fun.

One can go to dpreview to see the file from D800E to see which one of them is giving tons of false detail, making grids look like maze.

Jochenb wrote:
Don't feel offended. I wasn't just talking about you, but in general. Ever since Lloyd Chambers posted his flawed examples and the word "unusable" the forums seem to be overflowing with talk about the artifacts.
I agree that my example maybe wasn't the best, but here's another one I found on the web:

Fuji RAW file converted with C1 Beta

Look how much better the result is than those ACR ones from Lloyd Chambers.
To me, this performance is acceptable. I like what I'm seeing on my screen when I edit my Fuji files. I don't mind if people don't agree, everybody has their own preferences.

Jan 04, 2013 at 03:01 AM

Ripolini
Offline
• •
Upload & Sell: Off

p.2 #20 · New X-E1 owner joins the forum

Jochenb wrote:
... here's another one I found on the web:
Look how much better the result is than those ACR ones from Lloyd Chambers.

Thanks for posting that link.
If C1 beta is the best raw converter currently available for the X-Trans sensor, I'll not buy any Fujifilm camera with that sensor inside.

Jan 04, 2013 at 07:44 AM

1 2 3 4 end
FM Forums \| Alternative Gear & Lenses \| Join Upload & Sell 1 2 3 4 end