:-D
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
A XH1/A7 III comparison 6400/12800/25600 ...
- Thread starter baobob
- Start date
Um, worms? These images don't even show the kinds of things that provoke 'worms'. I think you're mistaking worms for standard noise artifacts (which by the way also appear on bayer sensors in exactly the same way).
The noise artifacts do appear a little sharper on the Fuji images, which may be an x-trans side effect, but I suspect it's actually due to the raw processor.
On the other hand if you can't pick out one camera consistency without having a crutch to look at - then that was sort of the point. I don't know if baobob's intent or not but he just gave us a "double blind test" didn't he?My takeaway: it's much easier to interpret results if the images from each camera are consistently posted on the same side and with a caption noting ISO and other settings.
If you can pick out one camera 3 out of 3 without any help then you can see a difference. if not - then you can't see a difference without putting on your reading glasses and reading the fine print on the test scenes. Isn't that the point?
Peter Jonas
Veteran Member
I note the discussion in this thread regarding differences between how Sony and Fujifilm define their ISO sensitivity.
In the end that should not matter.
However, it is important that images presented here for comparison have been exposed using the same shutter speed and same aperture settings for each respective ISO sensitivity setting. Once that is done, the images can be processed to represent the same brightness.
With their test images DPR explicitly states that this is the case.
For the sake of clarity can you please publish the ISO sensitivity, shutter speed and aperture settings for all images presented here for comparison.
That would be much more constructive for all of us not in the know..
--
Cheers,
Peter Jonas
In the end that should not matter.
However, it is important that images presented here for comparison have been exposed using the same shutter speed and same aperture settings for each respective ISO sensitivity setting. Once that is done, the images can be processed to represent the same brightness.
With their test images DPR explicitly states that this is the case.
For the sake of clarity can you please publish the ISO sensitivity, shutter speed and aperture settings for all images presented here for comparison.
That would be much more constructive for all of us not in the know..
RAW develpooed by ACR with default settings only
Rather unexpected...
--
Good judgment comes from experience
Experience comes from bad judgment
--
Cheers,
Peter Jonas
ialarmedalien
Veteran Member
- Messages
- 1,182
- Solutions
- 1
- Reaction score
- 863
True -- though a properly blinded test would have removed the labels from the images, so even putting on glasses (if applicable) and viewing at 100% would have left participants none the wiser as to which camera generated which image. Ideally it would also have some duplicated images (i.e. left and right are identical) to judge to what extent people are able to discern differences between images or whether they are relying on existing biases, such as the tendency to pick the last or rightmost item as superior, even if all items are identical.
To add to the experiment, it would also be interesting to ask participants which images they thought were produced by which camera, and whether they had a preference for one camera over the other, to see what extent their preference informed their judgements.
Sure, the initial post was to compare high ISO performance of two different cameras, but there's no reason this can't also be a psychology experiment!
To add to the experiment, it would also be interesting to ask participants which images they thought were produced by which camera, and whether they had a preference for one camera over the other, to see what extent their preference informed their judgements.
Sure, the initial post was to compare high ISO performance of two different cameras, but there's no reason this can't also be a psychology experiment!
Peter Jonas
Veteran Member
That is a fair point. Not telling the camera is fine.On the other hand if you can't pick out one camera consistency without having a crutch to look at - then that was sort of the point. I don't know if baobob's intent or not but he just gave us a "double blind test" didn't he?My takeaway: it's much easier to interpret results if the images from each camera are consistently posted on the same side and with a caption noting ISO and other settings.
If you can pick out one camera 3 out of 3 without any help then you can see a difference. if not - then you can't see a difference without putting on your reading glasses and reading the fine print on the test scenes. Isn't that the point?
However, ISO sensitivity, shutter speed and aperture settings should be stated for each image.
Canadianguy
Veteran Member
No one really cares what ISO they shoot at UNLESS they use an external light meter - then it becomes really important and this is the problem with Fujifilm - they are using a different standard than what what external light meters are using.
So you shoot at a recommended setting from the light meter and it turns out its doesn't correctly exposure the image based on the external meter because the camera is using a different ISO standard ... how silly is that?
If you don't use a light meter - sure no big deal - just some spin for bragging rights on how good their high ISO images are.
Don't know about you but I post process my images and the APS-C sensor images always falls apart faster than 35mm sensor images when I push and pull on the files.
So if you are are SOOC kind of photographer - no worries for you - if you like to process your files - 35mm sensor images have more latitude.
That is a fair point. Not telling the camera is fine.On the other hand if you can't pick out one camera consistency without having a crutch to look at - then that was sort of the point. I don't know if baobob's intent or not but he just gave us a "double blind test" didn't he?My takeaway: it's much easier to interpret results if the images from each camera are consistently posted on the same side and with a caption noting ISO and other settings.
If you can pick out one camera 3 out of 3 without any help then you can see a difference. if not - then you can't see a difference without putting on your reading glasses and reading the fine print on the test scenes. Isn't that the point?
However, ISO sensitivity, shutter speed and aperture settings should be stated for each image.
michaeladawson
Veteran Member
I went to the IR site and looked at the test images from the X-H1 and the A7rIII and looked at the exposure data they published.
The X-H1 and the A7rIII both used the exact same exposure settings at the same ISO.
ISO 6400: 1/500s @ f/8
ISO 12800: 1/1000s @ f/8
ISO 25600: 1/2000s @ f/8
Did they have the same amount of light falling on the scene for each camera? That I do not know.
The X-H1 and the A7rIII both used the exact same exposure settings at the same ISO.
ISO 6400: 1/500s @ f/8
ISO 12800: 1/1000s @ f/8
ISO 25600: 1/2000s @ f/8
Did they have the same amount of light falling on the scene for each camera? That I do not know.
Peter Jonas
Veteran Member
Yes, I understand that. But it would have been nice to state the ISO sensitivity setting for each image pair just for the sake identification.The Iso sensitivity on each camera was the same on each pair - that was sort of the point.That is a fair point. Not telling the camera is fine.On the other hand if you can't pick out one camera consistency without having a crutch to look at - then that was sort of the point. I don't know if baobob's intent or not but he just gave us a "double blind test" didn't he?My takeaway: it's much easier to interpret results if the images from each camera are consistently posted on the same side and with a caption noting ISO and other settings.
If you can pick out one camera 3 out of 3 without any help then you can see a difference. if not - then you can't see a difference without putting on your reading glasses and reading the fine print on the test scenes. Isn't that the point?
However, ISO sensitivity, shutter speed and aperture settings should be stated for each image.
--
Cheers,
Peter Jonas
I am fully aware of that too.
However, as one contributor to this discussion has claimed, due to the different ways these two manufacturers define the ISO sensitivity if their cameras, it is likely that in order to achieve the same image brightness, these cameras are likely to use different shutter speeds pr EC settings. (Assuming of course that their respective aperture settings are the same).
As you say it's not a perfect test, nor does it have to be one.
But it ought to be fair test. But right now, I am not in the position to decide if it is fair or not.
I am of the opinion that if one publishes test images for comparison, then those settings either must be the same, or if they not, then that fact is explicitly stated.
I agree, there is not much in it.
It was the the OP's choice to start this discussion of just how little that difference is. To support his point, he re-published someone else's test images without the exposure settings, nor did he state that they are the same for each image pair.
I just want to know that before I make my own conclusions about the differences.
--
Cheers,
Peter Jonas
Last edited:
Peter Jonas
Veteran Member
Due to reasons of practicality I think it's reasonable to assume that the light used for the scene is the same in both cases. However, it would be informative to know tough.
I have also noted that the IR website lists EC values used for each image. I wonder if those ware also the same?
--
Cheers,
Peter Jonas
Last edited:
C
Christof21
Guest
Peter Jonas
Veteran Member
John Gellings
Veteran Member
- Messages
- 11,472
- Solutions
- 1
- Reaction score
- 9,345
Exactly Bob... but many seem to like camera wars!
Last edited by a moderator:
5
57even
Guest
Just as silly as the fact that ANSI standard light meters can be set for middle grey at anything between 12% and 15% saturation, not 18% middle grey as stated by ISO.
And the fact that ISO only applies to output files not raw - ie images processed in camera.
A correct ANSI meter reading for digital will likely always give the wrong readings for raw.
Who needs a light meter these days? (And yes, I have worked in a lot of studios where they used to base exposure off polaroids more than light meters. Now, with digital, they just adjust it till the image looks right.
If you shoot at ISO12800 all the time, this might even be relevant.
5
57even
Guest
michaeladawson
Veteran Member
It doesn't make a whit of difference what the EC settings are. The EC settings play into how the metering calculates the exposure. The OP's post is not about how the two cameras meter.
You made the point that the exposure needs to be the same between each camera for the ISO comparison to mean anything. And in fact they are. The only relevant unknown here is how the scene is lit and whether it is the same for each camera.
5
57even
Guest
Fairly predictable response from the relevant quarters ;-)
The Sony is supposed to have a BSI sensor with a 1/3 stop improvement over the A7II
So we should still expect to see more of an obvious difference, even allowing for a difference in ISO calibration.
Which proves what most of us already know - there isn't really much difference between APSC and FF most of the time, and FF certainly won't suddenly make your images look miraculously better.
On the other hand, a one week course on Photoshop certainly will.
Some people won't ever be convinced though.
The Sony is supposed to have a BSI sensor with a 1/3 stop improvement over the A7II
So we should still expect to see more of an obvious difference, even allowing for a difference in ISO calibration.
Which proves what most of us already know - there isn't really much difference between APSC and FF most of the time, and FF certainly won't suddenly make your images look miraculously better.
On the other hand, a one week course on Photoshop certainly will.
Some people won't ever be convinced though.
Canadianguy
Veteran Member
Jhuffphoto
Well-known member
- Messages
- 108
- Reaction score
- 143
Keyboard shortcuts
- f
- Forum