Velvia 50 35 mm.. 120..4x5 in against D 700 and Hasselblad H3 31

[elliotn]

I always scan at the full physical resolution that the Imacon can do.
I get the quality that the film material has.

I didn do any upres or downres with this picture.... it comes out of the scanner this way with the highest resolution you can make..... go figure out how big the samples are !

Hmm, This is all getting very confusing. That is not a 100% crop of an Imacon scanning 4x5 film at native resolution.

I can't help wondering whether you're using the Imacon properly.

 

[rayman 2]

well what do you think I did with the smaller sized formats other then uprezing them ? they were uprezed to the same size in Photoshop using bicubic interpolation. I will have the 4x5 neg drumscanned so we will see if it was wrong.
What makes me more curious is why flickr is changing the res scale of my files.
I uploaded a big file and it came out 1000 px wide.

 

[rayman 2]

I can't help wondering whether you're using the Imacon properly.

I am using the Imacon properly but I am not using Flickr properly.
The file that I uploaded had 8902x 6282 pixels and came out at about 1000
pixels wide..... : (

 

[elliotn]

I am using the Imacon properly but I am not using Flickr properly.
The file that I uploaded had 8902x 6282 pixels and came out at about 1000
pixels wide..... : (

Thanks for the explanation. 8902 x 6282 pixels sounds like the right number of pixels for an Imacon 4x5 scan at native resolution. (I just opened one of mine, and it's 9774 x 7558 pixels, but that includes the neg rebate.)

 

[rayman 2]

Well i promised to make some pics for the comparison with a D3X.
note that its now 2 weeks after the first pics and the trees are now a little
more yellow and red thats not due to the camera thats fall !

note that i can not compare some haze either but we had comparable weather situation with very low contrast.
I got some posts that this lens is better and that lens well i made the pics
with 2,8 24- 70mm and micro nikkor d60mm and a 15 year old 3,5-4,5 28 - 70 mm
and the results are comparable the d 60mm is not bad or soft in the distance
All were used at F11 and 60mm focal length.
I also used the 50mm 1,4 ais....

 

[Marianne Oelund]

I got some posts that this lens is better and that lens well i made the pics
with 2,8 24- 70mm and micro nikkor d60mm and a 15 year old 3,5-4,5 28 - 70 mm
and the results are comparable the d 60mm is not bad or soft in the distance
All were used at F11 and 60mm focal length.

F/11 is a bad choice with the D3x. Your RAW converter is creating some severe artifacts: Why are solid lines rendered as dotted, or zigzags? Raw Therapee will do this; NX2 is much better.

What is the reproduction ratio for the samples you are showing? Did you upsample them?

All told, I have serious reservations about the quality of your D700 and D3x samples; they exhibit very poor per-pixel sharpness and I know I can do much better than this with my cameras. I'm currently doing a study to compare D3x to 6x6 film - we shall see if your samples are worth the time you spent on them.

 

[Negus]

Hi Rayman,

Thank you for posting your test results.

I have to say that in regards to your tests, I am a bit disappointed by the D3x.
I really want to see Marianne test now.

Also, I should soon receive some D3x raw files so that I can have a look at them, and this will certainly be interesting.

Have a good day,
Negus

 

[elliotn]

Thanks Rayman.

Your new medium format film pictures/scans look much better than the ones you posted on the first page of this thread. (Which makes me wonder whether you've got the best out of the 4x5 film.)

Re. the comment on the artifacts in the Nikon files. Rayman has upsized them considerably to match the 4x5 scan (which is itself upsized, I think).

Elliot

 

[rayman 2]

Regarding the Rawconverter it is Nx2 and nothing else.
Other then applying some unsharp mask there is nothing added to the results.

Yes they were upresed to match the 4x5 in results in Photoshop you couldnt compare them otherwise. Those are NOT 100 % crops !
I personally dont see the Nikon D3X to look bad in this comparison.

It is a very high enlargement and you normally never saw that highreproduction ratios. The D3x actually compares pretty well to the 120 film format.

Please dont tell me how good the D3x looks show us exactly these reproduction ratios... ; )
Let me tell you that I dont think the D3x files are that bad the results of the
H3 are that good. There is more to a camera then MP count !

If you look at the comparison pictures from Hasselblad -folder they show similar.
I wasnt alone with the last D3X comparison...... we were 2 this time.

Plus we used 5 different lenses but yes it was f 11 it would have been unfair against the other systems to use different f stops.
Peter

 

[Marianne Oelund]

Yes they were upresed to match the 4x5 in results in Photoshop you couldnt compare them otherwise. Those are NOT 100 % crops !

Ordinarily, it isn't even valid to look at 100% crops on a monitor, as this is far from the reproduction ratios used in most printing. Neither do I agree that the subjects must be presented at the same size, in order to make comparison possible. The whole point of using higher-resolution cameras, is to make larger prints.

Perhaps you could tell us what size the full prints would be at your chosen reproduction ratios, to put things into perspective. You haven't even shown us the full images from the cameras, so we can see how you have assigned the FOV matchup. The choices made in that regard can have a significant effect on the outcome.

Plus we used 5 different lenses but yes it was f 11 it would have been unfair against the other systems to use different f stops.

No, it would not have been "unfair." It would have been proper test procedure. Please research the principle of equivalence . Using F/11 on all cameras causes a DOF mismatch.

 

[semorg]

Yes they were upresed to match the 4x5 in results in Photoshop you couldnt compare them otherwise. Those are NOT 100 % crops !

Ordinarily, it isn't even valid to look at 100% crops on a monitor, as this is far from the reproduction ratios used in most printing. Neither do I agree that the subjects must be presented at the same size, in order to make comparison possible. The whole point of using higher-resolution cameras, is to make larger prints.

Perhaps you could tell us what size the full prints would be at your chosen reproduction ratios, to put things into perspective. You haven't even shown us the full images from the cameras, so we can see how you have assigned the FOV matchup. The choices made in that regard can have a significant effect on the outcome.

Plus we used 5 different lenses but yes it was f 11 it would have been unfair against the other systems to use different f stops.

No, it would not have been "unfair." It would have been proper test procedure. Please research the principle of equivalence . Using F/11 on all cameras causes a DOF mismatch.

You are being too harsh. If you can do it better, go ahead and show us what you can do. It's always easier to criticize than to come up with your own work.

That said, I agree with some of your concerns. PS enlargement may not be the best comparison vs RIP software that actually do the size conversion for printings.

I also think f11 may not be the best use, but I understand the reason. Might have been better to show the samples from f 5.6 to f11. Also, I think the D3X would have created sharper image with a lens like leica 100 apo converted for nikon or leica 180 apo. D3X has much higher pixel density than the Hassy and it needs a higher resolving lens.

All said, this is a tremendous effort by the OP and very valuable, at least to me.

--
must watch: http://www.climatecrisis.net/
My Latest Gallery: From Peru to Chile: http://www.pbase.com/salim/peruchile2005

 

[Bernard Languillier]

No, it would not have been "unfair." It would have been proper test procedure. Please research the principle of equivalence . Using F/11 on all cameras causes a DOF mismatch.

Indeed, f6.3 is the f stop to use on the d3x for best sharpness on most lenses.

The 100% sample below is a handled image of a moving helicopter shot with a 300 f2.8 lens... it was converted with a focus on sharpness that doesn't work well for all parts of the image, but is pretty representative of what I often get:



Cheers,
Bernard

 

[Bernard Languillier]

Regarding the Rawconverter it is Nx2 and nothing else.
Other then applying some unsharp mask there is nothing added to the results.

Yes they were upresed to match the 4x5 in results in Photoshop you couldnt compare them otherwise. Those are NOT 100 % crops !

Thanks for the test results. I was suprised by the gap between the d3x and HD31. It is closer to the gap of price than to what I have seen with my own eyes.

A few questions if you don't mind.

• Considering that the aspects ratio are different, d3x and HD31 do have very similar pixel density on their long side, how did you compute the enlargement factor if I may ask and what was it?

• How did you focus the d3x?

Thank you.

Cheers,
Bernard

 

[rayman 2]

Perhaps you could tell us what size the full prints would be at your chosen reproduction ratios, to put things into perspective. You haven't even shown us the full images from the cameras, so we can see how you have assigned the FOV matchup. The choices made in that regard can have a significant effect on the outcome.

Well you could have looked at all the files from the above flickr site you would have been able to look at this.

i know very well of these dof equivalents but there is a lso a question of what
would be the best f stop for the lens and that depends ......

whenever you compare things you have 10 people tell you that the results would be so much different if if if if if.......
In the end we dont see anything to look at here at all which leads to lots of

people telling you how much different things would be without ever showing anything.....

I would love to see your "professional" effort at a comparison.... one that you can see things at ....... 100 % crops might be technologically right but you aint
going to see anything side by side...

 

[rayman 2]

Well first to the focus we used tjhe autofocus on the windows of the house
and we used manual focus on the 50 1.4 ais lens with focus assist.

to the aspect ratio thats the hardest part of the whole thing because the 31 MP
hasselblad has a slight crop sensor with a multiplication factor near 1.2x

I looked at the D3x as if it had the same aspect ratio of the H3 which is cutting it at the long end. the size of the D3x would be 51,21 x 34,14 cm at 300 dpi
The enlargement factor is almost 3x so the full size of the print would be
153 cm x 102 cm at 300 dpi and each clipis 2,85cmx3,98 cm at 300 dpi.

You see it was initially intended to be printed out not to be posted.
We are doing this for our own reasons and share what we see with you.

So on my computer in the explorerwindow I would see a crop of a picture of about 5 meters wide

 

[tmascolo]

Hello,

Hope you are well.

Would it be too much trouble to reprocess the RAWs from the Nikon cameras with Nikon Capture NX2? I would like to see if there is a difference in the RAW conversion.

I know this is a lot of work, and can understand if you do not wish to do any more with this project.

I still believe that the 4x5 and 120 format would win this contest if using standard darkroom enlargement. Scanning film is not the way to go with large format films in my opinion.

Thank you for all the effort.

--
Later,

Tony

 

[rayman 2]

Would it be too much trouble to reprocess the RAWs from the Nikon cameras with Nikon Capture NX2? I would like to see if there is a difference in the RAW conversion.

I know this is a lot of work, and can understand if you do not wish to do any more with this project.

I still believe that the 4x5 and 120 format would win this contest if using standard darkroom enlargement. Scanning film is not the way to go with large format films in my opinion.

Thank you for all the effort.

--
Later,

Tony

I have no idea where all the misinformation comes from ...
I clearly stated all the time that all conversions were made with NX2.

I only crosschecked on the D700 with capture one and the NX2 is better by a margin.
I have done the Nx2 conversions for all lenses and the differences even between
the lenses at f 11 arent very big......
The difference between the formats is much bigger.
the difference between 12mp and 24mp Nikon is bigger then the difference
in between the lenses at f 11.
at f 2.8 that would be another story ......

Just for the records the people want to have different raw conversions + different
films plus different scans plus different f stops.
thats 12x the work i put into this and thats pretty much already.. ; )

If you are 100 % certain that enlargements are better then scanned film
just do a comparison and post it here.....
by the way the light path of the Imacon is that of an enlarger thats
different to drum scans...

but i am not trying to knock anyones beliefs over i just want to see not hear
the results.
Peter

 

[tmascolo]

Hello,

Hope you are well.

As I was reading through the thread, I had forgotten what was stated in your original post, I apologize. It does indeed state that you used NX2. I got confused while reading, it is late at night and I should not have posted.

As far as enlargement vs scanning there is no way to do that on-line, you would have to compare actual prints with a loupe. Otherwise you are just scanning a print, which just shows how good your scanner is unfortunately.

--
Later,

Tony

 

[bcphoto2]

I'm catching up on this thread -- thanks for your effort -- it's fun reading your comparisons. I don't believe that a person that has shot DSLR or 35mm film for their experience will understand the MF or large format comparison. Those who have shot all those formats know the advantage of going larger. As they say in racing -- there's no substitution for cubic inches.

 

[bcphoto2]

A tremendous effort that I applaud.

But I have yet to see anyone mention the real differences between the D3X and the H3 images -- they are contrast and saturation. The D3X images are flat and weak most probably due to the camera having an AA filter. At 100% with any DSLR with AA all we will see is a mushy image -- by design to prevent moire. The H3 is AA free and we can see the results.

I'd like to see a comparison between the H3 and the new Leica M9 -- neither have AA filter.