4K Mega-trumpet Chart vs. Dragon & Epic-MX

[Corey Robson]

For anyone interested in seeing the spatial resolution difference between Dragon and Epic-MX on a lifeless, monochromatic 4K chart here you go. The test was simple, both cameras set to 800ISO, 5600K, 6KFF and 5KFF respectively. Redlogfilm gamma. Cameras pointed at my 4K Mega Trumpet chart from DSC Labs.

Original R3D->TIFF exports were done from RCX-Pro 30. JPEGS were created at best quality using OSX Preview. JPEG samples are cropped from original RCX TIFF exports. The full size JPEGS & TIFFs were too big to post below but the links are there if you want to download the JPEGS. Nothing valuable that I can see was lost in the TIFF->JPEG export.

I didn't have my PL mount handy when I shot the MX sensor (shot them at different times) so I'm not playing entirely fair by using a $1000 piece of glass on the Epic-X. No surprises here but anyway, here's what happened:

Below - Dragon JPEG sample 6KFF 800ISO Leica Summicron-C 50mm T5.6

Below - Dragon JPEG sample 6KFF 800ISO CookeS4i 75mm T5.6

Below - Epic-X MX JPEG sample 5KFF 800ISO Canon 100mm MacroL F5.6

Full size chart JPEGS:
https://dl.dropboxusercontent.com/u...n_6KFF_800ISO_Leica_Summicron-C_50mm_T5.6.jpg
https://dl.dropboxusercontent.com/u/55061736/Public/Dragon_6KFF_800ISO_CookeS4i_75mm_T5.6.jpg
https://dl.dropboxusercontent.com/u/55061736/Public/Epic-MX_5KFF_800ISO_Canon_100mm_MacroL_F5.6.jpg

Samples from original full-size TIFFs:
https://dl.dropboxusercontent.com/u..._6KFF_800ISO_Leica_Summicron-C_50mm_T5.6.tiff
https://dl.dropboxusercontent.com/u...TIFFsample_6KFF_800ISO_CookeS4i_75mm_T5.6.tif
https://dl.dropboxusercontent.com/u...mple_5KFF_800ISO_Canon_100mm_MacroL_F5.6.tiff

 

[Jacek Zakowicz]

:001_rolleyes:Corey
If you make the Epic X trumpet dark and small- of course it won't look as good.
How about being fair to the trusty Epic X that has served you well and match the brightness and size to the dragon pic

 

[Corey Robson]

I intentionally did 1:1 crops from the FF images and didn't make any adjustments to keep the results as objective as possible.

 

[Jacek Zakowicz]

Nobody will look at the epic x on a smaller, darker screen on purpose. I'm not sure how it is "objective"

 

[Justin McAleece]

Does the Dragon expose a brighter image at identical ISO's? That's what I've seen evidence of in other examples. As I understand it, that's some of the reason why Dragon is "cleaner" than the MX. The way RED chose to register its ISO benchmarks is actually brighter than on the MX. That would make sense because other cameras often look brighter at identical ISO's than the M and MX sensors do and did.

 

[Mikael Lubtchansky]

Thanks for the effort Corey but it's difficult to judge spacial resolution without the exact same piece of glass between the sensors and the chart.

We can still conclude that 6K DRAGON with 50mm summilux resolve more lines than 5K MX with a macro lens used at non macro distance.

 

[Corey Robson]

Thanks for the effort Corey but it's difficult to judge spacial resolution without the exact same piece of glass between the sensors and the chart.

We can still conclude that 6K DRAGON with 50mm summilux resolve more lines than 5K MX with a macro lens used at non macro distance.

Yes, very true. I had the Epic 5K chart shot on hand from a while back and thought I'd add it to the Dragon chart I shot yesterday. In hindsight I should have just posted the Dragon 6K FF chart and left it at that. When I have a chance I'll reshoot the Epic-X chart with the same PL glass and brighter exposure and post it here.

 

[Mikael Lubtchansky]

When I have a chance I'll reshoot the Epic-X chart with the same PL glass and brighter exposure and post it here.

Cool. If you do get a chance to redo it, make sure you shoot both camera at the same time, using same lens and everything else exact same conditions.
When doing such a test it is also nice to record while slowly adjusting focus around the distance mark to select the sharpest image frame by frame in post.
Then you can compare 2 lenses or 2 cameras with maximum "objectiveness" if such a thing is possible.

 

[Robert Ruffo New]

Nobody will look at the epic x on a smaller, darker screen on purpose. I'm not sure how it is "objective"

agree

 

[Marc Wielage]

More contrast would have made the differences much easier to see, particularly in an online JPEG photo.

 

[Les Dittert]

My Epic without OLPF out resolves the Dragon easily . I like it that way.
I'm skipping the current sensor upgrade cycle. The noise is almost the same.

 

[Corey Robson]

More contrast would have made the differences much easier to see, particularly in an online JPEG photo.

True but I'd rather get criticized for that than get caught adding contrast and the potential false detail / exaggerated aliasing which would result followed close behind by passionate mud slinging. Any amount I could add might be seen as too much or too little for some. Played it safe by posting the Redlogfilm versions, feel free to grade these yourself.

Here are R3D stills for the Dragon and Epic-X MX taken from the sharpest frames in the original R3Ds:

Below - Dragon 6KFF Leica Summicron-C 50mm T5.6
https://dl.dropboxusercontent.com/u...mpet R3Ds/Leica/A002_C006_0506BZ.0000159F.R3D
https://dl.dropboxusercontent.com/u...mpet R3Ds/Leica/A002_C006_0506BZ.0000159F.RMD

Below - Dragon 6KFF Cooke S4i 75mm T5.6
https://dl.dropboxusercontent.com/u...mpet R3Ds/Cooke/A006_C018_05064I.0000065F.R3D
https://dl.dropboxusercontent.com/u...mpet R3Ds/Cooke/A006_C018_05064I.0000065F.RMD

Below - Epic-X MX 5KFF Canon EF 100mm f/2.8L Macro set to F5.6 (feel free to ignore this one because of the EF glass)
https://dl.dropboxusercontent.com/u...pet R3Ds/Epic-X/A002_C047_0505FN.0000033F.R3D
https://dl.dropboxusercontent.com/u...pet R3Ds/Epic-X/A002_C047_0505FN.0000033F.RMD

 

[Antoine Fabi]

Hi Corey,

Could you do the same test with the same lens on both sensors?

Lenses can (DO) vary a lot.

Thanks

Antoine

 

[Joel Arvidsson]

My Epic without OLPF out resolves the Dragon easily . I like it that way.
I'm skipping the current sensor upgrade cycle. The noise is almost the same.

How often do you run in to trouble for not having the olpf and can you to fix the problems in post?

 

[Corey Robson]

Hi Corey,

Could you do the same test with the same lens on both sensors?

Lenses can (DO) vary a lot.

Thanks

Antoine

I will eventually but my Epic-X is out on a job for a while. I'll post it here though when I get to it : )

 

[Antoine Fabi]

thanks Corey!

Antoine